Sites: GlobalSpec.com | GlobalSpec Electronics | CR4 | Electronics360
Login | Register
The Engineer's Place for News and Discussion®

Previous in Forum: Transformer Flux Linkage   Next in Forum: Electrical Motor Test Stand
Close

Comments Format:






Close

Subscribe to Discussion:

CR4 allows you to "subscribe" to a discussion
so that you can be notified of new comments to
the discussion via email.

Close

Rating Vote:







Page 1 of 2: « First 1 2 Next > Last »
Commentator

Join Date: Apr 2011
Posts: 98

How to Build an AVR for a Three Phase Generator?

04/15/2011 5:08 PM

I got one used diesel three phase generator with missing avr module (Lutian 5gf-lde3). Normally, it does not produce electricity at all but since the engine is in very good condition and the generator seems to be in a good shape (resistance measurements of the phases and the exciter windings are ok) I found challenging to try making an avr module. The main problem is the design of the circuit. I tried to google but I could not find any schematics. I would appreciate some help. Thanks

Mishel

Register to Reply
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Anonymous Poster #1
#1

Re: How to Build an AVR for a Three Phase Generator?

04/16/2011 11:40 AM

Think this type of generator is brushless self-excited, That means no AVR cirquit, just exciter AC capacitors. Check this: http://lutian.en.made-in-china.com/product/AbOnkBpdffcy/China-Diesel-Generator-5GF-LDE3-.html

S.M.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#2
In reply to #1

Re: How to Build an AVR for a Three Phase Generator?

04/16/2011 2:41 PM

Thanks, that also came across my mind! I saw on the same type of the new genset the square box with electronics (an avr?) in the inside. I tried to investigate the circuit but it was covered with black plastic layers above the plates on both sides. I guess the manufacturer does not want me to see what's inside. The wires coming out of the generator end there and I could only see a heat sink from a transistor or something at the surface of the circuit. The reseller was not interested to help at all and he told me that was a "week spot" of the product. He had only similar avr's on the shells that were kidney shape but he could not guarantee it would work for me. That's why I am trying to get other solution. There are six wires coming out of the generator, two (black and red) I guess should be from the slip rings for the rotor while the other four for the excitement winding for all three phases (parallel to the power ones in the stator). Power installations is not my prime so I cannot say for sure.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#3
In reply to #2

Re: How to Build an AVR for a Three Phase Generator?

04/16/2011 3:03 PM

This is the closest theory I could find, but again no particular design (solution)

http://www.alibaba.com/product-gs/358906117/Learnt_Power_alternator_with_brushless_self.html

"Voltage Regulators

With this self-excited system the main stator provides power via the Automatic Voltage Regulator (AVR) to the exciter stator. The high efficiency semi-conductors of the AVR ensure positive build-up from initial low levels of residual voltage. The exciter rotor output is fed to the main rotor through a three-phase full-wave bridge rectifier. The rectifier is protected by a surge suppressor against surges caused, for example, by short circuit or out-of-phase paralleling."

Register to Reply
Commentator

Join Date: Feb 2011
Location: Dubai, UAE
Posts: 63
Good Answers: 7
#4

Re: How to Build an AVR for a Three Phase Generator?

04/16/2011 11:02 PM
Register to Reply
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#7
In reply to #4

Re: How to Build an AVR for a Three Phase Generator?

04/17/2011 12:06 AM

A.A. Granted that the theory is similar, but you provided a nice diagram of a mechanical voltage regulator for a DC lighting generator. I believe the OP said it was a three phase AC generator and that he was looking for a solid state AVR module.

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#11
In reply to #4

Re: How to Build an AVR for a Three Phase Generator?

04/17/2011 3:35 AM

Thanks for the circuit but I think I cannot use it because I am searching for an avr for a three phase ac generator. Thank you again!

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#5

Re: How to Build an AVR for a Three Phase Generator?

04/16/2011 11:28 PM

First rule of technical writing: Do not use and acronym without first defining it.

Had to google it,, I hate that.

Martin

Register to Reply Off Topic (Score 4)
3
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#6

Re: How to Build an AVR for a Three Phase Generator?

04/16/2011 11:53 PM

The first thing that you want to do is verify that your generator can actually produce voltage. The safest way to do this is to separate and individually insulate all the leads that you found and fire up the diesel and using a voltmeter set on its highest scale, check for voltage on each phase of the generator output. If none is found then start lowering the scale on the meter until something is found even if it's only a few volts. The reason this usually works without any field current is that there is usually a small amount of residual magnetism in the steel of the rotor and this may provide enough to make a sensitive voltmeter move, if you have access to an oscilloscope then you can observe the waveform instead, it should be the same magnitude across all three phases.

When you're done with the main generator leads it's time to check the leads from the exciter windings, which is nothing more than a small 3 phase AC generator. Do the same tests as above, two leads at a time, again if everything is ok you may get a small voltage across all three leads.

If you can't get any voltage on any leads then there is not enough residual magnetism and you will need to use a variable DC power supply to provide field current to the black and red leads. The following steps require extreme caution since there's a chance of creating a shock and/or fire hazard! Standard lab precautions need to be taken; i.e., some form of overcurrent and overvoltage limiting is required on your DC supply, complete and total isolation of everything from each other and from ground. Since we don't know the field resistance you simply start at zero volts and raise the voltage slowly until your meter on the generator leads starts deflecting. When that happens start checking for uniform voltage on all leads, if everything checks out you can start increasing the field current until your meter shows the rated generator voltage on all three phases as well as some balanced voltage on the exciter leads. Caution-> if at any time the voltages aren't closely matched then there may be an internal problem; i.e., turn to turn shorts, turn to frame shorts, a winding that opens intermittently, etc. and the testing should stop.

Assuming everything is ok then you have proven that the generator can indeed generate and it is now worth your time and money to try and find the proper AVR module. Some of these generator excitation systems are very simple, consisting of some diodes, triacs, and capacitors, but it is not something I care to do without a lot more hands-on testing, but you may get lucky and find some simple circuits in the manuals of similar generators, or you could simply run open loop with a variable DC supply (sometimes a 12V battery and a high current variable resistor in series) providing the necessary field current. Good luck with your experiment.

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply Good Answer (Score 3)
Member

Join Date: Nov 2012
Posts: 5
#130
In reply to #6

Re: How to Build an AVR for a Three Phase Generator?

11/03/2012 1:06 AM

i have a rotary converter generator (208v, 400 hz, 3 phase, 30 kva)

can you please help me out with a circuit for designing its avr module,its old avr module has burnt out , it used to take input feedback from all three phases of main winding and neutral

line voltage = 208v u-v, v-w, w-u

phase voltage =120v u-n, v-n, w-n

the exciter field resistance =129 ohms
the three main windings are healthy and are showing equal ohmic value of 0.3 ohms

with previous avr module following were the exciter field current an exciter field

voltage at no load

exciter field current = approx 0.3 amps

exciter field voltage = approx 36v-38v

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#8

Re: How to Build an AVR for a Three Phase Generator?

04/17/2011 12:58 AM

Start with this page for an explanation of how brushless alternators work:http://www.pearen.ca/dunlite/BrushlessAlternators.pdf

It should help with your design/

Martin

Register to Reply
Guru

Join Date: Nov 2006
Posts: 663
Good Answers: 3
#9

Re: How to Build an AVR for a Three Phase Generator?

04/17/2011 2:37 AM

Most manufacturers cover it as u have mentioned with a black resin compound which hardens after application.This is to protect the components from vibrations they may encounter and other thing is not to allow component level service.However there are AVR's available without this covering which can be used .By the way is make is the alternator?then the manufacturer can supply the AVR and u can use the set.Why own the cow when milk is available?

__________________
To avoid crticism do nothing,say nothing,be nothing
Register to Reply Score 1 for Good Answer
Commentator

Join Date: Apr 2011
Posts: 98
#10
In reply to #9

Re: How to Build an AVR for a Three Phase Generator?

04/17/2011 3:33 AM

Thank's a lot folks. I will do the testing of the generator asap and let you know about the results. As I previously said, I've been told these avr's are cheap and unreliable and there are may companies that sell avrs and 5GF-LDE3 gensets (with different wires coming out of them). The reseller told me if I use the wrong one I could burn the generator or so and he could not confirm the "right one for me",since they are covered and unreadable. I will try to carefully measure and excite the generator so I will let you know the measurement so we try to find a solution together. I also have a homework to read, thanks again.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#12

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 5:16 AM

Something like this...... but we do not know voltages, yet. The winding 1-2-3-N is the main machine output. The field will be the last two wires of the six???

Register to Reply Score 1 for Good Answer
Commentator

Join Date: Apr 2011
Posts: 98
#13
In reply to #12

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 10:15 AM

Thank you all for the inputs,

today I've done some measurements of the generator. The main windings seems to be ok, there are symmetrical resistances on all three phases, 3.8 Ohms each. When engine is on I can measure only 2vAC on each to the zero (4v between R-S-T), no avr or any excitement connected.

The excitement part is a different story. The AVR slot has a six wires out, two separate and other four on the connector. The two separate outs with their 3.8 ohms on cold give some 2V AC. The other four have two alive with 24 Ohms and give 2.2V AC(blue and amber)and last two are dead (black and red). No resistance to any direction from these last two or any voltage.This way I cannot figure out the schematic of the generator. This weekend I will have to dissassemble the electric part and look inside. I will try to get the wiring solution so I know exactly what sort of excitement windings it has and what is the cause of the dead outs. I think that would help you to help me. I could see that the aftermarket AVRs have six wires (2+4) outs so I assume they are inter-changeable for the same type/size of generators. That means we are looking for a solution of a six pole circuit.

I don't have means/variable and safe source of AC/ to try to excite the generator and measure the behavior of the main windings. I will try to figure out the generator first and find out how many and what sort of excitement windings it has and let you know soon.

thank you

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#14
In reply to #12

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 10:17 AM

Sorry, I should have mentioned earlier, it is 220V AC, Three phases , 5 KW

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#15
In reply to #14

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 12:23 PM

Thanks for Info #13, #14.

  1. Is there any resistance between what you call main windings and excitement windings? 3.8 ohms on the separate pair seems too much like the 3 .8 ohms per phase of main winding - it might be a voltage sensing feed from main winding to AVR board.
  2. Red and black suggest direct current field winding positive and negative ; the lack of any resistance is ominous.
  3. If your output is 220 V/phase-neutral , then resistor R2 on my AVR circuit becomes 220k.
  4. If there was a permanent magnet exciter in this generator, I would expect 10s or 100s of volts on some windings with engine at governed speed, but without any DC excitation.
  5. Does any wire have a resistance to the earth or frame terminal of the generator? If you do not have an actual frame terminal, some scraping or filing may be needed to get a good contact for your voltmeter.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#16
In reply to #15

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 12:55 PM

Thank you, first thing tomorrow ( day light:) I will provide answers for your questions. I can see additional diodes between the phases, does it mean I would be able to draw the power from the points 1,2 and 3 on your circuit as 220vAC with the change mentioned on the R2? Any details about D1-2-3 and the diode parallel to the exciter field?

When I say main winding, it is the resistance at the outlet, between the phases and zero. ( where I connect the load)

I cannot say if there are any magnets inside before I brake in but I expect that to happen soon.

Thank you

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#17
In reply to #16

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 4:24 PM

The diodes D1,2,3 make a 3 phase rectifier to feed the 220k with a direct current voltage proportional to [average of 3 phases] peak voltage of the machine alternating voltage output. Actually with 220 volts root mean square phase output, this DC will average about 295 volts and the actual machine output would regulate at roughly 170 volts rms with R2 = 220k. Once you get regulation at that value a variable resistance, say 50k, added in series with R2 would enable you to raise volts to rated. For D1- D3 type 1N4007 should be OK. The field diode and transistor will depend on the field current, which is not yet known (and the +20V DC and R2 = 220k are only a first "stab" at values - we do not yet know the field voltage!).

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#18
In reply to #17

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 5:25 PM

Thank you for the details. This sounds good to me but I have to do the first thing first and dig into the generator so I see the red/black wire problem. By the way, how big resistances I should use (1/2 W,1 W or bigger), especially the variable one?

thank you for the quick response

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#19
In reply to #18

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 5:49 PM

P=E**2/R, about 200V**2/220e3 = 0.181W, so 1/4 watt

Martin

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#20
In reply to #19

Re: How to Build an AVR for a Three Phase Generator?

04/20/2011 6:05 PM

thanks for your answer.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#21
In reply to #15

Re: How to Build an AVR for a Three Phase Generator?

