Previous in Forum: Transformer Flux Linkage   Next in Forum: Electrical Motor Test Stand
Close
Close
Close
Page 2 of 2: « First < Prev 1 2 Last »
Commentator

Join Date: Apr 2011
Posts: 99

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!
Commentator

Join Date: Apr 2011
Posts: 99
#101
In reply to #100
Find in discussion

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
Commentator

Join Date: Apr 2011
Posts: 99
#102
In reply to #91
Find in discussion

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

08/24/2011 3:25 PM

Hello again 67 model,

I have just finished the "bench test" and found the FET1 burned. I have changed it with the new one. All the steps from 1 to 7 have been followed and all went as you have described in your post. Just for 6.step the 6W 12V lamp started to glow at 4V and it can be regulated but just with the pot at the LM317 power supply. When I measure the voltage at the gate of FET1, there is a voltage drop and the bulb light decreases. Changing the value of the RV1 does not affect the brightness of the lamp nor the voltage at the field terminals. I measured the value of the RV1 and the connections at the board and it seems as it is according the schematics. If this is OK and expected, I can hook the avr MK4 during my field trip this weekend and see how it works "together" with the genset.

thank you for your time

regards from Macedonia

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#103
In reply to #102

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

09/12/2011 6:28 AM

Hi Mishel, I have returned from holiday.

All you report is OK. It is not expected that RV1 would have any effect, because the current through RV1 is very small - only the leakage current of FET1 gate, the zener diode and C1 capacitor. Connecting a digital voltmeter - usually 1 megohm resistance - would have a small effect on FET1 gate voltage. about 10.000 ohm for RV1 + fixed resistor versus 1 megohm causes about 10,000/1,000,000 about 1% volts drop in 4 volts i.e. 40 mV. The change in FET1 drain current versus change in gate - source voltage is about 1 amp per volt, which means about 40 mA current change in lamp for 40 mV gate voltage change. As you wrote, this causes a visible change of lamp brilliance.

I have thought of something I have not mentioned, since I have been doing it for so long.

When attaching FET1 to a heat sink it is essential to cover metal base of FET1 with silicone/zinc oxide mix or similar grease to conduct heat from FET1 to the heat sink. Without this, temperature rise will be too much and FET1 will overheat. Alternative is to use a modern heat-conducting washer made for the job. You may be able to find a washer in a failed audio amplifier or TV which used the same physical size transistor (usually known as TO220 - TO stands for Transistor Outline). View the following as example....

http://www.rapidonline.com/Electronic-Components/Semiconductor-Hardware/Transistor-Mounting

The grease is better if there is no washer (insulation of FET1 from heatsink not required if the heatsink itself is mounted with insulating pillars or similar), because adding a washer usually adds its own temperature drop per watt of heat. If you use mica washers as insulation for FET1, the washer needs the grease on it.

I wish you success with your next test.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#104
In reply to #103

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

09/12/2011 8:57 AM

Hi again 67model,

I am glad you are back. Hope you had a good (and long) vacation. I already started to think that you gave up on me:) I read your post and have learned something new again! I have not used any of the recommended procedures but sure will do in order to avoid any next burning of the(by the way expensive) FET. So,to make it short, I will provide the FET with a good contact at the heat sink, do the test again and let you know about the results.

regards

Mishel

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#105
In reply to #103

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

09/20/2011 4:20 PM

Hi 67model,

I did the test again and this time it worked just as you had described in your previous posts. I put good heat sinks to both FET1 and voltage regulator, hooked up all together and the regulation worked just fine. Both pots changed the voltage and I fixed the output voltage at 230 V. I tried several loads at my summer house ( lights, TV, Stereo and even a camp stove with oven which is almost 3.5KW) and the AVR was doing the job properly. No overheating, no voltage drops or surges. The engine did not even change the RMP significantly while the voltage was +/- 5% fluctuating under load. I just thought that my "Never ending story" was history but when I tried my welder the regulation failed again. The inductive load, I assume, has caused a problem to the AVR and again, something burned. Turning the pots now does not have any influence at the voltage which is constantly 230V AC but it does not provide the current needed. The fact that I had used the same arch welder machine with my neighbor's HONDA 3.5KW (5.5HP engine) without a single problem encouraged me to hook it to my (claimed) 5KW diesel. The Honda doesn't even have an electronic avr, just condenser and diodes at the rotor but it does the job while my Chinese is still fighting back.

I will do the "bench test" and diagnose the problem.

any thoughts, if you quit replying my post I will understand you...

Regards

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#106
In reply to #105

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

09/24/2011 1:55 PM
  1. Sorry to hear of problems. I wait to hear what failed.
  2. Guess it was FET1 - but did it fail "short circuit source to drain" or another way. FET1 can stand a high current at a low source-drain voltage or a high voltage at an amp or two. It is both for too long which leads to failure.
  1. The usual field currents and voltage should not be a problem, but a welder gives a bad combination of a high current followed by sudden interruption of current.
  2. The stationary windings of the generator and the rotor windings behave as a transformer. Considering one stator phase e.g. A, depending on the radial position of the rotor, the transformer coupling to the rotor can be high [rotor winding and phase A winding axes coincide] or low [rotor winding axis perpendicular to stator A winding Axis].
  3. So if the phase A stator current is high and is suddenly interrupted, being an inductive circuit, the current will not stop instantly, it will find the easiest path to keep flowing in - it it seems as if it were the AVR and rotor!
  4. In these cases it is necessary to provide a path - trying to stop the current instantly just causes a higher inductive voltage surge, which leads to failure of something - even insulation.
  5. In the normal way of a mains supply, there are lamps and water immersion heaters and electric cookers and fluorescent lamp power factor compensation capacitors - all of which can absorb a sudden surge of current without a voltage surge more than a few times normal, they also absorb the energy dumped by the inductor (and the resistive loads burn it up as heat).
  6. So just having a single load on a generator can be a bad situation. Crudely, having a "2 bar incandescent heater" [resistive] on while the welder is used may provide the "sink" to absorb the surge energy and make them short-lived.
  7. The inductive surge is not unlimited in energy - a given volume of iron can store BH/2 joules. For the usual irons, B cannot exceed about 1.5 Tesla - and that requires lots of H -ampere-turns.