04/21/2011 12:09 PM

1.Yes, there are some readings between the main (power) windings and the two separate wires of the excitement windings.( The other four at the connector do not have any connections with the mains).

The first one is in short connection with the zero (common) connector at the main windings and it reads 1.5 Ohm to every phase (R-S-T) while the other one shows short connection to one of the phases (middle one "S"), 2.5 Ohm to R and T phases and 1.5 Ohm to the zero ( common).

2. I will try to find out what is the reason for the dead black/red during the weekend.

3. Thanks for the 220 V solution of the R2.

5.There is a ground terminal on the frame and the power outlet and there is no resistance to any of the wires of the (imagined:) AVR and the wires connected to the power outlet ( main winding). I guess that means no shorts to the frame from any winding.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#22
In reply to #21

Re: How to Build an AVR for a Three Phase Generator?

04/22/2011 6:38 AM
  1. This definitely seems a voltage sensing (and power?) feed to AVR.
  2. The literature to which you referred mentions a "12 V DC 10 amp output". Could be red/black for DC. Maybe battery charging? Does set have starter battery (with separate battery charging alternator??)?
  3. The 24 ohm winding may be the field. Certainly 10 amps x 24 ohms = 240V drop at 10 amps does not make a "12 V, 10 amp " output. Could try connecting 3V (2 x D size alkaline primary cells, with 1 amp fuse.
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#23
In reply to #22

Re: How to Build an AVR for a Three Phase Generator?

04/22/2011 3:53 PM

1. Ok, whatever you say , if so-does it mean that the circuit above is usable?

2. Yes it does have a start lead acid battery which has two separate wires (other than red and black from the avr that are stone dead) I have not mentioned them earlier because I didn't think it was relevant for the generator. I am sure they are supposed to charge the battery but they can be used only to start the engine, after starting they are dead (no voltage at all, but I think it should be aprox 14V DC on them just as in every vehicle installation) No 12 V at the front panel as well ( claimed to charge any outside battery). I assume there must be a separate 12V installation to be used for the charging the inside battery and feeding the front panel 12V terminals. I am not sure if that is connected any way with the actual 220 V generating. It has a separate starter on a side of the engine, so if there is any winding for 12 V in the generator it's not used for cranking ( like some dc generators that I know)

3. I like the idea to connect the 3V to the those 24 Ohms. It sounds safe and logical. Will do asap and let you know about the behavior of the generator.

thank you

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#24
In reply to #22

Re: How to Build an AVR for a Three Phase Generator?

04/26/2011 12:40 PM

I connected 3V batteries to the wires that show 24 Ohm resistance and I got 30 V at the main instrument at the panel. Only in one direction +/-, opposite connection ( when I switched the poles) there was no voltage at all.

I am little late with this answer due the Easter "no-working-days.

Register to Reply
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#25
In reply to #24

Re: How to Build an AVR for a Three Phase Generator?

04/26/2011 2:27 PM

Progress! There may be a diode in the field circuit to prevent reverse polarity and/or across it to absorb the field collapse, in any event does your multimeter show 30 volts phase to phase across all three phases? I'll bet a 12 volt battery on the field gets you near 120 V if everything else is ok. How about any voltages across the 4 other leads in that group?

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#26
In reply to #25

Re: How to Build an AVR for a Three Phase Generator?

04/26/2011 4:14 PM

That was quick reaction, thank you!! I thought myself the same but did not have the courage to try without an expert suggestion. I will try with 6V and 12V soon ( I should be careful with noise and the neighbors, their fence is painfully close and they have children). I will also read the complete situation with voltages at the power phases and the rest of the wires at the "avr port" in these two situations. I guess I should try all these before I actually start dismounting the generator and figure up the cause for the dead black/red wires.

Register to Reply
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#27
In reply to #26

Re: How to Build an AVR for a Three Phase Generator?

04/26/2011 7:14 PM

They're not "dead" and assume that anything electrical is alive, or you'll be the one that gets fried!! If you have the means take a picture of the nameplate and the end-bell of the alternator then try posting for us to take a look at. I wouldn't dismount the alternator yet because if the external measurements check out they should tell us all we need to know.

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#28
In reply to #27

Re: How to Build an AVR for a Three Phase Generator?

04/27/2011 4:08 AM

Yes, good point, I guess in time we get careless. I have camera and I will do as you say, try 12v on the field and find the nameplate of the generator. The end-bell of the generator is behind the double pot muffler with a LOT of mineral sound insulation and it will take some time to get it out, but I will do it!

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#29
In reply to #28

Re: How to Build an AVR for a Three Phase Generator?

04/27/2011 8:37 AM

Hi Mishel, glad to see the progress!

  1. Looks as if rated volts will need 24V on field, zero load current. "Rule of thumb" is that rated output at unity power factor [pure resistance, incandescent lamps or equivalent] will need twice that (48V): rated output at 0.8 power factor lagging, three times [75V]. Allowing that field resistance will increase about 1/3 when machine is hot means rated output at 0.8 power factor lagging needs 100 volts at 3 amps. Assuming all phases have equal volts, as Ramconsult writes, showing we have found the field winding and the machine is healthy!
  2. "Field forcing", to cope with two or more times rated current at low power factor e.g. 0.3 power factor starting a 3 phase motor, will need a proportional increase of field current. Hence it is probable that the AVR is designed to work from about 240V - that is the full machine phase voltage or a separate exciter winding of that voltage.
  3. As Ramconsult notes, you might have some diodes hidden in the field circuit - referring to my diagram in reply #12 - it may be in parallel with the "field"winding, as drawn on my diagram [the way diode symbols are drawn, they conduct current - positive to negative, the direction of the "arrow" of the symbol - if positive is applied to the "triangle" and negative to the "crossbar"]. The way to tell would be to apply the 3V battery to the field with both possible polarities , measuring the DC voltage at the field wires. With the polarity which gives 30V AC at main terminals, I would expect to get 3V; swapping the battery polarity would give about 1 volt, because the diode would conduct, with about 1 volt drop, bypassing the winding. Actually, two new alkaline D cells would have an internal resistance of about 0.2 ohms, suggesting a current of (3V - 1V)/0.2 = 10 Amps, which is a lot and could overheat even a large diode without a heatsink.
  4. Bearing in mind possibility of last sentence of 3. , it is important to get right polarity before applying any more than 3 volts to the "field" wires, else some damage could be done. If possible, measure DC current from 3V battery to "field" with both polarities [no need to run engine to do this]. Be sure your meter range is 10 amps full scale - the working rule is always start with your voltmeter or ammeter on the highest range.
  5. I do not recommend putting a voltmeter in parallel with the field winding, then connecting and disconnecting the battery by prodding a wire on - use a switch to connect the battery, then prod the voltmeter on to check the voltage. Switch off the battery as soon as possible. Breaking inductive circuits with amperes of current can give VERY high voltages. A couple of years ago, I destroyed a cheap but useful multimeter, by connecting it to measure a DC motor field voltage while it was running - without thinking, I stopped the motor by opening its main isolator. This meant the tens of amps flowing found a lower voltage circuit to flow in, which was the the motor "running" lamp (briefly) - which immediately fused, followed by smoke from the meter. A convincing demonstration that you cannot stop the current in an inductive circuit instantly - you can only remove the stored energy in something at a rate which depends on how high the voltage gets. Inside the meter it was found that, the 1 megohm range resistor had popped, the current flashed over to less repairable electronic parts, removing some PCB track to finish.
  6. As Ramconsult says "Be warned...." - dangerous voltages and things may not be as "dead" as you think.
  7. When you break an inductive circuit, taking a wire off a terminal by hand, the air gap breaks down at about 300V with a tiny gap. The arc formed burns at about 20V, not increasing much with gap. So the arc burns off the energy as heat and light without too much drama at a few amps. But isolators are designed to break many times full load current in 1/100 second or so - they do this by increasing the gap and arc voltage quickly - hence the effect of the voltage on my poor voltmeter.
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#30
In reply to #29

Re: How to Build an AVR for a Three Phase Generator?

04/27/2011 11:59 AM

Thank you so much. Please don't mind, It will take time for me to read several times and comprehend all these before I start to work.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#31
In reply to #29

Re: How to Build an AVR for a Three Phase Generator?

04/28/2011 4:38 PM

1. I hope you are right about the state of the generator. I will test with higher voltage the "field" winding and let you know about the readings at the outlet and the avr connector. I think I understand the supposed dynamic reaction of the future AVR but I cannot go deep into it because I hope that the circuit will do it without my knowledge.

2. Same as above, just to add that I intend to use the generator mostly for electric arch welding /demanding inductive loads/. So please if you help me with the circuit bare that in mind. I do have 1KW Honda that covers perfectly all other needs at my off-grid summer house.

3. and 4. I will determine the polarity of the battery and try not to fry it.

5. I will try not to fry the multimeter!

6. I will be very careful with my fingers, too!

7. I will prepare one insulated "on cord" switch for the power source, small improvised board with connectors of all the wires so I can quickly measure the voltages with different excitement voltages at the "field" and I will prepare some photos of all these.

If I survive, most probably after the 1.and 2. of May (when we usually have picnics, BQ and we don't work) I will let you know about the results of my work.

thank you so much for your help

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#32
In reply to #29

Re: How to Build an AVR for a Three Phase Generator?

05/06/2011 2:58 PM

I am sorry for being late. We had some weather here, first sunny day and here I am with camera and tool box. I must admit I am a little bit confused and this is why:

1.I excited "field winding" (first two on the left) with 6V/DC and I got 120V/AC at R;S;T to N (all the same) and aprox 55-60 V/AC at the brown/blue wires (last two at

the right side. With 13V DC excitement I got almost 270 V/AC output and aprox 150V/AC at brown/blue. No load attached.

2. I could not detect any voltage at the black and red wires.

3. There is no classic name plate on the generator.

I am anxious to find out why those two are dead ( black and read) but I will wait for your comments before I start to work "inside". I don't mind missing 12V for the battery charging, if you think the circuit given above will work I could try to build it and test it after this readings.

thank you

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#33
In reply to #32

Re: How to Build an AVR for a Three Phase Generator?

05/08/2011 1:47 PM

Discrepancy between your post #21 item 1 "Two separate wires to AVR......

First one short to zero of main windings....other one shows short.... to phase S."

and

latest post #32 " With 13V DC excitement.....270 VAC output and approx 150 VAC at brown/blue"

If #32 "Brown/blue" and #21 "two separate wires to AVR" are the same wires, one would expect 270 VAC at brown/blue (which were, according to ohms, connected direct to N and phase S of main windings).

Are the ohms values of post #21 the same now, if you repeat the test ?? Is the blue wire short to N of main winding and brown short to S of main winding?? Is it possible this brown is a part voltage tap into main winding?? It might be a design in which AVR is "120V", direct to main winding on 120V/60 Hz machines and to a tap on 240V/50 Hz machines.

Does look as if you have a functional machine, anyhow.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#34
In reply to #33

Re: How to Build an AVR for a Three Phase Generator?

05/08/2011 4:32 PM

I am sorry for the discrepancy that you noticed in your last post!! I must admit now I am completely lost. I did what you told me to do. I put 6V DC and 12V DC at the wires that showed 24 Ohms on cold (field??). I put all the wires on the new connector in order to avoid any shorts but you can see at the picture (32-1) that the first four wires on the left are part of the "imagined/missing" connector while the last two on right are "two separate"-brown/blue. I measured all the voltages at the wires and told you the readings. I will measure again and re-check the values! It is possible that somewhere I scr..w it? I am going to provide digital multimeter and do the measurements again and again with the diesel of and on. It is possible that my old analog instrument is not very accurate but I doubt much different results.

I cannot say anything about the design of the (missing avr) and I admit I do not understand the part "to a tap on 240V/50 Hz machines" but I can say that there is a small sticker at the generator saying "230V/50Hz". I DO understand the last sentence about having a functional machine and that's why I am going to bother you a little more!!

Will be back very soon and thank you for your answer!

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#35
In reply to #33

Re: How to Build an AVR for a Three Phase Generator?