So await result of failure analysis.

regards; 67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#107
In reply to #106

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

09/25/2011 4:59 PM

Yes, it is the FET1 that failed, it is short circuit source to drain but I got also weird reading at the gate as well. I will go to Capitol during the next week and buy a couple more FETs so I can compare the readings and tell you for sure.

Anyway, I understand your explanation from 1 to 6, while 7 is a little to much , I remember Tesla=induction, joule=energy but that's where it ends. I understand that the unit has a "φ" factor that shows the capability of the apparatus to "cope" with inductive loads. It did not come across my mind to load the generator with two different loads at a time. It makes sense to me (now) but than I thought it will be "easier" for the generator to work with the single (inductive) load.

I spoke with my Godfather who's been in electronics for ages and he suggested 2µF block capacitor at the output in order to compensate the inductive pick.

In your opinion, is there a way to protect further more the FET1 and add some more elements that will improve the resistance of our avr and protect it from surges? I could build a new board with the voltage regulator on it and apply the (eventual) new solution.

thank you

regards

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#108
In reply to #107

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

09/29/2011 6:24 AM

A 2 μF capacitor on the phase you attach welder is a good idea - it gives a current path for load surges other than the generator.

Item 7 --- A good lamination steel can achieve a field strength [B] of 1.5 Tesla with 1000 amp-turns per metre of magnetic circuit [magnetizing force H]. It is well saturated at that H and it is about the maximum field strength in a generator.

So BH/2 is 1.5 x 1000/2 = 750 Joules

[it is all metre-kilogram-second (MKS) units, so you know energy must be in Joules = watt-seconds! Much better than inches/pounds, lines/square inch...or even cm-g when it is "head scratching" work guessing the units of energy].

A cubic metre of iron is about 8500 kg - so the energy per kg is 750/8500 = 0.09 J/kg approximately.

The purpose of this 0.09 J/kg is that, faced with a transformer or a contactor which needs spark suppression for contact protection or radio-frequency interference purposes one never gets an inductance value to help size the suppressor (due to saturation, an inductance value would help little). But you can always weigh the item, assume 75% is iron, rather than copper, and use J/kg and size your voltage dependent resistor VDR etc joules rating from that.

There is another way of expressing BH/2, since B = μH, that is ½μH². Looks like ½LI², inductor stored energy or ½CV² capacitor stored energy or even resistive loss I²R does it not?

But once you are in magnetic saturation, μ is not useful and a B versus H table or graph is the way to do it, so BH/2 is used.

I think a current limit circuit should make FET1 survive, I will do some diagrams and send later.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#109
In reply to #108

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

09/29/2011 8:43 AM

OK and thank you for your detailed explanation. The new FET1 and the board are waiting for the updated circuit.

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#110
In reply to #109

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

10/07/2011 7:04 AM

Hi Mishel,

New circuit is below....

I hope it is all readable. The changes are R4, new value 0.33 ohm 3 watt and new parts TR1, R5, R5A which current limit FET1 current. Transistor TR1 is any small silicon NPN junction transistor similar to BC107. R5 is 1k and R5A about 8k -adjust to get 2 amp current limit.

Will send more explanation later.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#111
In reply to #110

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

10/07/2011 1:24 PM

Hi 67model,

thank you for your efforts. The new circuit looks good and I am looking forward for some details before I start soldering. The explanations of the elements are small and it would be nice if you could increase the font or something. I tried to magnify but it is not readable again (blur).

How do I regulate the current limit? Must be under load I assume?

Regards,

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#112
In reply to #111

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

10/07/2011 7:00 PM

Sorry, resolution on CR4 is not good, I tried myself. I have expanded the new part of the circuit below.

It looks clear at this size. The new parts are..

  1. Resistor R6, 1 kohm, between C2 and ZD1.
  2. +12V DC regulated is provided by an LM317L, which is the low current version of LM317 in the TO92 small plastic transistor case. Fixed resistors R1, R2 will give 12V near enough. You can use your existing 12V regulator, of course!
  3. R4 will give 2 x 0.33 =0.66 volts drop at 2 amps.
  4. A small NPN transistor, used for TR1, will have a base - emitter voltage Vbe of about 0.6V [600 mV] at a collector current of 1 mA [the data sheet for a BC107 gives 620 mV typical at 2mA, with a tolerance range of 550 to 700 mV]. Each 25 mV change of voltage will change collector current by 2:1 - so the BC107 Vbe value will be about 600 +/- 75 mV at 1 mA.
  5. This amount of current in TR1 would over-power the 1 mA approximate coming from VR1 making FET1 gate voltage zero which will turn it off.
  6. But at the normal maximum current in FET1 of about 1 amp (determined by the DC supply voltage and the field winding resistance), the voltage across R4 will be about 300 mV - about 12 steps of 25 mV, for each step of which TR1 collector current will halve - reducing TR1 collector current to 1/4096 mA = 0.25 microamps - which will have negligible effect on FET1 gate voltage.
  7. So summarizing items 5 to 6, at normal FET1 currents, TR1 will have no effect, while at a higher current it will conduct, pull FET1 gate voltage down, giving a current limiting effect. In fact, the gain of TR1, acting with the sharp change of current as FET1 gate voltage is changed, makes the combination into a very good current source, FET1 current changing very little as FET1 drain voltage is changed.
  8. The temperature of TR1 has an effect on the current limit, because its Vbe falls about 2 mV for each degree increase of its temperature. But TR1 will be almost at local air temperature because less than 4 volts gate voltage, and , say, 0.25 mA is only 1 mW - which will only heat up TR1 junction by 1 degree.
  9. You can check the current limit "on the bench" by applying a 12V battery supply in place of the "30V" supply and an ammeter in place of the field winding. It will be OK with R5A left out. Connect a 3k3 resistor across C2, to simulate the current normally drawn by the two 150k resistors, when the generator voltage is present. With VR1 at maximum resistance, FET1 current may be limited by inadequate gate voltage, but as VR1 is reduced, the current will rise and then settle at the current limit. FET1 will get rather hot, so best not to have current on for many seconds, not knowing how good your heat-sink is on FET1.

It is late so I will finish now, read it tomorrow!

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#113
In reply to #112

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

10/08/2011 2:23 PM

Hi again,

the heat sink on FET1 is quite good but I will be careful. I intend to study your last post and build a completely new board with the over current protection. It will take time but since the weather has turned bad I'll have long afternoons and a good homework to keep me busy. Most probably I will have some questions meanwhile :)

thank you so much,

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#114
In reply to #113

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

10/09/2011 7:14 PM

I read my last post - I left out some things I should have written.

item 6 - should have written "about 12 steps of 25 mV below 600 mV,".

item 8 - should have added "It might seem a "non-perfect" current limit when it changes with temperature. However, since the current, for the same voltage, is proportional to the heating watts the FET gets, and the watts the FET can stand before it gets too hot falls as ambient air temperature rises, a current limit which falls with ambient looks good [in fact, some current limits could bolt the current limit transistor on to the heat sink of the protected transistor, so that the current limit tracks what the protected device can withstand]."

Anyhow, I was going to give an explanation of why a current limit protects FET1.

  1. You have seen evidence that the generator field winding is not just like a resistor - when heavy loads are thrown on, power is being fed back from the generator into the field winding circuit and into FET1 - causing it to overheat and fail.
  2. If the voltage fed back into the field is such as to reduce FET1 drain voltage to zero, this will reduce the voltage across FET1 and also the current it takes - which is not going to cause FET1 to overheat and fail.
  3. It is possible that FET1 drain voltage can be driven negative to the common. But in that case, the drain to source diode inside FET1 will conduct in its forward direction and limit the voltage to about 1 volt. This diode has a current rating of about 30 amps continuous so is not going to fail easily.
  4. So, feedback from field with polarity as examined in 2 & 3 is not going to damage FET1.
  5. If the feedback has the other polarity, however, it increases FET1 drain-source voltage and its watts loss for the current FET1 can take [which depends on its gate-source voltage].
  6. There is a limit to how high FET1 drain voltage can go - if it exceeds the supply voltage (about 30V, as you know) then diode D5 will conduct with a volt-drop of about 1 volt, clamping the voltage to little over 30V.
  7. So, feedback with polarity examined in 5 & 6 will give a limited voltage on FET1 - but the current FET1 will take will depend on its gate-source voltage.
  8. Unfortunately, when a heavy load like the welder is thrown on, this causes a large generator voltage drop which reduces the negative voltage fed back to FET1 gate - making it more positive and increasing the current FET1 is able to draw. So FET1 is operating at about 30 volts and a higher current.
  9. FET1 can survive up to 150 Celsius junction temperature. Its data sheet gives 0.8 degrees/watt junction temperature rise above its metal tab and typical 0.5 degrees/watt rise of tab above heat sink clamped tight with conducting grease between - that is a total of 1.3 degrees per watt heat loss in FET1. With the heat sink at 25 Celsius, the allowable rise is 150 - 25 = 125 degrees - so the allowable watts is about 125/1.3, near enough 100 watts (for a while, till the heat sink heats up!).
  10. So with 30 volts applied, the maximum current FET1 can sustain without failure is about 100 watts/30 volts = 3.3 amps.
  11. The heat sink temperature is not going to stay constant with all this heat fed into it, but the numbers in this case are favourable - a heat sink 5cm x 5 cm x 0.5 cm thick is 12.5 cm³ and for pure aluminium its heat capacity is such that 96W of heat will heat it up at only 3 degrees per second. I realize the heat sink is bigger than that, but for more than 2.5 cm from FET1 the temperature rise due to heat conduction in the aluminium will become important [For Al, the temperature gradient is 43 degrees/cm at 100W/cm² heat flow, considering the metal surface area of FET1 is only about 1 cm², one can see that the heat conductivity does have an effect].
  12. The length of surge to worry about is, I guess, less than 0.1 second, so the actual power FET1 can survive, due to the thermal capacity of the heat sink, if the heat sink temperature is only 25 Celsius at the start of any surge is much more than the "DC" continuous value.
  13. FET1 does have some heat capacity of its own, but it is only specified as 0.5 watt-seconds to heat it from 25 to 150 Celsius - e.g. 50 watts for 0.01 second.
  14. Considering the above, I have set the current limit at 2 amps to provide a "safety margin". It will not affect normal steady operation, because 30V supply with 25 ohm field only allows about 1.2 amps steady state.

Finally, there is the issue of how much current FET1 could draw. To see this, one must look at the data sheet typical gate-source voltage against drain current graph.

The vertical scale does not reproduce well, but the bottom number is 10° = 1 ampere Drain current, while the next is 10¹ = 10 amperes [horizontal scale numbers are 2, 4, 6, 8]. One can see that on the 25 °C (junction temperature) curve, about 0.5V increase takes the current from 1 amp [about the normal field current] to about 5 amps, while 0.85 volts increase would give nearly 10 amps. In steady operation of the AVR, FET1 gate voltage is about +3.5 volts to common, that is 12 - 3.5 = 8.5 volts below the +12 volts supply. If the generator output voltage, represented by the -280V DC output from the 3 diodes D6,D7,D8 connected to the stator, falls 10%, the 8.5 volts will fall in almost the same proportion, to 7.65 volts, which will mean gate-source voltage rises to 3.5 + 0.85 volts. As noted, this can give a drain current of 10 amps - which at 30V supply would pump 300 watts into FET1, overheating it within milliseconds - hence the need to provide a current limit circuit for FET1.

Regards,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#115
In reply to #114

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

10/10/2011 5:12 PM

It is obvious that you are good at what you do!! I've read your explanation several times and I think I get it. At least most of it I will start with R5A=8K2 and place TR1 close to the FET1 heat sink at the new board. I will see how the current behave and try to implement the theory and build the MK5.

It wouldn't be me if I don't ask question

1.I am now sure I understand about "OK with R5A left out" Do you mean to remove it (disconnect) and than proceed with the procedure explained in #9?

2.The current limit should be set at 2A exactly?

3. The value of C2=1μF?

thank you 1o³ times

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#116
In reply to #115

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

10/11/2011 7:27 AM

If you did not understand part of the explanation, I did not do it well, so ask about the parts you are not sure about.

Starting without any R5A would limit the current at the lowest value, so it would be a safe way to start.

As explained, TR1 base voltage will be around 600 mV when current limiting. For a given value of collector current, TR1 base voltage will fall about 2 mV/degree increase of its temperature. Since 2 mV in 600 mV is about 1 in 300, TR1 current limit will fall by about 30% for 100 Celsius rise of its temperature.