05/09/2011 2:10 PM

I tried to measure the values several times and I always got a different readings. I bought a digital instrument and did the test again, these are the readings with engine off:

Amber Blue Black Red

BLUE

BROWN

N R S T
Amber X 49 Ω 0 0 0 0 0 0 0 0
Blue 49 Ω X 0 0 0 0 0 0 0 0
Black 0 0 X 0 0 0 0 0 0 0
Red 0 0 0 X 0 0 0 0 0 0

BLUE

0 0 0 0 X 1.5Ω sc 1.5Ω 1.5Ω 1.5Ω

BROWN

0 0 0 0 1.5Ω X 1.5Ω 2.8Ω sc 2.8Ω
N 0 0 0 0 sc 1.6Ω X 1.5Ω 1.5Ω 1.5Ω
R 0 0 0 0 1.5Ω 2.8Ω 1.5Ω X 2.8Ω 2.8Ω
S 0 0 0 0 1.5Ω sc 2.8Ω 2.8Ω X 2.8Ω
T 0 0 0 0 1.5Ω 2.8Ω 1.5Ω 2.8Ω 1.5Ω X

First four avr connector, next two separate wires from the connector BLUE/BROWN and last four power outlet,

sc short connection, 0 open circuit.

I excited the "field" winding (49ohm)with 13.3V DC but the voltage dropped to 10.3V when wires were connected to the winding. With engine on the output voltage was a nice 240V/AC (N to any R;S;T) and 150 V/AC at the BROWN/BLUE (separate wires from the avr). I went further more and tried my 1KW grinder at the 220 V AC outlet and it worked just fine but the 240Volts dropped approximately 30V under load.

I apologize for the initial inaccuracy, I can always blame the instrument? I still don't know are this results of the test expected or there is (again) some "no-match to the math issue"? Please, if there is something that you think I should do just tell me and I will do it in order to revive this tool and put it in function.

Thank you for being patient and cooperative despite the mistakes.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#36
In reply to #33

Re: How to Build an AVR for a Three Phase Generator?

05/10/2011 3:11 PM

Hi again,

I have removed the mufflers and I found out this:

Obviously I was not the first one to have the idea

1.The wires from the so called field that we used for excitement of the generator are actually wires from the slip rings that go to the rotor.

2.The "dead" red and black are really dead because the connector has been disconnected and there is a free hanging four pole connector at the upper side and one red wire extra. It fits together with the two pole connector with the red/black wire.

3. I tried to measure the resistance at the connector that "hangs" and it seems to me like it is (as somebody said) a small alternator inside the windings, it is 3+1 wire but the resistance is so small, I would say an ohm or less but I cannot say for sure because I cannot measure accurately.

Any hints? Can I use the circuit above and when we talked about field did you think about the rotor field or excitement winding at the stator? Can the avr be build and used without the usage of the upper (loose)connector?

thank you

Register to Reply
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#37
In reply to #36

Re: How to Build an AVR for a Three Phase Generator?

05/10/2011 6:42 PM

Congratulations! The mystery is almost solved. What you have is a (more or less) universal 3 phase self-excited alternator complete with auxiliary windings for battery charging. I've had a couple of these (single phase only) but the principles of operation are nearly identical.

First let's examine the "self-excited" circuitry. In one of my earlier posts I talked about residual magnetism of the rotor and how it usually provides enough magnetism in the poles of the rotor to generate a few volts without any field current. That small voltage then comes out on a connector (probably the 4 pole one) that would normally be rectified into DC and then applied to the field circuit via the slip rings (the two pole connector). During startup the residual magnetism causes a small voltage across the output of the alternator which is rectified and applied to he field windings which in turn increases the field current which in turn causes more voltage which in turn causes.... Well you get the idea, it's a positive feedback loop.

So how do we "regulate" the terminal voltage? Remember the four pole connector? In these small alternators there are usually a few auxiliary windings embedded in the stator, one or two for generating/controlling the excitation voltage and one for the battery charging circuit. The output of these is fixed by the number of turns in the windings so that the maximum field current and hence the maximum terminal voltage is limited by design. In most of the small alternators I've seen the circuitry could hardly be called an Automatic Voltage Regulator (AVR) by industrial standards, but it does provide more of a voltage limiting function.

Typically the output of the self excitation winding is rectified and then placed across some circuitry that includes a zener-diode/resistor combination which limits the voltage across the field (and hence the field current) and subsequently the terminal voltage of the alternator. The voltage regulation is provided by the second winding, it may be wound to sense the output voltage of the alternator and provides the feedback (sensing) signal to the AVR module causing an increase in the field current as the voltage drops under load. In some designs there isn't any real AVR, the design of the alternator provides a sagging voltage as it is loaded.

The battery charging winding is attached to a pair of diodes that provides roughly 13-15 volts dc to the starting battery if there is one..

I found this book that may be a good reference... http://www.amazon.com/Small-Generator-Service-Manual-maintenance/dp/0872888118

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#38
In reply to #37

Re: How to Build an AVR for a Three Phase Generator?

05/11/2011 4:41 AM

Thank you for your explanation. The theory is OK and I understand the processes since I have some experience with the car alternators. It's the same just the stator has an extra winding to initialize the process (and to charge the battery-if) while with the automotive installation is easy because we do have a battery. The founding and your post has encouraged me but the problem is that e-shopping does not work where I live. It would be nice to have that book, it is affordable and worth it but I think I have to ask if somebody could provide a possible usable circuit? I hope there is enough info in the previous posts somebody to be able to scan/post or email a circuit that I could build?? I bought the original AVR as well (the man who provided claimed so) which is a kidney shaped black gadget with six wires out, two separate +/- (field??) and other four at the four pole connector. The other solution is if somebody could give advice how to connect the factory AVR?

this is the link with the same avr as I got it

http://www.alibaba.com/product-gs/315164933/Generator_Spare_Parts_AVR_5KW/showimage.html

thanks in advance

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#39
In reply to #38

Re: How to Build an AVR for a Three Phase Generator?

05/11/2011 11:21 AM

I was thinking over what to do with a generator which seemed to have its "pilot" or "starting" AVR supply generator dead. Attached diagram is the result.

It works as follows (note by convention on circuits, if two wires cross without a big black blob, they are NOT connected, but if there is a blob they are connected):-

  1. D1 - D3 give a "3 phase rectified" DC supply with a mean value of about 286 V. There is some AC "ripple" on this DC, but with 3 phase rectifiers, its rms value is only about 18% of the DC value.
  2. L1 was, at my first guess, a domestic 240V AC mains tungsten filament lamp. But because 286V is too high, start with two 240V 60 watt lamps in series. This will feed about 1/4 amp to the field at rated generator voltage of 240V AC. The lamps will still glow (yellow) at half voltage, giving a visible indication of operation - usefully, they will flicker if voltage fluctuates, showing variations which will be hidden by meters, which have a slow response.
  3. If the machine voltage gets too high with the excitation delivered by 2. the shunt regulator FET1 comes into operation.
  4. FET1 will have high resistance between its s (source) and d (drain) connections, until the voltage between s and g (its gate terminal) reaches about 3.5 volts. At 4 volts g to s, it will draw about 2 amps, fairly independent of the voltage d to s. This s -d current is stolen from the lamp current, reducing the current the field gets - causing a fall in generator voltage.
  5. Resistors R1,R2 and R3, VR1 (VR1 is rotary, hand adjustable) tap off a proportion of the DC voltage (which itself is a proportion of the machine AC output voltage).
  6. VR1 is adjusted so that, with rated output, there is about 3.5 volts on FET1 gate (my initial guess was 286 volts total, make R1 + R2 total 280 kilohms, R3 + VR1 = 6 kilohms - total 280 + 6 = 286).
  7. If the machine voltage voltage increases, the increased voltage at FET1 gate draws more current away from the field, reducing the machine voltage to compensate. At 0.25 amp current in FET1, the current changes about 1 amp per 1 volt change of gate voltage. Similarly a fall of voltage will cause a compensating rise of field current.
  8. R1 R2 could be one resistor, But the usually available small resistors are only rated for 350V maximum (which is the peak voltage of 240 VAC rated voltage); there would be no margin for surges over rated value. So use two, which halves the wattage in each also, which is more reliable.
  9. ZD1 clamps the voltage g to s of FET1 to about 15 V max. The maximum rated s -g voltage is usually about 30V - much more causes permanent g-s short circuit.
  10. I missed out a spring-return push-button switch switch and 3 volt battery to flash voltage onto the field to get the machine into generation along the lines of the build-up described by RAMconsult's "post" to CR4

Before proceding to giving types and values -

  1. The field resistance you give is a little suspect (two values 24 ohm and 50 ohms) - brushes have about 1 volt drop, but resistance varies rapidly as voltage is changed. Ohmmeters apply less than 1 volt (i.e. range where brush resistance is very variable with voltage) - it would be best if you measure the field voltage with 12 V applied through an ammeter, using your two meters (digital and analog). Field Volts/Field amps = Field ohms
  2. Can you give us some idea of where you can get electronic parts, perhaps a web site. Best to suggest parts you can get easily.

Finally, regarding the 3 +1 wire connector you found under the covers. Please Measure voltage there with machine running but no field excitation - this may be the missing excitation power source. If no volts try with 12VDC applied to field. machine running.

Regards

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#40
In reply to #39

Re: How to Build an AVR for a Three Phase Generator?

05/11/2011 2:46 PM

thank you hundred times for the post! It will take a day or two of reading so I can comprehend the idea. I do have a electronic part store in the neighborhood so parts should not be a problem. I will read my homework and reply soon with the results about the last part of your post. My both instruments seems to be "Junk", I need to provide a "real one" ( I feel so stupid with all these different readings). I will have to extend the wires from the new discovered connector and measure the voltages with the engine running. I need to put the mufflers on because I am sure it will be quite noisy without them.That means that I will read your post, do the measurements of the rotor as well and let you know soon.

thank you for your efforts again

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#41
In reply to #40

Re: How to Build an AVR for a Three Phase Generator?

05/12/2011 9:43 AM

I very much doubt your instruments are junk!! And you have no reason to feel stupid!

Engineers only expect checks with a multimeter on motor/generator windings to find big problems like open circuits or major differences in resistance between windings which ought to be the same.

But always starting with the basics is good practice and essential when one begins without any labels or diagrams and cannot see what is connected together or separate. You have already found that wires may be open circuit just because they are unplugged or disconnected!

Teachers sometimes give the electrical student a "black box" with two or more terminals on it and ask them to find, by electrical measurement, what is between the terminals. There may be a transformer, a resistor, an inductor, a capacitor or combination of them in the box. It tests theoretical knowledge and use of test instruments and is good preparation for real life (where the problem is usually that one knows what is in the box, from the circuit diagram, but need to know if they are still connected and working as they should - sometimes the trouble is that something has been added).

Until you "opened the box" and found the slip-ring it did not occur to me that brush voltage drop might cause variable ohms readings [a brush-less generator would have the field winding, which the AVR feeds, on the stator] .

In any case, measuring low resistance with a multimeter is erratic and low accuracy for many reasons:-

  1. The leads from the meter to the test clips have a resistance of around 0.3 ohm or more, digital meters do not usually compensate for this. Try joining the clips together and reading the value. Analog meters usually have a "zero" knob, mostly because the changing voltage, and resistance, of the internal battery directly affects the zero and the reading on the tested resistor.
  2. The resistance of the contacts in the meter sockets and switches and the test clips or prods are no longer negligible. Try joining the test clips and seeing how the reading varies if you switch in-out of the ohms range several times or rotate the plugs in the meter sockets.
  3. Making a contact onto the tested item, which has a much lower resistance than the item becomes difficult.
  4. Any corrosion on terminals makes good contact more difficult.
  5. Your meter and leads can warm up in the sun, or cool in the wind.
  6. Electrical copper increases its resistance about 34% between zero and 100 Celsius (electrical resistance is a standard way of finding the temperature rise of windings or cables). So what you are measuring is not constant to begin with! In particular, running the machine will heat up windings due to heat from the current or the engine.
Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#42
In reply to #41

Re: How to Build an AVR for a Three Phase Generator?

05/12/2011 1:20 PM

34%??? I don't think so; R(100)= R(0)[1 + 0.00393(t2-t1), filling in 100 degrees C for t2 and 0 for t1, we get at change of R at zero times 1.039, less than 4%,, these numbers are for the International Annealed Copper Standard. See the Electrical Engineers Handbook, section 2.9.

Martin

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#44
In reply to #42

Re: How to Build an AVR for a Three Phase Generator?

05/13/2011 6:16 AM

Absolutely right! A formula of the form [ (234.5 +t1)/(234.5 + t2) ] was somewhere in mind, but wrongly recalled. I was only thinking approximate and 34% is about right for difference between ambient temperature and 100 C. Mind you, 0.00393*(100 - 0) = 0.393, equivalent to 39.3% - not "less than 4%"! In CR4 our speed of typing often exceeds the speed of our brains

67model

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#45
In reply to #44

Re: How to Build an AVR for a Three Phase Generator?