If TR1 were glued onto FET1 metal tab, it would follow FET1 temperature, but not instantly, because its plastic case is [relative to metal] a thermal insulator. On the other hand, if FET1 were dissipating 30W, its junction would be about 25 degrees above its tab - if heat sink were at 25 degrees, junction would be 25 + 25 = 50 degrees. Raising heat sink temperature 100 degrees would make it 150 degrees - the maximum rated. So FET1 safe dissipation falls from maximum to zero for [roughly] 100 degrees rise, but TR1 temperature effect only reduces current limit 30% compared to an ideal 100% per 100 degrees rise.

C2 is still 1 microFarad.

The current limit does not need to be 2 amperes exactly.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#117
In reply to #116

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

10/11/2011 2:05 PM

sorry , no questions no more!! I get it. Better to start working on it.

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#118
In reply to #116

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

11/28/2011 2:32 PM

Hello 67model,

Sorry for being late (again). The preparations for coming winter, trip out of town, bla bla...here I am again with the new board according to the last schematic. I've decided to start with 8K2 at the BC107 and I did the "bench test". So far - so good. The heat sink at the FET1 is 3.5x3.5cm and I am not sure it will be big enough or I should put the other, bigger one that I have (15cmx10cm). During the (short) bench test the FET1 went quite hot while the Voltage Regulator LM317 was stone cold. The board is now waiting for a decent box with the cooler and the actual test at the generator set.

It would be nice to have the complete last schematic in a good resolution

thank you

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#119
In reply to #118

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

11/29/2011 4:50 PM

Hello Mishel,

Did you mean a bench test as post #112, item 9 ??

Did you measure the current? What was it? Best to start without R5A (R5A = infinity ohms!), just R5 - which gives lowest current.

Since you are using TO220 size LM317, it does not need a heatsink for this application.

Much safer to use big heatsink 15 x 10 cm for FET1, 2 amps at 12 volts is 24 watts. The small sink must be about 10 degrees/watt (including grease and junction to metal tab of 1.3 degrees). At 24W that is 240 degree rise in theory, so FET1 is staying alive on the heat capacity of the metal.

It may be convenient to scavenge a heatsink with 12V DC fan from a computer Pentium type microprocessor. For test, putting heatsink and FET1 (vertical) into a glass bowl, filled nearly up to the FET1 source/drain leads with drinking/rain water is useful N.B. For about 80 years big radio transmitter valves working off 10,000 volt supplies have been water cooled (via pipes of insulating material), pure water is a reasonable insulator, so it will not matter if you get some water on the FET1 leads, or even get them underwater.

Trying to get better picture, but not good.

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#121
In reply to #119

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

11/30/2011 3:19 AM

Thank you for the quick response and the new schematics. It is much better now.

About the test now: I did not do the complete test with measurements as explained in #112, the FET1 got hot and I did not want to risk and waist time and money ( I need to travel 100km in order to buy the new FET :) I checked only shortly if the (remaining old) voltage regulator (with 5k pot) regulates the field voltage. But Yesterday - I did the test at the generator directly and guess what? the avr worked just fine. I left the engine to run for 20min without load, the both pots were changing the voltage and the FET1 temperature was OK. I tried several resistive loads, not bigger than 2KW, and everything was cool.The temperature of the FET1 under load was again ok, I could touch and hold the heat sink.

But I will play safe. I will make another bigger box with the cooler and change the heat sink with the much bigger one, pull out the R5A and leave only the R5 as you said. I will hook the instrument box and measure both the voltage and the current at the field with many loads including the welder. I just hope it won't take another month again

thanks a lot

Mishel

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#120
In reply to #118

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

11/29/2011 7:06 PM

Try this for diagram.... looks better, I enlarged all small text. But original file is much clearer.

Register to Reply Score 1 for Good Answer
Commentator

Join Date: Apr 2011
Posts: 99
#122
In reply to #120

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

12/20/2011 12:58 PM

67model,

I followed your instruction and put the FET1 on the big heat sink in a box with cooler.

I hooked the AVR to the generator using the "control box" in order to monitor the values of the current and voltage at the field,

-voltage is ok, the regulation is ok, the current at the field 0.7A voltage 24 VDC, with resistive loads connected it works fine, the stove 2KW does not change the values and the output voltage is 230 VAC....the AVR can regulate the output voltage both with and without load. I could not see that the field current changed once the load was on. The FET1 is cool and it seems that it can work in this regime for a long time without problems (It worked 30-40 minutes smoothly)

BUT - the same story again, with the resistive load the generator behaves OK, once the welder is used the FET1 burns. It does generates usable voltage after but the pots do not regulate the voltage any longer and the spark, even with the thinnest electrode rod - not usable.

I changed the FET1 and now waiting for your opinion.

Any other approach that might be considered? I have Briggs and Stratton generator 3.5KW, single phase with the condensers only that give better performance. Is it possible to use the same method and hook condensers parallel to the RST windings of this generator, maybe just for the welder application while for other loads the AVR might be used?

thank you

regards

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#123
In reply to #122

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

12/21/2011 6:25 AM

Most sorry to hear of your trouble.

  1. You did not report what value the current limit on FET1 actually gave. It is important the current limit is low enough to keep FET1 within its limits
  2. How did FET1 fail? Resistance source - drain? Resistance drain-gate?
  3. I am thinking that 3 lamps in series of 12V 21 watt [motor car turn flasher] type connected field to negative [FET1 source to drain] would give enough field current ( around 0.75 amp at 3V each, 9V for 3, visible red colour) for normal volts no-load, cold machine (at standard brightness, they would absorb 1.75 amp at 36 volts and they should be durable at 42 volts).
  4. Try them on their own, without AVR - see how voltage goes with resistive load, then the welder. Take note of what using welder does to their brightness - which will indicate any field current surges and when they occur.
  5. Using the AVR with the "lamp ballast" source-drain is possible.
  6. Have you tried capacitor in parallel with welder?

Also ZD1 anode should go to source of FET1 not common negative, my mistake.

Best wishes for a Happy Christmas and a Prosperous New Year,

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#124
In reply to #123

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

12/23/2011 2:39 AM

1. As you said I started without 8K2 (I did not put R5A) at the base/emiter at BC107.

2.FET1 failed source-drain-gate (all to all )

3;4 and 5- I am not quite sure, are you saying just to hook those three lamps in series without the actual board (no rectifier at all?)

6.I did not put any additional capacitor. Should I add it at the board (was is it 2μF?)

I will change the ZD1 position

Merry Christmas

Best wishes from Macedonia

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#125
In reply to #123

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

02/19/2012 6:01 PM

OK, I assume it's time to recap the thread:

I did the last change of the ZD1 and the AVR is functional. The output voltage is adjustable and the fluctuation is minimal (aprox +/- 5%). I have tried again with various loads except the welder. It starts and feeds the air-compressor with 1HP motor, 3KW stove, power tools and other loads without burning the FET1. I have tried up to 4.5KW load at a time. No problem at all.

This board has put the generator in function, otherwise it would have been useless. I could not find any factory AVR that would match/work.

It is quite enough for me and I give up experimenting with the welder. I am happy with the accomplished.

I can recommend the board to be used for a 5KW three phase generator with brushes if there is an external voltage (to be used as excitation of the rotor).

I thanks to 67model and admire his endless patience. Without his expert's advices during the project I wouldn't do much.

Mishel

Register to Reply Score 1 for Good Answer
Power-User

Join Date: Oct 2008
Location: Thousand Islands, Ontario, Canada
Posts: 187
Good Answers: 9
#126
In reply to #125

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

02/19/2012 6:05 PM

Mishel, thanks for the followup, I have been following this thread from the beginning.

Cheers

Martin

Register to Reply Off Topic (Score 5)
Commentator

Join Date: Apr 2011
Posts: 99
#127
In reply to #126

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

02/20/2012 2:24 AM

Welcome

Register to Reply Off Topic (Score 5)
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#128
In reply to #127

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

02/25/2012 5:37 AM

I am glad the AVR was useful at the end and you think your persistence was worthwhile.

I hope the design will be useful to others.

Your thanks are appreciated.

Regards,

67model

Register to Reply Off Topic (Score 5)
Participant

Join Date: Mar 2018
Posts: 1
#172
In reply to #125

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

03/14/2018 3:35 PM

Its an old thread, but ....

you wonder why your fet getting dead while using the welder ?

well, whats the difference in terms of load condition .. a welder producing high load pulses .. resulting in the Fet getting fequently open and closed, with the inductance of a generator this will be sufficient to produce enough energy to lift the voltage of 32V to dangerous levels .. there is no real load on this powersource consuming this buildup.

i would say to prevent that, use a Transistor with an zenerdiode limiting the Powersupply Voltage to 37-40V this Transistor need to be cooled of corse, and the SOA has to be taken in account while choosing an proper Transistor.

my guess is while soldering there has to be 100W disipated.

an SOA of 200W will be in the secure side. There are a lot of Transistor available in this range. Any NPN with an beta of 30+/50V and an 1W zenerdiode will do the job

screw the transistor to the generator chassis will be enough to swallow the energy.

If you chose an allready insulated Transistor, your able to go arround isolation.

But, take care about the Generator windings at all with those kind of loads, they will heat the whole generator more up than a normal constant load, as there are higher peaks, and the fast changing of the Magnetic flux which all will even producing more heat.

AND: think about an 5Kw 3 fase generator is not able to deliver 5Kw on only one or two phase, he may getting into magnetic saturation, raising the temperature in the iron, the windings getting hot as well and the mechanic stress on the windings could cause some unpredicted failures.

Welder sessions longer than 20 minutes should only done on Generators which provide more than 3 times the maximum Power of the welder.

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#173
In reply to #172

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

04/04/2018 6:40 PM

Phunny,

I agree a welder is a heavy load - tending to a short circuit, low power factor, with high stator current. When the field winding couples with the high stator current, transformer action will put high currents into the field winding.

These currents will want to take the easy path through the FET and associated circuit. In one direction, thanks to D5 diode, high drain voltage will be prevented - in this direction the current limit circuit should prevent > 2 amp drain current, but there is still 32V or more supply volts, so 60 watts loss in FET. In the other, current can flow through the inherent inverse diode drain-source in the FET, but limited only by the field winding resistance.

60 watts at 2.5 W/'C = 150'C, so that's junction at max temp. 2.5'C/W from memory is 125 x 125 x 20 mm sink with heavy extruded fins; to which 1.3'C/W junction to sink must be added. It is not continuous safe - unless perhaps a PC type 12V cooling fan, but needs vibration mount for on-engine.

As you point out, even if AVR survives, a welder can overheat the stator/rotor if it is more than a fraction of gen rating.

I tried to make a simple, easy to understand, AVR. However, the usual commercial solution is a thyristor (SCR), since these are very robust, ON voltage is low - even anode overvoltage just fires them on & they survive if the circuit is inductive.

67model

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

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 3230
Good Answers: 444
#174
In reply to #173

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

04/04/2018 8:28 PM

Hi, I've been following this thread for quite awhile (is it really 7 years old??), and I'm coming to the conclusion that it's a combination of basic machine theory/design that's causing/reacting to the welder load.

This is a self-excited generator with no external power source for the field. That means two things, the output of the excitation winding is a function of the airgap flux, and there is no way to support the field voltage/current if/when the airgap flux is suddenly decreased. So when the airgap flux is suddenly reduced as it tries to supply the stator with enough flux to be converted into nearly 100% inductive current, the voltage across the excitation winding drops, the airgap flux continues to drop, then the output voltage of the generator falls further, and the output races to the bottom. In IEEE excitation modeling terms, there is no field-forcing source available to hold up the voltage/current required to adequately supply the field and subsequently the airgap flux drops further.

This one of the purposes of the PMG (Permanent Magnet Generator)/pilot exciter found on larger separately-excited brushed/brushless excitation systems. It supplies full power to the field regardless of what happens to the rest of the generator as long as it continues to turn. The easiest way around your problem may be to put a 36 volt battery in parallel with the excitation winding, but then you'll need additional circuitry to charge the battery and regulate its output, the design of which I'll leave in 67Model's capable hands.