05/13/2011 8:09 AM

Blast! I sit corrected,,, I wish they'd turn the air conditioning on in my office. That's my excuse, and I'm sticking to it!

Register to Reply Off Topic (Score 5)
Commentator

Join Date: Apr 2011
Posts: 98
#46
In reply to #44

Re: How to Build an AVR for a Three Phase Generator?

05/13/2011 2:37 PM

All that math scares me to death. Anyway, I did the test as I said and these are the results:

1. The field winding ( rotor slip rings) give definitely 43 Ohms +/- 5 Ohms. I used several instruments on a 24°C ambient temperature and the readings are almost the same. The auxiliary windings (new discovered four pole connector)show very small resistance between themselves, most probably less than 1Ohm and I cannot measure the value (#36; 1 and 2). The "lonely" red wire however shows 27 Ohms measured to the each wire from the same connector.

2. With the diesel on and no external excitement of the rotor the voltages at the four pole auxiliary windings are very small, aprox 200 mV, I cannot measure accurately.

With external excitement of the rotor ( car battery) 13 V DC:

2.1.diesel off, the rotor draws 1.25 A,

2.2.diesel on it draws 1 A ( I expected the opposite?)

2.3. 5V AC between 4 wires at the auxiliary winding, with the engine spinning and rotor excited as above,

2.4. 80V AC between the red wire (alone-out from the connector) and any of the four at the connector from the auxiliary winding.

Output voltage at the power winding 280V AC!! N to any R,S,T. NICE NICE NICE

I wish to emphasize, last time I used small transformer to excite the winding of the generator and under load the output voltage dropped for 30V. I assume the weakness of the source was the reason. This time, with the car battery and steady 13V DC excitement, the output voltage did not change significantly, at least I could not see any change at the instrument even with the 1KW load ( grinder again).

I hope this will be sufficient for you experts to determine the values of the elements at the circuit. I live at the Balkans (Macedonia) so any European standard of the elements should work. The shops are well supplied with these stuff and catalogs as well so I am sure that I can provide the material.

thank you

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#47
In reply to #46

Re: How to Build an AVR for a Three Phase Generator?

05/16/2011 6:39 AM
  1. The "80 volts AC" you discovered on the red wire might be an "exciter" winding driven by permanent magnets on the rotor.
  2. The 4 wires with 5V (running) on them are a mystery at present. Do you mean you get equal voltages between any of them? Not clear what 5V might be required to power. One way to test their resistance is to connect a 12V battery to them through a 12V, 5 watt motor car "side/tail" lamp. This will deliver about 1/2 ampere, which will give about 50 mV (millivolts, 1/1000 volt) across 0.1 ohm resistance. Your digital voltmeter probably reads 200 mV [199.9] full scale on its lowest range.
  3. The way to find out about "80V" winding is to change the excitation current to the field while the set is running. Connect your 12V battery to the field through a 12 volt 21 watt motor car stop/turn indicator lamp. This will reduce the field voltage to about 6. Fit a switch so that you can short out the lamp while running - so that you can switch easily from one strength to another. It will help if you can measure the battery (field) current with an ammeter.
  4. measure 80V and 5V and the N-R-S-T main winding voltages (running engine) with maximum and reduced field current, using switch.
  5. If voltages change in close proportion to field current, (expected for R S T to N, roughly 2:1), then the winding is wound with the main windings. If it keeps same voltage you have a separate exciter.
  6. Another question. Does ohm meter test find resistance between 5V and 80 V windings and any other winding? - or are they independent? If they are independent they can be used, as we want, for AVR.
  7. Machines have separate exciters because, if you throw on a heavy load current which drops the voltage right down, this chokes off the field voltage at the same time when it is powered from the output. This further weakens the voltage, so you have a no-go situation. With pilot power to AVR, it will boost excitation to meet the demand.
  8. It has occurred to me that the new 4 wires connector you found MIGHT be CTs "Current Transformers" to indicate R, S and T current. If they are, they must operate with a load connected or be shorted-out (but if they are not CT windings, shorting them will be bad!). If they do not change their voltage as you switch the field strength, but change when applying a load to the generator, then they are CTs. The trouble is that CTs will deliver a high voltage if they have no load, which can cause their insulation to fail, or could damage a voltmeter. Best to connect 12V 5 watt motor car lamps to them.
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#48
In reply to #47

Re: How to Build an AVR for a Three Phase Generator?

05/16/2011 7:02 PM

Thank you for your post. Before I go on I have to be sure about following:

2. I will measure again the voltage at the 4 pole connector. I understand the circuit (5W lamp, 12V battery and winding in series), it is not clear to me shall I measure the voltage at the ends of the winding when all above is on (with the engine of)?

3.You want me to measure the fluctuation of the voltage (initial 80V) with the controlled (voltage/current) change of the excitement (field) and record the changes with the engine on?

4. I think this answers the question above, you need these three parameters with the two possible excitations (with details recorded).

6. Will measure, with all the three instruments!

7.I think most probably it is a different winding for the exciter but will re-check.

8.It refers to the procedure in number 1. but with the engine on with and without load?

Where do you draw the energy to deal with people like me?

thank you

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#49
In reply to #48

Re: How to Build an AVR for a Three Phase Generator?

05/17/2011 5:01 AM
  1. re- your item 2. Yes, voltage with engine off.
  2. Re- yr item 3 Yes.
  3. Yr item 4, yes please. Current value is sufficient for field. Amperes x number of winding turns defines excitation.
  4. item 8. Correct, I should have made clear.

I think most of the folks on CR4 have had the urge to make something work and have been helped with that by others.

I am onholiday till 24th so do not expect quick reply

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#50
In reply to #49

Re: How to Build an AVR for a Three Phase Generator?

05/17/2011 8:18 AM

OK,thank you and have a good time. It seems that "holiday" is the answer at my last question

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#51
In reply to #50

Re: How to Build an AVR for a Three Phase Generator?

05/17/2011 4:21 PM

Thank you for your good wishes and I wish you progress with your genset

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#52
In reply to #47

Re: How to Build an AVR for a Three Phase Generator?

05/23/2011 2:24 PM

Hi again, here I am with the results of my last experiment:

2. I connected the 5W/12V lamp as you told me and these are the readings between the wires (I named first three of them "A1" to "A3" and the separate one at the connector "N":

A1-A2=46mV; A1-A3=42mV; A2-A3=60mV;A1-N=40mV; A2-N=60mV; A3-N=58mV

This part worked well.

3.I connected the the 21W/12V lamp but it did not change the field as we expected. I could not manage to excite the field with the two different voltages/currents while the engine was running so I used two batteries, 12V car battery and 6V NiCd battery. I connected the - pole together and put a switch so I had 13VDC/7VDC excitement:

Don't get confused with the color, it is the same generator !!

At least I can paint without anybody's help!

3.1.13VDC excitement, engine running, current through the field 0.6A; A1-A2-A3-N = 5V AC between themselves and N (common at the power outlet), the "red alone" (No1 at your post) gave 75VAC, output voltage R-S-T to N = 170VAC )

3.2.With the 7VDC excitement, engine still running, current through the field 0.3A; A1-A2-A3-N = 2.5VAC between themselves and N (common at the power outlet), the "red alone" gave 36VAC, output voltage R-S-T to N = 82VAC.

(the panel voltmeter at the genset turned very inaccurate. I was using it to measure the output but for the 170VAC measured with two other instruments it showed 250V) I am not going to use the readings from the original factory instrument any longer.

4,5 and 6. I think the above shows that we have a "wound" windings because "on cold" with no excitement or engine on, there is a connection between the "A1-3-N" connector and the main windings, the instrument shows small resistance.

I found out that the engine itself has a device to generate the 12VDC for the battery charging that is completely independent from the generator. It is like at any motorcycle or other engines. It had a regulator that was sc.... so I did not have those 12VDC at the generator 12V terminal and I had to charge the battery with the external charger.

The 30VAC comes out from the engine and goes to the device and a single red wire provides flat 12VDC/8.3A, directly to the + terminal at the panel. I found a new appropriate power size stabilizer (from the ATV vehicle), put it on and now I have a 12VDC all the time. It charges the battery that I use to start the engine and it is ( I guess) possible to use the voltage and charge another battery as well.

I don't want to abandon the track that this project is going on and I think we should try to find a solution with the facts above but I was just wondering, what would happen if this 30VAC are used (for an example with the regulator) to excite the field of the generator? Aren't we doing that right now with the car battery? I am not an expert but I have a feeling if I put regulated 15-20VDC with the stabilizer to the field(fixed voltage/various current depend on the voltage drop) I will have a usable 230 VAC?

I am sorry for so many questions

Register to Reply
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#53
In reply to #52

Re: How to Build an AVR for a Three Phase Generator?

05/26/2011 2:28 PM

The short answer is yes, any source of the appropriate DC power will do fine, you basically "set it and forget it" and the generator will happily supply load all day long (in my world this is known as "Manual Excitation"). Your next task would be to start loading the generator and watch as the terminal voltage drops as the load increases, you come to a point where the voltage becomes unacceptable and/or large motor loads such as a refrigerator won't start due to low voltage. If you had an AVR it would try and maintain the voltage by injecting more current into the field windings, but since you don't you simply manually increase the current from your (adjustable) DC source until the terminal voltage is acceptable. Once you do that you start taking load off and chart the terminal voltage rise. If you're lucky you may find that the inherent voltage regulation of the generator is good (=/-5%) and all you have to do is pick a field current that produces a no load voltage that is between the full load and unloaded terminal voltage and leave it set there. If the inherent regulation is poor then you may have to babysit the field current as you increase or decrease the load off of the midpoint. Sooner or later someone will locate a circuit that you can use to provide the desired AVR functionality.

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#54
In reply to #53

Re: How to Build an AVR for a Three Phase Generator?

05/26/2011 4:38 PM

Yes, I understand that part. Thank you for your post but again, if the variable source is set to deliver the appropriate fixed field voltage for the desired output voltage and, at the same time, it is quick and capable enough to deliver appropriate current /in order to keep the voltage stability/ - as a result I should have stabilized output voltage no matter the size of the load? If that theory works I would have to experiment what would be the best inherent voltage and - no baby sitting ever after.

My point is, I can build a good 0-20V/5A regulated source and hook it to the 30V AC port at the engine? Any suggestion for that circuit while we are waiting for the "real" AVR?

thank you

Register to Reply
Guru
Engineering Fields - Power Engineering - New Member

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 1725
Good Answers: 228
#56
In reply to #54

Re: How to Build an AVR for a Three Phase Generator?

05/26/2011 9:55 PM

A fixed power supply has no "smarts" in it, it just keeps cranking out the same voltage/current up to its design rating. What I was describing was taking advantage of the regulation inherent in the design of the generator, basically more iron in the rotor and stator equals better regulation. So run your experiment and see how much the load depresses the output voltage.

You should check out this link, it might have exactly what you need ...

http://www.kitsrus.com/projects/k68.pdf

__________________
Curious minds want to know, engineering minds get answers....
Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#57
In reply to #56

Re: How to Build an AVR for a Three Phase Generator?

05/27/2011 3:43 AM

Now I see thank you for your quick response and explanation. LM317 was already on my mind and I have similar circuit but I will use this one. Seems better to me.

It is time to start soldering, I guess ! I will let you know about the results of the experiment soon.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#58
In reply to #57

Re: How to Build an AVR for a Three Phase Generator?

05/27/2011 1:07 PM

I see you have got the starter battery charging going! It is not clear where you are measuring the 30 VAC. I guess it is a winding input to the 12 VDC regulator box. There may be problems with just connecting a second bridge rectifier to feed an LM317, because there must already be a bridge rectifier (to get the DC) inside the regulator box - or even something more complicated. The will already be a connection to engine chassis via the 12 VDC output. The best guess is to connect the anodes of two diodes, one to each end of the 30 VAC winding, connect their cathodes (ends with rings) together to get 30V DC. Must go, got to go out...

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#59
In reply to #58

Re: How to Build an AVR for a Three Phase Generator?

05/27/2011 3:14 PM

Do you mean to bridge/rectify those (we call this sort of sources "wild")30V AC with two extra diodes before I hook to the actual board? And Yes, this unregulated voltage comes out from the engine winding (magnet+winding like motorcycles?) and it goes to the 12V regulator box. I cannot say how the engine generates that voltage but this is the most probable way. This installation does not have any connection with the generator!