__________________
“Tell me and I forget. Teach me and I remember. Involve me and I learn.” Ben Franklin.
Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#134
In reply to #115

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

04/01/2013 5:18 AM

hello I just learned about this avr, I want to try stringing avr MK5, but I do not know the transistor and fet1 in use.please its support part number that is in use in transistors and fet1thanks before

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#135
In reply to #134

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

04/01/2013 8:18 AM

Hello Lare Dusun,

you should go through all the comments and read details about the process of building this avr, it's all in there :) good luck

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#55
In reply to #53
Find in discussion

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

05/26/2011 4:47 PM

By the way, I've found a way to order the book you recommended!! It's on the way!

thank you again

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#43
In reply to #41
Find in discussion

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

05/12/2011 1:55 PM

Thank you for the explanation. I've tried to "zeroize" the analog instrument before each measurement with the knob but again I always got "close" but different readings. I posted the average reading calculated from at least five single tries. The digital one shows 2.5 to 2.9 Ohms at short connection so I deduced the average 2.7 from the readings. I am glad that my "black box" is no secret any more and further more I am glad that my generator has slip rings. I read somewhere that those machines are better for inductive loads comparing with the brush-less ones.

I will measure the resistance on the auxiliary winding ( even it's very low), the rotor (again) and do the dynamic measurements with the engine on and let you know.

thank you

Register to Reply
Member

Join Date: Nov 2012
Posts: 5
#129
In reply to #12
Find in discussion

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

11/03/2012 1:01 AM

i have a rotary converter generator (208v, 400Hz, 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.

previous avr 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 and exciter field voltage

at no load

exciter field current = approx ( 0.3 Amps)

exciter field voltage = approx ( 36v-38V)

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#131
In reply to #129

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

11/03/2012 6:18 AM

The trials and developments with Mishel went on for a long time and ended up with the following as a workable, simple but not perfect AVR. It worked OK with normal loads but broke if a welding set was used as the load.

You start with knowing which terminals are the field etc!!

You have, I guess, a motor - generator set in which the generator is driven by normal power of 50 or 60 Hz. This makes it easier, because you can get the excitation DC supply by using a normal mains transformer to get the AC voltage at WNDG on my circuit diagram above.

The first step would be to apply a 12V car battery to the field winding and see if you can get output voltage from the 400 Hz generator output when it is rotated by the motor. In case there is a diode across the field winding, it is necessary to check with the battery and a lamp or resistor to find if there is a diode

However, it would be good first to hear from you if I am on the right track, what your requirements are for the final set performance, and how much experience you have in electronic construction.

Good Luck,

67model

Register to Reply
Member

Join Date: Nov 2012
Posts: 5
#132
In reply to #131

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

11/04/2012 12:22 AM

sir you are in a very correct track. sir i have done my diploma in electronicsbut am not such a bright student.sir thank you for replying and helping me out.

1. i had used a dc power pack to inject the exciter field current after starting the converter set. i got rated output from converter 208v, 400 h, 3 phaseat 0-25 percent load. during which the dc power pack was reading 36v-38v voltageand .3amp-.32amp current.

2. their is no diode across the field winding. the diode was incorporated inthe previous avr module.

3. to construct the circuit i need to know the wattage of all the resistors, specifications of diodes, FET, transistor, zener diode

4. i will study the circuit and try to understand its functioning clearly.

5. sir in my case the output voltage is 208v instead of 230v, how is it goingto affect the circuit and i am curious knowing that how the required field voltage and field current decide the values of components in the circuit.

5. sir if i have understood it right i need to give 30 v ac using a transformer in place of WNDG

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#133
In reply to #132

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

11/05/2012 6:30 AM

Welcome Roymor,

I have read your reply but must go out shopping now.

There is information about voltages on my circuit diagram. So you can calculate the power in resistors by ohms law yourself. If you want some help, I will give it

Regards,

67model

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#136

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

04/03/2013 9:48 AM

sorry .... if i did not read the thread that lies ahead ..and thanks for the guidance.

there is no question in my heart ..

1. AVR maximum MK5 can plug in how many watt generator?2. eg AVR Mk5 in pairs on the generator 40Kva by increasing the number of FET 1 is in parallel may allow avr Mk5 whether this could work well on the generator 40Kva?

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#137
In reply to #136

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

04/03/2013 2:17 PM

the actual AVR worked well on a 5KW generator. I am not sure that just simple increasing the number of FETs will work....I know when you put a number of transistors in parallel you need to put resistance (0.1/5W) at the emiter side in order to load them all equally,.. so in short-I can't answer for sure, but I am also confident the 67model can :)

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#138
In reply to #136

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

04/04/2013 6:36 AM

Hello Lare (is that own or family name?)

The first step is to identify the field winding of the AC generator and check it is not short circuit or open circuit. Then apply fixed DC excitation from a battery to the field and see if you have an AC output. If the field is damaged, then it is a strip/repair job on the generator.

Unless it is written on the generator rating plate or in the manual, finding the approximate field rated voltage and current is the first step to an AVR specification. Is there a generator make or type available?

Greetings to Mishel, I hope you are well, despite the economic gloom around Greece.

67model

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#139
In reply to #138

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

04/04/2013 9:57 AM

Hello 67model and thank you for the greetings and interest. Yes, we are just fine, the economic crisis is something that we've learned to live with :)))

all the best from Macedonia

Mishel

Register to Reply Off Topic (Score 5)
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#140
In reply to #138

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

04/08/2013 5:44 AM

hello 67model ...
avr I want to wake up I will apply to generators that can still work, just the generator using a diode without using avr,
well this time I want to modif generatot it using avr in other words replace the diode with avr mk5,
specification generator. output 10kVA single phase,
because it is only capable in 5kva mk5 I want to be strong in my modif 10kva or more, by adding the number FET1 in parallel, for example, 4 pieces or more.
whether it was all I could do?

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#141
In reply to #140

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

04/08/2013 8:04 PM

Greetings Lare Dusun,

OK, it is 10 kVA single phase. That will need some change to AVR mk5 for 1 phase.

Questions for vital data...

  1. What is rated output voltage of generator (rms - root mean square - what AC voltmeters usually indicate, even if they work on mean value like AVR)??
  2. What is rated frequency??
  3. What field DC voltage and current give rated AC output voltage (no-load)?? Or field DC resistance and voltage at rated AC output voltage??
  4. What source is used by generator to power DC field?? Separate permanent magnet excited winding in machine or what?? What is the voltage and frequency of this source?? With no field load and at current for rated "Main machine" AC output voltage, see question 1 above??