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#60
In reply to #59

Re: How to Build an AVR for a Three Phase Generator?

05/28/2011 2:40 PM

In reply to your post #59 ---

  1. The parts inside the "broken" line are probably what is in the regulator box. The negative is probably connected by the bolting of the regulator unit to the frame/casing/chassis of the engine [ indicated by the symbol with the three horizontal lines, which means chassis, rather than earth/ground. Earth symbol has triangular outline.].
  2. The all-black diodes are those you add, your negative will be the chassis.
  3. Note, if you put a full bridge externally, those diodes "chassis connected" would be in parallel with diodes in the "regulator box". Sharing current between diodes in parallel is so " hit and miss" [without a lot of effort] that it is better to just use the [well rated] diodes which are already there.
  4. It is possible that diodes are buried in the engine and DC comes out to the regulator - a check with voltmeter [set DC, then AC] should find this out.
  5. Diodes type 1N5406, rated 3 amp 600 volts PIV (Peak Inverse Voltage) are probably OK.
  6. A warning, the high current ratings of these "wire end" diodes, with their thick wires, assumes they are well cooled by conduction through those thick wires - the wires must be short and soldered to large terminals to lose the heat - being in a blast of (cool) air helps a lot. Rating 6 amp would be better, for reliability, since up to about 3 amps field current is expected.

Going back to the AVR. This "30V AC" supply from the engine is connected to the chassis on one side.

a) Is the N terminal of the R-S-T-N main alternator connected to chassis??

b) If it is not, there may be safety problems with connecting N to chassis, which would make it easy to feed the output voltage back to an AVR/field connected to chassis via the "30V" supply. But, if the 5 volt and 80 volt (red wire) windings are not connected to output N (ohmmeter test), then one could connect their "common" to chassis and use 80V winding to feed a proportion of the machine output voltage to the AVR.

c) Your posts #46 and #52 actually give 4 values for field resistance, 10.4, 21.7, 22.3 and 43 ohms (the first 3 by volts/amps - the fourth, by ohmmeter can be ignored, due to the low voltage). It makes it difficult to decide the AVR current rating! It is possible you have brushes which are worn or sticking, or trouble with spring pressure mechanism. Best to measure field current and voltage, simultaneously, at intervals to check if value really is unsteady. Also look for signs of excessive sparking at brushes.

Finally, supposing one can connect output "N" to chassis as well as AVR/field, I can propose an AVR circuit with values, as below...

I drew the diagram thinking the 80V winding with the red wire would have to serve as excitation power. But I was worried we did not know any current rating for the 80V winding and 27 ohms felt high. I finally realized that, if the main windings of 230V were 1.5 ohms, then 80V of winding with the same wire would have resistance 1.5 x 80/230 = 0.5 ohms approximate resistance - so an 80V winding with 27 ohm resistance would have wire with 27/0.5 = 54 times smaller cross section. Finally, if current rating is proportional to wire cross section area, and main winding is 10 amp rating, the rating of the "80V" winding would be 10/54 = 0.18 amps - NO USE to supply 1 amp field.

But you can replace winding A-B of the diagram with DC supply got by diodes from engine "30 VAC" winding, and replace D6 by wire.

Note that the on-charge voltage of a 12V battery is likely to be regulated to about 14 volts by your regulator, like a car [auto] alternator.

On to component types for AVR (best to be generous on voltage/current ratings) -

  1. Diodes - type 1N5408 3 amp 1200 V PIV.
  2. ZD1, zener diode BZX55C16 [500 mW, 16V @ 5 mA] or similar.
  3. FET1, MOSFET transistor IRF640 [200V, 18 amp, TO220 package] or BUZ31 or similar.
  4. C1, 1 microfarad 25V aluminium (or better, tantalum) electrolytic. Same microfarad, higher voltage is OK.
  5. R2, R3 resistors 1/2 watt, 150 kilohms each. 5% or better tolerance, preferably metal film, although cracked carbon or metal oxide will be OK. "TRUOHM type MF60 0.6W 1% tolerance are fine.
  6. R1, as R2,R3 but 8.2 kilohm.
  7. RV1, 10 kilohm linear rotary variable resistor (also known as rotary potentiometer). Any rating, with spindle with convenient knob for trial, but finally, since vibration and temperature can cause setting to vary, you may need a high-grade one.

The on/off switch you need to avoid draining battery when engine is stopped. If you turn it on only when engine is at full speed and turn it off after you disconnect loads, before stopping, it will avoid problems. The better AVRs do not turn on excitation till generator is at half speed and, below 90% speed, reduce voltage in proportion to speed. This avoids trouble like overheating the field, due to no airflow, from battery power when the set is stopped or over-fluxing potential transformers (directly connected to main windings) at below normal frequency (big turbine machines can take 30 minutes to stop, just due to inertia). You can add these automatic features once you have working voltage regulation.

Regarding FET1, a good heatsink is essential, since power loss might be 25 watts, together with a suitable thermally conducting, electrically insulating washer kit between FET1 metal surface and heatsink (and to insulate mounting bolt from FET1 metal tab). You need a proper finned heat-sink, about 100 x 50 x 50 mm with a temperature rise of 2.5 degrees Celsius per watt to get this because the internal temperature rise in FET1 and the washer add. This heatsink will probably cost more than any other single item! If you have 125 x 125 x 3 mm pure aluminium plate available, fix FET1 at centre - warning, aluminium alloys can have much lower thermal conductivity.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#61
In reply to #60

Re: How to Build an AVR for a Three Phase Generator?

05/28/2011 4:18 PM

Thank you for your efforts!! Please, give me a day or two to read and understand your last explanations. It is very educational and challenging!

I will be back soon, most probably with just a few details to clarify .

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#62
In reply to #60

Re: How to Build an AVR for a Three Phase Generator?

05/31/2011 3:46 PM

As first thank you again for all the efforts. This is what I've done today:

1.Minus 12V/chassis/grounding terminal are all in short. N is in no connection with any of these.

2;3. and 5. I found 5A 1N5408 diodes and I will use them as you suggested.

4.It is only 30V AC coming out from the engine. NO DC!

5. I will experiment with the newly build board (with LM 317) to see what will happen with the generator and if it works I will make a good box with the cooler.

This is about the regulated "manual excitation" version of the solution. I intend to feed the board with those 30V AC (after being rectified) because with all the 13V from the battery I don't get 230V AC output.

about the avr:

a) NO, the N terminal is not connected with the chassis.

b)If I think twice (and I have) the red wire seems to me to thin to be able to deliver all the amps needed for the field anyway. Including the fact that 5V and 80V are connected with the main winding (as we said "wounded") this solution is no good??

c)I understand your point very well. Different readings frustrates me the most but I did it everything as you told me to. It is not possible to observe the brushes because the generator end is behind both the mufflers. That means I need to run the engine without silencers and I am sure my neighbor with the newborn kid wouldn't like the idea.

OK, I will put the regulated 0-30V 1.5A with the LM317 to the field. I will feed it firstly with those rectified 30V from the engine and I will measure the voltages and currents through the field and the output power! I may build a stronger version with the LM338T if all fine.

I will connect the battery only with the engine running, thank you for the detailed explanation of the boards.

I am in the process of collecting the material for your proposed AVR (MK3)circuit and any suggestion or change regarding the latest input would be nice before I go shopping.

Is MK(3) given on purpose?? It means MAKEDONIJA???

/this is our state acronym/

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#63
In reply to #62

Re: How to Build an AVR for a Three Phase Generator?

06/02/2011 6:32 AM

Hello Mishel,

I have thought some about your project before replying.

Your points up to 4. Noted.

5. LM317, a good idea to try, and the board may be useful for other things, as well as getting an excitation/ output voltage curve (which is not practical with an AVR). Such a record would be useful in dealing with another set of similar type and as a reference if you later have a fault on this machine.

This will settle the issue of how much excitation current and voltage you need to supply the loads you want to use. At present, you know that 12V on the field is about right, but it is not clear if the field is 20 ohms or 10 ohms.

An LM317 makes a good constant current source. I have pointed out before that machine output voltage depends on the field current, not the applied voltage and that the field resistance will change as the machine heats up. So ideally an excitation regulator [in contrast to a voltage regulator] delivers an adjustable current independent of field resistance and supply voltage. The top part of the diagram below shows an LM317 current regulator......

....forget the add-on at left. All you need is the LM317 and a resistor, which is 3R3 [3.3 ohms] in the diagram. The resistor value needs to be 1.25/(current wanted, in amps) ohms.

Because low value adjustable resistors are rare beasts, it is impractical to use a "pot" to change the current. The add-on at the left using a TL431, [which behaves like a zener diode whose voltage is adjustable from 2.5 volts upwards] allows a settable current from about 0.4 to 1.2 amp, by adjusting the 1 kohm pot (3R3 changed to 1 ohm). You must disconnect the ADJ pin of the LM317 from the bottom of the 3R3 and connect it to the REF terminal of the TL431. Note the LM4041 is an improved TL431 which can regulate from 1.25 Volts up, rather than 2.5 V.

One thing about LM317, its maximum input to output voltage is 40. A sinusoidal 30V has 42 volts peak and a magneto may have a greater peak to mean ratio than sine. But it is in series with the field, so all may be well. The LM317 is self protecting against overcurrent and overheating, but does not claim over-voltage protection - so if it should fail, it will probably be the voltage.

Your a), --- as expected, see below.

b) agreed, it always helps when looking at a problem from 2 directions suggests the same truth.

c) No fault on your part. Here we have something called "Murphy's Law". It applies generally, but in engineering means that anything you really need to look at is hidden behind the muffler (everything working OK is easy to get at) or the measurement you really need gives a different answer each time you test....

Concerning the N terminal not connected to chassis - I noted before that connecting it to chassis would be a safety issue. An example of the danger runs like this......

  1. N is connected to chassis.
  2. Set is connected to water pump in pool. Operator is giving attention to pump.
  3. Unknown to operator cable is damaged, exposing R conductor.
  4. Damaged cable has fallen into water at edge of pool.
  5. R is now well earthed, N terminal and all exposed metal on genset are at 230V to earth.
  6. Set is on rubber feet and wheels, so nothing happens yet.
  7. It has been raining, so ground is wet.
  8. Child without shoes, who has been paddling in water, touches set to see if it is vibrating.........

Consequently N, R, S, T are insulated from everything else, a single earth fault is harmless so long as the machine insulation is good.

Provided you understand the above, if you connect N to chassis temporarily, the AVR can be tested and proven using the MK 3 circuit with the 30V DC source (chassis connected) powering the field. From your personal viewpoint, it is safer if the AVR parts (except D1-D3, R2, R3 which can hidden under an insulating cover) are no more than 30 VAC to chassis while you are reaching into them (wearing rubber boots).

Once the AVR is working, one can try a 1:1 isolating transformer from the 30 VAC source winding or 250:30 V transfo from the main windings and "float" the AVR and field and N terminal insulated from chassis.

Does the set have built-in "30 mA residual current" earth fault protection?

If not, feeding any load you connect through the "residual current" safety circuit breakers (maximum 30mA trip current) which plug into a standard power socket is highly recommended. Where I live, the electrical regulations for buildings now require that a socket outlet provided for outdoor equipment, like hedge cutters, has such a breaker built-in.

Best wishes for further progress,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#64
In reply to #63

Re: How to Build an AVR for a Three Phase Generator?

06/02/2011 9:35 AM

Hi and thank you for your post. I should have joined this place much, much earlier!! It is very educational and fun doing all this things.

I will (again) print and read your post before I comment or ask something.For the time being I know for sure following facts because today I've been busy with the generator. I wish to share the founding before I go on with the next step.

I tried the LM317 board made as the recommended project in the pdf file and I fed it up with the 30V AC through 2 diodes as told. It simply did not work. I then disconnected the 12V regulator because I doubt that the regulator somehow might affects the voltage. I directly put those 30V AC from the engine and through the full bridge rectifier fed the board with 30 V DC directly. The output from the LM317 board went directly to the field. I don't care for the battery charging right now, it is the goal to achieve steady 220V AC output.

With 24 V DC in the field I got nice 220V AC output. With 1KW grinder on as a load the output voltage dropped aprox 10V. As I said the panel voltmeter at the generator set "lied" to me about the output voltage. Today I corrected the scale according the digital instrument and (finally) the information is accurate. So 12V not enough, 24V field = fine. No current measurements. I am building a box with two high grade instruments (Volt and Ampmeter) just for this project. I will be measuring the voltage and current with different loads and let you know. Now the output voltage could be regulated (with the knob at the lm317board) and I could get from 0 to 260 V AC.