67model

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#142
In reply to #141

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

12/03/2013 7:43 PM

hello 67model ...
new resume sorry this post
data for the generator I have
GENERAL brands syincoronus generator type ST-10
230VAC output, 43.5 A, freq 50HZ, 1500rpm, phase 1
for information exiter volt ampere've not read anymore because I've erased attachphoto caption generator ..
please help avr schematic for this type of generator

Register to Reply Off Topic (Score 5)
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#143
In reply to #141

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

12/03/2013 8:30 PM

schematic from my friend..good morning .....
from my friend there who gave such a scheme but I do not understand please explanation this scheme .. whether it can be applied in my generator?

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#153
In reply to #138

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

04/30/2014 5:59 PM

67model,

Thank you for mentioning, I am doing fine :) I was away for quite a while but here I am again. It is nice to see that the thread has been interested for other members...

I hope you are well, too !

Mishel

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#144

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

12/04/2013 10:14 AM

good evening sir ..sorry if my thread is more than oneI am still confused with this schemeI do not know the part number of components in the red circlepinout is also available in a green circle will be in connect to where?please enlightenment 67model friends and also friends all ...

Register to Reply
Participant

Join Date: Feb 2015
Posts: 1
#156
In reply to #144

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

02/23/2015 10:49 AM

Hi Laren,

Can you send for big schematic ?

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#145

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

12/05/2013 9:48 PM

still waiting ....Still no new threads about the AVR ...

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#146
In reply to #145

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

12/14/2013 6:23 AM

Dear Lare,

I am going out soon, so do not have much time now.

The N terminal connects to the neutral and L1 to live of the generator output, if single phase.
N goes to star point if 3 phase.

The two green ringed terminals [on right] go to the generator field winding.

RL would be an electromechanical relay with an AC coil rated voltage/Hz equal to normal supply at L1 to N.
Contact rating for same voltage at an amp or two, preferably greater than field current.

Red-ring diode would be same type as that to left side of SCR gate.

SCR (thyristor) would need to be rated to suite L1-N peak voltage with a current rating exceeding the
maximum field current calculated as half the rms L1-N voltage divided by the field resistance -- volts/ohms.

My first impression is that there is a connection & components missing on the left side of diagram.
Also, making connection from anode to gate of SCR via RL contact is not good practice, gate current voltage
ratings are far less than anode - it may have survived where the field current is low, that does not make it good.

Cannot see how zener (3V3?) gets a proper bias current. Also, the purpose of the relay is not clear, probably to get start
if residual voltage of generator too low for SCR pulse driver 2N2646?? to work.
Nor are connections to SCR gate/cathode readable.

67model

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#147
In reply to #145

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

12/20/2013 11:46 AM

Further to my post #146.

This modification has a proper voltage reference, 15 volts TL431 integrated circuit, from neutral N to common supply.

The 2 kohm potentiometer is connected direct across the bridge rectifier and 3V3 zener diode is deleted.

67model

Register to Reply
Participant

Join Date: Feb 2016
Posts: 4
#167
In reply to #147

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

06/02/2017 6:11 AM

Hi, you could send me a bigger picture where you can read the list of the contributors

thank you

Register to Reply Off Topic (Score 5)
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#168
In reply to #167

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

06/03/2017 10:28 AM

If you mean being able to read component [contributers?] kilohm/microfarad values, then the original circuit I modified is in post #144 by Lare Dusun as .jpg and more readable with care. I suggest right click mouse on image, select "save to file", then view that file with Windows Picture & Fax Viewer e.g. by double left mouse click on file in Explorer file list. Using + option to make image bigger does help.

Also see comments in post #146

67model

Register to Reply Off Topic (Score 5)
Participant

Join Date: Feb 2016
Posts: 4
#169
In reply to #168

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

06/03/2017 4:09 PM

Hi I thank you for the answer I tried to download the image but in low resolution I can not read anything

thank you

Register to Reply Off Topic (Score 5)
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#170
In reply to #169

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

06/06/2017 3:38 PM

Reading from left in diagram from Lare.............

  1. (Primary Volts) 240?
  2. (Secondary V) 12V
  3. (bridge) 1 amp
  4. 4k7
  5. 10μ 35V
  6. 2kohm pot; 1k5
  7. 15k
  8. 1k3 ; 10μ 35V
  9. 22k ; 3k3
  10. 4k3 ; 100n
  11. 1N4007
  12. 220R ; 560R
  13. 1N4007 ;1N4007
  14. 22k 5 watt
  15. Last two diodes & SCR to suit voltage & current of field/supply e.g. 10 amp 600V.

key : 4k7 = 4.7 kilohms : 10μ = 10 microfarads [electrolytic]: 100n = 0.1 microfarad [must be plastic polyester or similar]: 220R = 220 ohms.

UJT = 2N2647 ; transistors any npn/pnp small signal silicon planar Vcbo 30V 200 mW

Register to Reply Off Topic (Score 5)
Participant

Join Date: Feb 2016
Posts: 4
#171
In reply to #170

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

06/06/2017 4:05 PM

thank you so much

very nice

Register to Reply Off Topic (Score 5)
Associate

Join Date: Feb 2014
Posts: 28
Good Answers: 1
#148

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

02/12/2014 12:42 PM

Hi, my name is Sude. I came across this thread whilst looking for an AVR circuit for a 100KVA three phase brushless alternator. The alternator is a Chinese in origin and the AVR stopped functioning a few months ago. I was wondering if the AVR described in this thread could be adopted to my requirements. Any help would be greatly appreciated.

Many thanks/sude

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#149
In reply to #148

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

02/13/2014 1:22 PM

I am willing to help. But this thread has got so big [>140 posts] that it is unmanageable.

I suggest you do a new question with a different title.

If you can run your genset off-load, the first step is to understand the field power requirement DC voltage and current to get rated voltage off-load. This begins with measuring the field resistance (ohms) and finding the rated field voltage, this might be in technical information manual or on rating plate of generator.

67model

Register to Reply
Associate

Join Date: Feb 2014
Posts: 28
Good Answers: 1
#150
In reply to #149

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

02/16/2014 8:28 AM

Hi 67model,

Many thanks for your kind offer to help, much appreciated.

As suggested by you I have been trying start a new thread but do not have much success. I wrote a fairly comprehensive note but could not submit. Each time I tried to submit I got a message saying can not process. Obviously I am doing some thing wrong. Will try again and see what happens.