I understand (initially) your point about the current and I am ready to build it up but that resistor scares me as hard to find. I will check all the elements in the store.

We are not abandoning the other possible solution with the "real" AVR?

N is not connected with the chassis right now and as I can see there is no need to do it?

I am not an engineer, just a technician but about Murphy, we say

"If something might go wrong - it WILL go wrong"

I will be back soon

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#65
In reply to #64

Re: How to Build an AVR for a Three Phase Generator?

06/02/2011 12:22 PM

Thanks for new info about required field voltage. I see you sorted out the supply problem to the LM317 board well!

I most certainly have not abandoned the real AVR .

If you are not worried about the battery charging, a possible way is to use the 30V winding to feed the AVR direct and, later, use a transformer from the 30V winding to charge the chassis connected battery as another project.

Resistors of 1 ohm usually come in the "wire-wound power" type such as :-

http://www.rapidonline.com/Electronic-Components/Resistors-Potentiometer/Power-Resistors/5W-Radial-ceramic-resistors/66474

or http://www.rapidonline.com/Electronic-Components/Resistors-Potentiometer/Power-Resistors/Aluminium-clad-wirewound-resistors/65217

The second type can be screwed onto a chassis as a mounting/heatsink [stops them getting "hand-burning hot" at a few watts] and soldered to without needing extra terminals. Also they are "military" type, well tested for damp and hot conditions. So the extra cost compared to the ceramic type soon pays for itself when you are installing in a control panel.

Sometimes one puts, say, 3 of 3R3, 1 watt resistor in parallel, to make 1R1 3 watt.

Finally, I meant to reply to your mention of the "MK 3". Where I live, aeroplanes etc come in "Mark 1", "Mark 2" versions, abbreviated to Mk 1 say. Hardware like that often gets up-graded during life and it is the practice to fit identity plates with a row of numbers 1, 2, 3... in a grid. When it is upgraded, the 1 is marked (crossed-out) with a tool, hence the description "mark", I guess. It saves changing the plate, which could be a deadly source of error, say putting the wrong plate back on an engine, getting serial numbers in error, even running a Mk1 engine at Mk 2 power! So, that is why my 3rd proposal was MK3.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#66
In reply to #65

Re: How to Build an AVR for a Three Phase Generator?

06/04/2011 3:46 PM

Hello again,

regarding your last two posts (63 and 65) I would like to clarify some details:

1.About the LM317 that now works as a "voltage regulator" at my board, changing the voltage at the field I change the output voltage. I understand your point that the current is more vital and it's value should be regulated (by an avr) depending on many factors (heat, load...). Does it mean I don't need the voltage regulation at all? As I understood the circuit after several readings, using a simple LM317 with just a high power resistor (10W or more, applying the math with the 1.25 constant) as an ex.= 3R3, the given simple devise (without the add-on at left) would deliver the constant 0.38 Amps in the field/no matter the factors mentioned above. Does it mean the input 30V DC would not cause very high output voltage and I don't need a part that actually regulates the voltage coming in? If so (theoretically) experimenting with the working value of the resistor I could found a solution without the TL431.

I intend to build the board as you suggested anyway, I just want to understand the possibility. I will use the TL431(or better LM404-if I find). Can I use stronger LM338T instead LM317 without any changes?

Is the resistor below the pot 2K7?

I intend to solve the battery charging otherwise after all this done.

I don't think that the generator has "30mA residual current" earth fault protection. I assume the function but have never seen or heard the standard. I will google to learn that safety standard.

2.All concerns related with N terminal noted. I understand the possibilities and always try to keep Murphy away, so, if and when I use the second board (step 3 MK3) - I will be careful.

Next week I need to travel to the capital to by all the elements since the shop here does not have the most of the components. Please, if you have any other suggestions before I go shopping, do not hesitate to order in good time

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#67
In reply to #66

Re: How to Build an AVR for a Three Phase Generator?

06/05/2011 2:44 PM

In reply to your post #66 Mishel,

Your 1. You are right, if the current is regulated, the 30V input does not need to be voltage regulated. You are also right about finding the required value, then using a fixed resistor.

I did not think simple enough, you can do things without TL431 [TL431s are useful, being like an adjustable zener, with a better regulated voltage]. The LM317 operates to always keep its ADJ terminal 1.25 volts negative to its OUT terminal. So if you put a 1.5 volt torch cell, (with positive to the end of the field connected to the LM317) you will have -1.5 V available.

Tap off 1.25V of this with a potentiometer resistor, 1k or more, 10k would be fine, one track end to each battery pole, connecting the potentiometer moving wiper to ADJ via 1 k fixed resistor [1k is just a precaution, because ADJ terminal might take high current if you force it to -1.5V], then, in theory, the LM317 current could be brought down to zero (actually LM317 needs about 5 mA to function, so you will not get zero).

For example, if you set pot to get -1.00 volt, there will be 0.25 volts across the current setting resistor instead of the usual 1.25V and consequently 1/5 the current. And an "AA" size Duracell (alkaline) 1.5 volt torch cell will feed a 1k load for 500 hours above 1.25V!!

Since you can adjust the voltage, you can be relaxed about the resistor value, which is as well, because low value resistors are difficult. You need less than 1 ohm, to be able to get 1.2 amps. I have used coils of hook-up wire. 50 metres of 1 sq mm copper wire is about one ohm, good for up to 10 amps . Do you have some old "door bell" wire around, it is about 0.5 mm diameter copper, 0.2 mm2, so you need 50 x 0.2 metres = 10 metres?? Or telephone cable?? Remember you just need to measure the field current with your meter, the resistor value is not needed.

If you need a small amount of variation, +/- 10%, you can use (in place of the battery) a diode or, better, a transistor (base tied to collector) as an 0.6volt [at 25 Celsius]voltage regulated source (working like a zener diode shunt regulator, with - 2mV/degree temperature influence), fed by a resistor (e.g. 4k7 1/2 W) from + 30V or common. Note also, the usual small silicon transistor, around in broken radios, TVs and such, has an emitter-base diode which makes a good zener diode, of about 7.5 V, just try them till you get a good one (most are good).

If you want to try this, best to decide the resistor ohms value [1.25/current in amps] you need before your Oddysey to the Capital, after measuring the field current which gives 240V generator output.

The "resistor below pot" in circuit diagram was 2k7. But if you got 2k4, that is closer to the ideal 2k5 ("ideal" is based upon 2.5V, which is the REF voltage value in the TL431).

Since European Electrical Codes [for electric power in buildings] are now mostly based on the same document, translated into local language, you may find "Residual Current Device ", [RCD in english] in Part 2 "Definitions" and clause 412-06 and others, should you have a copy of such Code at your employment. The term "Earth Leakage Protection" might also apply. On a single phase device, the RCD sees, for example, 10.03 amperes going out on the live wire and only 10.00 amperes coming back on the neutral (30 mA difference) when it trips - but that the 30 mA is going to earth via a person is not certain - it might be leaking to another circuit, so the "RCD" description is more accurate than "Earth Leakage".

I hope you have a successful visit

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#68
In reply to #66

Re: How to Build an AVR for a Three Phase Generator?

06/06/2011 7:15 AM

Further on post #66

LM338T - higher current is fine but looking at Linear Technology data MAX voltage In to Out = 35V. National Semiconductor data 40V. In view of your 30V rms AC input, 35V may not be enough, best to be sure you get 40V.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#69
In reply to #68

Re: How to Build an AVR for a Three Phase Generator?

06/06/2011 10:02 AM

Ok, in short that means No LM338T because of just 30V AC in?

I am reading about the RCD standards and your last explanations. Before I start to build another board which should be flexible and accept changes, I am in a middle of search for any 10W resistors that might be useful (in between 1 Ohm and 5 Ohms) so after I found the material needed I will have the idea how to solve the board. I will travel this week to the capitol anyway and until then I should know, hopefully, what is missing and what is to be bought. I understand your explanation about how the TL431 part of the board serves to avoid /hard to find and match/ high power resistor and how to make a 1 Ohm resistor if the LM431 is used . As I said, I will spend some money and time to find components and after the "Oddysey" I will let you know about the results. My idea is to move the generator to the summer house where the closest neighbor lives a mile away and I can do some dynamic tests with the boards and the new built control box.

Until then, thank you and take care

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#70
In reply to #69

Re: How to Build an AVR for a Three Phase Generator?

06/06/2011 3:52 PM

In response to Michel's post #69 ------------

Ref. LM338T, the possible problem was :-

  1. The 30 VAC, if sine wave, would be 42.4V peak.
  2. LM317 is rated 40V input to output, but did not fail.
  3. One Manufacturer gives 35V and another 40V for LM338.
  4. 40V is same as LM317 - but would dropping to 35V be a "move too far"??

So the only problem was 30 VAC input, if that was what you meant by "just 30V AC in".

The real unknown is the wave form of the 30V AC. You would need an oscilloscope to find that - or, it is possible to get programs [free trial] which use the sound microphone or Line inputs of a PC as a useful oscilloscope - most useful if it is a laptop, since the microphone input is not tied back to mains earth. An alternative, for peak voltage, is an OP amp in a level detector circuit which turns on an LED lamp when its input voltage exceeds the reference voltage from a potentiometer.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#71
In reply to #70

Re: How to Build an AVR for a Three Phase Generator?

06/06/2011 5:26 PM

Thank you for your quick response

Yes, I thought the problem would be 3O VAC input ("in"), anyhow I understand your point and I will hold to the LM317. I really couldn't go that far into it and try to find out the wave form. Sorry for distraction from the main course

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#72
In reply to #70

Re: How to Build an AVR for a Three Phase Generator?

06/14/2011 6:42 PM

Hi everybody, I've been busy these days and I apologize for being late. Today I've been working on the project and I wish to share with you my findings:

1.As first, friend of mine brought the book that you have recommended. It is good to have it.The first ten pages have good hints about generators and their construction and the rest of the book covers various types of generator sets with schematics and things. I could not find any part that would throw light on the AVR, they are all represented in block diagrams.

2. I build the "current" excitation board. It was easier for me to find a bunch of strong resistors to experiment with the value needed to mach the 24oV AC output. It is good to have a friend in a local Radio-Amateur club?

You can see the new board at the left and the existing "voltage" regulator at the right side of the paper.

It simply did not work. Even with a resistor as small as 0R5 I got just 30-40 V AC output.

I am a little confused with the input on 2 at the LM317 but I think I build it as you said. Not much at the board but again I maybe s.... it somewhere? I tried several resistors and the generator was not excited enough to get output. Anyhow, on Friday I go to by the TL 431 and build the complete board.

3. I was curious and put the "voltage" regulator back and - voila

with 20V DC , 0.9 A delivered to the field I got nice 230 V AC output.

I put the drill on and it worked, I intend to try bigger loads like compressor and the welder tomorrow. I hope I wont damage something.

In short, I will test further the "voltage" regulator and push harder to see what happens, I will provide elements needed for the full version of the "current" regulator and the MK3 AVR as suggested.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#73
In reply to #72

Re: How to Build an AVR for a Three Phase Generator?

06/15/2011 9:33 AM

Hi, Mishel,

Apologies for the trouble with the current regulator! Most DC regulators are intended to work from smoothed DC supplies, similar to batteries. The circuit I gave is as Fig.4 of the data sheet :-

http://focus.ti.com/lit/ds/symlink/lm317.pdf

- and I forgot that an input capacitor and preferably an output capacitor are needed as shown in fig. 1 and its notes A and B [the circuit of your voltage regulator included those, they avoid the LM317 going an on/off oscillation behaviour]. Also an inductive load (the field) is a problem - what probably happened was that the LM317 was switching ON/OFF at such a high rate the current in the inductive field never got above a small value. With 12V battery input and capacitors and a 12V lamp load, the LM317 would be OK. What it needs with the field would be a resistance in parallel with the field, so the LM317 can draw enough current to keep its internal circuits working, say 1000 ohms 1 watt, or two 2k2 1/2 W in parallel.

Since the 30V AC engine winding and bridge rectifier have been chosen to get DC for the field, it would be worth adding a reservoir capacitor. This would help with the voltage regulator, because LM317s do not expect to be fed from a supply which falls to zero twice each cycle - they usually expect an input always 2 volts higher than the required output.

For test purposes, to get well over 20V, you can put a 12V battery between the field negative and the negative of the bridge rectifier/voltage regulator. I will send diagrams later, but must go now.

N.B. The AVR circuit I gave does not have to have a pure DC supply.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#74
In reply to #73

Re: How to Build an AVR for a Three Phase Generator?

06/18/2011 4:57 AM

Hi 67model,

Now I understand that the voltage has to be "ironed", please advice about the possible solution. I went to the capital yesterday and I have all the components for the mark3 avr and if you have any final suggestions it would be good to know before I start working on it

thanks and have a good day

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#75
In reply to #74

Re: How to Build an AVR for a Three Phase Generator?

06/18/2011 6:43 AM

Hi Mishel,

Here are the diagrams. The effect is like putting two 12V batteries in series to get 24V etc. - but one battery is made adjustable.

V and I are the voltmeter and ammeter. Capacitor C needs to be 3000 microFarads [50 V DC or more voltage rating]. You could put 3 of 1000μF each in parallel. Note 3000 μF would give about 3 volts peak to peak "ripple" with 50 Hz supply and 1 amp load - actually your engine generator may be higher frequency, which would give smaller ripple. N.B. Using a digital multimeter, it is usually possible to measure the DC voltage across C - then switch to AC volts and measure the mean ripple, the indicated AC ripplevalue will be around 1/4 the peak to peak.

Diode D1 [1N5400 or any higher volts of 1N5400-1N5408 family] is ESSENTIAL. When the engine is stopped, current will flow from the battery through the field and through the regulator in the opposite direction to that the regulator is designed to take! Unless there is another, easier path, as provided by D1.

The next diagram shows the effect of the capacitor C on the output voltage. A bridge works the same way [with an extra diode voltage drop], without needing a tapped transfo secondary.

Finally, you will probably find it difficult to get currents over 1 amp, because the LM317 will current limit [or overheat and current-limit]. A solution would be to fit one or two 12 V/ 5 watt motor car lamps across from IN to OUT of the LM317 (they will draw about 0.4 amp each at 12V). Also, be careful you set the regulator to get 20V or more before you connect the lamps and that the IN - OUT voltage of the LM317 is likely to be less than 14V, else the lamps will fail. Since you actually need 1 amp field current for normal 230V AC, a permanent "ballast" of resistors may be useful, to relieve the AVR of some of the heat. Last point - consider fitting a fuse in the feed from 30V winding.

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#76
In reply to #75

Re: How to Build an AVR for a Three Phase Generator?

06/18/2011 1:42 PM

67,, no no,, D1 will be forward biased and the battery will discharge through D1 and the exciter field.

Martin

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#77
In reply to #75

Re: How to Build an AVR for a Three Phase Generator?

06/18/2011 5:01 PM

Hi 67,

I assume this is the second version of the diagram as on your post 63 using LM317 and the 10W 3R3. I understand the diagram but I have a couple of questions before I start building it:

1.I would kindly ask about the proposed value of the other resistor and detail if it still stands that in is 2; out is 3 and 1 would be the adj with the (still) unknown resistor that goes before the 3R3?

2.Will the battery be charged automatically and what if it is week and draw higher current from the 30V winding? While experimenting with the battery attached (with the "voltage adjusting" board) I noticed significant voltage drop from the 30V to only 12V and something, depending on the state of the battery. This board means "no battery- no output"!!

3.Any comment about D1?

4.Does the "Mark 3" still stands as operative and good to be built as parallel project to the board above? I am willing to skip this and get to the main!!

regards

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#78
In reply to #77

Re: How to Build an AVR for a Three Phase Generator?

06/19/2011 2:38 PM

Hi Mishel.

In reply to your post #77,

First paragraph, the block marked REG is the LM317 Voltage Regulator you built - which works successfully to vary field current. The terminal numbers were from one data sheet, but use the connections you have for IN(put), OUT(put) and ADJ(ust) you used for the "voltage regulator".

Your question 1 needs no answer now.

Your question 2 - With about 1 amp needed for the field, I would expect battery charging to work well enough. Once a battery is well charged, it draws little current after it has been on charge for a short time. But I thought you wanted to get the excitation working and worry about battery charging later - a battery charger to work from 230 V AC can always be arranged.

Your question 3 - baxterm in post #76, is absolutely right and I am glad he is keeping a "eye" on things! However, in post #73, I only meant that connection to be a very temporary one to apply greater current than 1 amp to the field coil (needing more than 20 Volts), because once big generator loads are applied more excitation voltage and current will be needed.

Your question 3 - Certainly AVR MK3 stands [noting A - B and D6 would be the "30V" winding and bridge rectifier] and no problem if you try that now. It could be improved by feeding RV1 from a voltage regulator, removing the battery voltage variation effect on generator AC output voltage.

The object in being able to vary the actual field current is to get 1) a generator no-load output AC volts versus field current graph and 2) a generator short-circuit output current versus field current graph (nearly a straight line, so one or two test points are enough).

Given 1) and 2) [which are standard tests for AC generator performance], one can estimate the field currents needed to supply loads of different sizes and power factors.

The tests take a little time, but one then has a reference point in future problems and for similar machines. Also, it is less trouble than trying out various loads up to maximum output, since load banks are seldom available!

N.B. Doing a test with the generator output terminals [R-S-T-N] shorted together sounds alarming, but, if you run a machine up from standstill like that, it happens without drama, and it could be run continuously at rated current.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#79
In reply to #77

Re: How to Build an AVR for a Three Phase Generator?

06/19/2011 6:13 PM

Mishel,

AVR mark 4 diagram - For the avoidance of doubt.....

I have drawn this with the "30V" engine winding and bridge rectifier. Also I have added the reservoir capacitor C. Resistor "r" in the source (s) lead of FET1 is an addition. It should be 0.25 ohm approximately. It adds "field current feedback" and also reduces gain. It may be needed to avoid voltage instability - all may be OK without it - but best to introduce the idea now. The problem is the field inductance, current change lags behind voltage change - maybe a good part of a second before current settles after a step change of voltage.

I just remembered an important point ZD1 anode must be returned to the S terminal of FET1, not common, if r is inserted.

Regards,

67model

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#80
In reply to #79

Re: How to Build an AVR for a Three Phase Generator?

06/19/2011 7:38 PM

67, looks good, except I think you have D1,2,&3 drawn backward.

Martin

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#81
In reply to #80

Re: How to Build an AVR for a Three Phase Generator?

06/20/2011 6:33 AM

Hi Martin,

R1, RV1 [to +12V] bias FET1 gate positive, turning it on. Negative voltage from D1-D3 pulls gate negative when AC output voltage is high enough, reducing FET1 drain & excitation current to bring AC mean voltage into control with gate at about +3.5 volts for IRF640 FET. Actual control point depends on individual FET [threshold 3, +/- 1 volt for IRF640] and battery voltage, as well as R1 + RV1 value. The +4.5V marked on diagram was an extreme value to decide R1 value.

67model

Register to Reply
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 8
#82
In reply to #81

Re: How to Build an AVR for a Three Phase Generator?

06/20/2011 8:32 AM

Ah, I misunderstood your cct, i see now.

Register to Reply Off Topic (Score 5)
Commentator

Join Date: Apr 2011
Posts: 98
#83
In reply to #79

Re: How to Build an AVR for a Three Phase Generator?

06/21/2011 5:57 PM

Hi 67model,

thank you for both the circuits. Now I have homework to do. Just before I start please clarify the position of ZD1 with the r/0.25ohm inserted and if the reservoir capacitor value is critical?

thank you

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#84
In reply to #83

Re: How to Build an AVR for a Three Phase Generator?

06/22/2011 4:57 AM

Best place for ZD1 is at FET1, directly between its gate (g) and source (s) pins. Cathode (coloured band end) to gate. Reservoir capacitor value is not critical - the usual tolerance is +/- 20% of nominal capacitance anyhow ---- 1500 to 10000 microfarad will do fine. 67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#85
In reply to #84

Re: How to Build an AVR for a Three Phase Generator?

07/04/2011 9:18 AM

Hi 67model,

I've built the MK4 AVR and it does not function. I checked and rechecked the board and connections several times but the generator does not get any excitation and output voltage. The board gets the 30V rectified and filtered but the field terminal is simply "stone cold".

The only thing I did not do was the battery connection ( I tried the board without the battery charging) but I think that is not a problem.

Please clarify the post 75 and the function of the D1 (1N5400) so I try that board.

thank you

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#86
In reply to #85

Re: How to Build an AVR for a Three Phase Generator?

07/04/2011 3:33 PM

Welcome again Mishel! I was worried you had met an electrical accident! Apologies for the problem. It is essential with the Mk 4 AVR to have a +12V supply at the junction of D4 and RV1.

Without that, there is nothing to bias FET1 gate (g) terminal positive to turn it ON. FET1 will not conduct until its gate is about 3 volts positive to its source (s) terminal. No current through FET1 = no field current.

You can get the +12V by just connecting a battery or by feeding your LM317 VOLTAGE regulator board from the +30V supply, adjusted to give about 12V output.

The good thing with the voltage regulator is that it will be a constant 12V, unaffected by the state of charge of a battery or change of the 30V supply (down as low as 12V input to LM317 anyhow, if you adjust the regulated voltage to 10V). The voltage regulator is also a permanent way to get +12V, while the battery was just a quick way to get 12V when I drew the diagram first - since then you have built the LM317 voltage regulator.

Post #75 was a way to get a hand-adjusted field current, for test purposes, for a few minutes or hour only. It did not matter that the battery would discharge in some hours (through D1). The purpose of D1 was to stop negative voltages at the LM317 output (when there was no voltage at its input terminal), which might damage LM317. As I wrote, if you can make, from test results a graph of generator open-circuit voltage against field current and a graph of generator short circuit current against field current, the field current for any load can be estimated.

Must finish now, I have a job to do.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#87
In reply to #86

Re: How to Build an AVR for a Three Phase Generator?

07/07/2011 12:26 PM

Hi again 67model!

As you can see I am still breathing and ready for (many) questions! I could not resist laughing about your worries regarding possible electrical accident. Whenever there is a major blackout in the neighborhood the first thing that all the neighbors do is they look if I've been playing with my gadgets at the balcony!! They always first ask me if I have something to do with the power shortage.

I don't know why such an idea come cross their minds??

1.About the post 75, understood, thank you, I put that one on a waiting list.

2.I used the lM317 regulator to drive the FET1 and I succeed to excite the generator. It gives 250-260 VAC output but without possibility to adjust. Turning the pot 10K on the MK4 or/and the pot 5K on the Voltage Regulator (used to bias the FET1 gate) does not affect the output voltage.

3. I decided to take a risk and hooked the air compressor. This device did not get enough power last time when I used just a Voltage Regulator to excite the rotor of the generator. Since it is a demanding load I put it on and this time it started and worked well and the output voltage of the generator dropped to 230 VAC under load. No smoke, no overheating elements or signs of possible trouble, even just for couple of minutes spinning.

The problem that worries me is the fact that when I disconnected the compressor as a load /while the generator was still on/ the output voltage of the generator rose above limit of the panel voltmeter which is 260 VAC. I did not want to go further more and risk experimenting with some other loads like welder or similar. I assume it would be better if and when we succeed to stabilize the output power regardless the size of the load connected/disconnected.

Should I increase the value of the POT 1OK or/and r1 which is 0.33 Ohms now?

I think we are close to finish this project successfully...

thank you

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#88
In reply to #87

Re: How to Build an AVR for a Three Phase Generator?

07/07/2011 4:44 PM

From what you write, that the excitation does not fall even with LM317 voltage regulator set to minimum (which should be 1.25 volts) it is possible that transistor FET1 has failed short circuit from s to d. This should be easy to check with an ohmmeter, a value of a fraction of an ohm means death.

Some questions to know what we have now.

  1. Do you have reservoir capacitor C on 30V DC supply?? If yes, what is its value in microfarads (μF)?? What voltage do you get on C, at what field current??
  2. What voltage do you get at the anodes of D1, D2, D3 - relative to common?? Should be about -280 volts, with half that at the junction of R2,R3.
  3. Is FET1 on a heat-sink??
  4. Is diode D4 "OK"?? If it were short circuit, or the wrong way round, you would get + 30V DC to RV1 (once the engine ran up) - regardless of setting of the LM317 [which would mean RV1 at maximum would not get you voltage regulation].

Good luck!

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#89
In reply to #88

Re: How to Build an AVR for a Three Phase Generator?

07/08/2011 9:02 AM

Regarding your last post:

FET1 seems to be ok. I checked and compared the readings with the new(spare) one and they are the same. Just s to dr in one way conducting, no other readings, so I think this should be ok.

1. The DC supply includes full bridge rectifier and reservoir capacitors 470µF and 2200µF in parallel, total of 2670µF. It shows 32VAC when engine on, I don't connect load any more ( I don't want to burn something because the higher voltage output) so I cannot tell the field current. This 32VAC are with no load.

2. There is 240VAC at the D1, D2 and D3 to common and 120VAC measured at the spot between R2 and R3.

3. Yes, the FET1 has mounted alu heat-sink.

4. Diode D4 is ok and not turned way round, the anode at the + terminal of the LM317 voltage regulator and the cathode to the positive field connector.

I tried it several times and checked the connections and wires but it's always the same story. The voltage is too high and no regulative. I will think about the board and what could be the possible mistake that I make.

Any suggestion appreciated

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#90
In reply to #88

Re: How to Build an AVR for a Three Phase Generator?

07/08/2011 12:22 PM

Sorry, I rechecked the components again and it turned that the voltage regulator didn't work properly. The LM317 has failed and I wonder is it possible that FET1 drew more current that the voltage regulator can deliver? I replaced the LM317, the voltage regulator works fine again and I am ready to do the test but will wait for your answer about the expected bias current first.

thank you

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#91
In reply to #90

Re: How to Build an AVR for a Three Phase Generator?

07/09/2011 6:32 AM

This is in reply to your posts #89 and #90. I was replying to #89, so I will let what I wrote stay below :-

<< In reply to Mishel's post #89 :-FET1 seems OK.

Your item 1. --- Capacitor value is good. Description of voltage as 32VAC is confusing, because rectified voltage should be DC. You should use your meter on "DC Volts". It may indicate if set to "AC Volts" and connected to a DC voltage - it might give different readings if you swap the meter probes over, to reverse the DC polarity. In any case, when a multi-meter reads AC, it usually does it as a DC meter which gets the mean value of the rectified voltage when it works on AC [unless it is labelled "True rms", and costs a lot more] and multiplies that mean value by 1.11 to get an indicated value which is correct for the "rms" value of a sine wave alternating voltage. In summary, If you measure DC volts using a meter on the AC range, it will probably read 11% high!

Your item 2 ---- The expected value is -280 volts DC (negative!!), with 240 volts rms sine wave output by your generator [and I saw from the image of your component board you posted that the diodes are the right way round]. So you should get DC, unless 2 of the diodes are open circuit and one is shorted (please check with ohmmeter, just to be sure)! So your VAC is again confusing, and the value seems low even if your meter was on an AC volts [V~] range. A possible, but unlikely, reason for a low value is that the generator gives an output which is not a sine wave, but something like a sine wave with the positive and negative peaks "flattened" down.

Your item 3,4 --- Good. >>

Regarding post #90, FET1 gate cannot draw more than a few milliamps, because R1 limits the current, even if FET1 gate were to fail and short to drain (if FET1 is good, its gate current is so small as to be difficult to measure). This would not damage LM317. It might have been damaged by overcurrent when supplying the field, even though its overcurrent and overheat protections are good. N.B. I am assuming you have a reasonable heat-sink on the LM317.

I suggest the following tests, which can be done "on the bench" away from the genset......with reference to the AVR Mk 4 circuit.

  1. Get your 12V starting accumulator battery and, preferably, a second 12V battery to put in series with it (negative to positive) - to get 24V DC. If you do not have a second starter battery, 3 "AA" alkaline dry cells or a 9V transistor radio battery will serve to get over 15V.
  2. Connect this battery to your LM317 voltage regulator input. Adjust the voltage adjust pot to get maximum voltage. You should get well over 12V out. Reduce the output voltage to 12.0 V. N.B. You cannot rely on the LM317 giving a regulated voltage out unless you have an input voltage more than 1.5V above the required output voltage at no-load and 3V at maximum current.
  3. Connect your 12V output to the common (-) and RV1/D4 junction (+) of the AVR(Mk4). The voltage to common (-) of the g (gate) terminal of FET1 should be nearly 12V (check with voltmeter).
  4. Connect the cathode of D4 (D4/D5 junction) direct to the input battery (input of LM317). Note that D4 will be reverse biased and will not conduct.
  5. You should now be able to adjust your "12V" supply from minimum (1.25 volts) to maximum and get almost that voltage at g of FET1.
  6. Set variable voltage to minimum. Connect a 12V 1 watt lamp in place of the "field" of the generator. Increase the voltage until the lamp just begins to glow red (this will be a current of about 0.04 amp, 40 milliamps and 2 volts for a 12V 1 watt lamp) . Measure the voltage at FET1 gate (with care! maximum gate voltage is 20V - any more, e.g. accidental connection to 24V, could destroy it. ZD1 is there to protect gate from overvoltage but cannot fight a 24V battery and win). It should be in the range 2 to 4 volts, typical 3V per IRF640 data. This is the "gate threshold" voltage given in the FET1 data sheet. It should be sharply defined as you increase the gate voltage, increasing from red to full brightness for around 0.1 V change of LM317 voltage. N.B. If your battery is weak, the regulated voltage might drop, check the LM317 input voltage is 3V greater than output. You could use a higher wattage 6W lamp for test if your battery is strong.
  7. Remove lamp, adjust voltage regulator for 12.0 volts DC. Adjust RV1 to maximum resistance.

The preceding tests will show that the LM317 voltage regulator is working, set it to 12V before use on engine and show that FET1 is working and measure its approximate gate voltage at which it will work as a generator voltage regulator. Setting RV1 to maximum resistance ensures the engine runs up with voltage regulated at (should be!) well below normal 240V. When running, if voltage regulates OK, you can increase voltage to normal with RV1.

Good luck,

67model

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#92
In reply to #91

Re: How to Build an AVR for a Three Phase Generator?

07/09/2011 12:40 PM

Further to 91.

That you were able to start your compressor with 30V DC available for field (albeit with cold field winding) is good [Aagh! Sorry - albeit = English word that means "though" and probably began as "all be it" - your English is so good I forget to keep it simple].

The voltage jump when throwing off that compressor load is expected. The long time delay before voltage falls back to initial no-load value is quantified as the "field open circuit time constant" of the generator. It has a value in seconds. The good news is that the time is much reduced when there is still a load on the set.

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#94
In reply to #92

Re: How to Build an AVR for a Three Phase Generator?

07/09/2011 7:20 PM

67model

You keep answering my posts howbeit I always report unexpected readings

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#95
In reply to #92

Re: How to Build an AVR for a Three Phase Generator?

07/18/2011 3:54 PM

Hi again 67 model,

I am little late with the response because we were hit by a hit wave from Africa and it was 40°+C outside, not a good weather for work outdoors. I have removed the generator to the remote cabin where I can do the things in proper way. The engine was wormed up first, the electronic was spread and easy to control and various load were prepared for the experiment. These are my latest results:

1. Regarding your post 88 I did the measurement again and I have to apologize for bad presentation from the last time, it is U=35VDC at the reservoir capacitor at the DC supply.

2.Voltage at the diodes is U=-272VDC and at the junction of R2,R3 U=-135VDC.

Regarding your post about how to test the LM317 I have not done the bench test but I replaced the burned LM317, checked it with the instrument prior the actual hook up with the MK4 and it worked fine. Further more, I hooked an instrument that will measure the regulated voltage that comes between the pot and the common all the time with engine on.

I rechecked the pot at the MK4 (10K) and it worked fine "on cold" just like other readings of the elements used.

With the diesel on the pot at the LM317 board changes the voltage from 1VDC to 25VDC but it does not have any effect at the output voltage which is constantly 240VAC. The field always gets 30VDC no matter the voltage rate given by the LM317 board and the voltage at the gate at the FET1 was always 2.5VDC.

The 1oK pot at the MK4 does not change the output voltage neither.

Further more, I disconnected the + from the LM317 regulator that goes to the anode of the D4 to see what happens and it happened nothing. The output voltage remained the same 240VAC and the voltage at the gate at the FET1 was always 2.5VDC..

With all these the generator can feed smaller loads like drill, grinder etc but it cannot supply enough power for utilization of welder or air compressor. The voltage drop was significant and it went down below 150 VAC with the demanding load connected.

Frankly, I was sick and tired and decided to disconnect the MK4 and the LM317 board. I hooked the rectified 30VDC directly to the field terminal curious to see what will happen and again-it happened nothing. I got again the same 240VAC output with the same behavior of the generator regarding bigger inductive loads as with the both boards hooked before.

Is it possible that this 30VDC supply is not strong enough to give all the amps as declared (let's not forget it is Chinese and it's possible that declared power is a little inadequate)? I have been wondering how the rotor can not be excited enough with ALL the rough power from it (8.3A).

I have an idea to try with two car batteries hooked serial and see what will happen if the rotor gets 24VDC from a supply that can "deliver" currents. Both batteries would be 50 Ah?

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#96
In reply to #95

Re: How to Build an AVR for a Three Phase Generator?

07/19/2011 9:29 AM

Hi Michel.

I suggest you follow the tests "on the bench" suggested in post #91. This will prove the gate voltage can be varied etc.

But I wonder if you have the zener diode ZD1 (cathode [colour band] to gate) between drain and gate of FET1, rather than source to gate.

Running, FET1 would finish up with about +3V (you get 2.5V) on gate relative to source, about +3.6V on drain. That leaves about 30V from the "30V" supply across the field if there were 33V on reservoir capacitor C. Trying to vary gate voltage via R1 would have have little effect on field current, nor would feedback of output volts via R2,R3 do anything.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#97
In reply to #96

Re: How to Build an AVR for a Three Phase Generator?

07/19/2011 10:00 AM

Will do the "bench test" and recheck the zener diode!

thank you

best regards

Mishel

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#98
In reply to #96

Re: How to Build an AVR for a Three Phase Generator?

08/11/2011 3:24 PM

Hi 67 model,

I was away from home for quite a while. I am ready to continue with the project now! Will be back with the results of the bench test soon.

regards

Misel

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#93
In reply to #91

Re: How to Build an AVR for a Three Phase Generator?

07/09/2011 7:05 PM

67model

I need to apologize asap for using VAC instead VDC in my post!! The value of the readings were as written in the post, I just made a mistake writing AC instead DC, anyhow, I will read again all your later posts and follow your instructions. It will take a day or two to get back to you with the results.

thank you for being patient

Mishel

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#99
In reply to #91

Re: How to Build an AVR for a Three Phase Generator?

08/17/2011 1:30 PM

67 model,

I tried the bench test several times but it seems I've made a mistake somewhere. It is OK up to step 3. and the 12V DC ends at the 8K2! If I disconnect the R1 from the point where it contacts the gate and the voltage is measurable at the end of the R1, once R1 connected to the gate - the voltage drops to 0. I did not go any further, I should check again the board, there is a short somewhere I guess.I will also change the zd1!

if I don't get any results - will make another board, :) I don't give up!

regards

Register to Reply
Guru

Join Date: Dec 2010
Posts: 791
Good Answers: 56
#100
In reply to #99

Re: How to Build an AVR for a Three Phase Generator?

08/18/2011 6:58 AM

Post #100....must be a record?

From what you write,

  1. FET1 gate is shorted to its d or s.
  2. or there is an accidental short in wiring.
  3. or ZD1 is short.
  4. or C1 is short.

You must just disconnect each one in turn from R1 and test for short.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 98
#101
In reply to #100

Re: How to Build an AVR for a Three Phase Generator?

08/18/2011 8:31 AM

I am not sure that I should be pride with this "100th post" record, it has taken a lot of your energy/time and I am still struggling with shorts.

You definitely deserve a big thanks for being so patient with me!!!

will be back soon, hopefully with not shorts....

Register to Reply
Register to Reply Page 1 of 2: « First 1 2 Next > Last »
Interested in this topic? By joining CR4 you can "subscribe" to
this discussion and receive notification when new comments are added.

Comments rated to be Good Answers:

These comments received enough positive ratings to make them "good answers".

Comments rated to be "almost" Good Answers:

Check out these comments that don't yet have enough votes to be "official" good answers and, if you agree with them, rate them!
Copy to Clipboard

Users who posted comments:

67model (50); A.A.Khi (1); Anonymous Poster (1); baxterm (9); Lare Dusun (7); Mishel (71); nesubra (1); RAMConsult (7); roymor (3); sude (3)

Previous in Forum: Transformer Flux Linkage   Next in Forum: Electrical Motor Test Stand
You might be interested in: Motor and Generator Winding Services, IC Waveform Generators, PCI Products