I have called the new thread "AVR for 100 KVA Alternator"

Sude

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#151
In reply to #150

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

02/16/2014 5:06 PM

Months ago, there were changes to CR4. These seem to block a "post" with an "unable to process" message when logged-on longer than a certain time.

Mostly [like now] I have to copy all my message (by highlighting all of it, using shift and arrow buttons on the keyboard: then hold Ctrl key and press C key), log out, log back in, open a "Reply" window again, then do Ctrl-V keys to copy it all into the window. Then "submit" works OK.

67model

Register to Reply
Associate

Join Date: Feb 2014
Posts: 28
Good Answers: 1
#152
In reply to #151

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

02/16/2014 9:17 PM

Hi 67model,

As you say, I probably had a time out when I wrote all the explanations to my request. I was able to successfully start the new thread "AVR for 100 KVA Alternator".

I have all of the posts on Mishels' AVR in hard copy format, and have built the Mk5, without current limit and LM317, to understand how the system works. I have incorporated a variac feeding a 230/12 volt transformer to simulate the generator side and used a 12 volt battery for the exciter voltage with a 5W auto lamp in circuit. When the variac is at minimum the lamp glows with full brightness and as the variac is turned towards 230 volts the lamp progressively diminishes. I can now understand how the 'AVR' functions. What is now required is to couple the "system" to my application.

Please be kind enough to look at the thread I started. I am confident you will understand my problem better in that.

Many thanks for your kind assistance.

Sude

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#154

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

12/16/2014 9:36 PM

Good afternoon 67 models
sorry old does not make this forum ..and thank you for all the knowledge that is given in this forumwho once said by you for avr mk5 3 phase when mounted on the generator singlephase nothing should be changed. what should be changed? may be depicted in the schematic? plan will be applied on syinchoronus 3kVA single phase generatorguidance please ...

Register to Reply
Participant

Join Date: Mar 2015
Posts: 1
#157

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

03/15/2015 9:45 PM

Old thread I know, but today I managed to dissect a Kipor6700T Avr. Its almost identical to the generator talked about earlier in this thread, except for the fact its a single phase version.

I have taken the last couple of hours to draw the schematic of it. I have all the values except for one cap. It may be a DC blocking cap, but it forms an RC network so not sure, and a diode that's 1Nxx07, possibly a 1N4007 or more likely a 1N5407? Same size as a 1N5408 however... I'm guessing its purpose is to protect the FET from inductive loading, as highlighted as a problem earlier in the thread.

Essentially it works like this.

The generator in addition to the main AC 120vpairs has two additional windings and a field winding. The two pairs of windings always seem to be blue OR green and yellow.

Both pairs of windings blue and yellow go through a full wave rectification. The blue pair are then used as v. ref I think. The Yellow pair are then fed via a large 450v 680uF cap and via a FET to the output field winding.

Now the fun part, here is the schematic:

I don't guarantee this is 100%. Its a bit of PITA trying to do this, especially since it came from a potted AVR board. Last time I tried this everything was smashed to pieces - this time I was more patient and got lucky.

Anyway this may be of use to people out there like me! Please let me know if this is of any help! I wondered for ages what was inside this black box!

Register to Reply
Commentator

Join Date: Apr 2011
Posts: 99
#160
In reply to #157

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

08/13/2015 11:17 AM

Definitely will keep this schematic and would appreciate 67 model comment :)

Register to Reply
Participant

Join Date: Aug 2015
Posts: 3
#161
In reply to #160

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

08/13/2015 3:12 PM

Demagnitized maybe.. Been testing the electric drill method now, but it doesn't pick up. My 2 wires to the AVR (sample winding voltage?) give 2,5 vAC which should be 140 V (if I read the manual right) What voltage do you get? Maybe the demagnitisment causes it, I will try later with 12 v battery connected to rotor.

Here is a link to all the documentation and photo's:

http://ppl.ug/OKk8mGtfYEs/

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

Join Date: May 2007
Location: NYC metropolitan area.
Posts: 3230
Good Answers: 444
#162
In reply to #157

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

08/13/2015 10:27 PM

Well done! The bottom half of your diagram is the AVR including the comparator, feedback, and reference/setpoint stages; the top half is the power stage which controls the output voltage/current to the field by following the output of the AVR.

Why two windings? To prevent interaction between the two stages. If only one winding powered both the field and the AVR then as the field current changed the voltage drop across that winding would change as well, and the AVR wouldn't be able to distinguish whether a change in output voltage was due to the loading of the stator or the field.

By keeping them separate one winding is sensing the combined effect of the rotor MMF and the stator (hence load) MMF. This in turn provides the comparator with a proxy of the output voltage, as well as powering the AVR circuitry, while minimizing the interaction between the output stage and the AVR.

The RC circuitry is there to provide a relatively ripple free field current and dampen out any oscillations from the minor perturbations and/or rapid load shifts. The diode across the field is there to absorb inductive kick from rapid changes in the field current and prevent damage to the circuitry.

__________________
“Tell me and I forget. Teach me and I remember. Involve me and I learn.” Ben Franklin.
Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#163
In reply to #157

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

08/15/2015 5:26 PM

Well done, beyondhelp!

I have re-drawn your diagram with component references - to make it easier to refer to each component. Please check if I got it right. Sorry the image contrast is not as good as I wished.

I have some questions & comments below.

  1. Is R7 value 110 or 910 ohms?
  2. The transistor should be NPN ? Arrow as you drew it would be PNP.
  3. Is FET1 a type IRFS640A - difficult to read?
  4. Have you tried to measure the value of capacitor C2?
  5. What are the measured AC outputs of the two windings and DC outputs of their bridge rectifiers?

As I see it, TR1 and FET1 make a non-inverting amplifier with very high gain, R2 and C2 give positive feedback. FET1 could be switched hard on and off by an oscillation, modulated by ZD2 feedback to regulate field current.

67model

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#164
In reply to #163

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

08/17/2015 2:23 PM

Trying circuit image of post #163 another way, to see if CR4 treats it better - original is ~2600 x 1100 pixels.

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#166
In reply to #164

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

02/05/2017 8:33 PM

please link this image pack, for high resolution, and a list of components of thanks :)

Register to Reply
Participant

Join Date: Aug 2015
Posts: 3
#158

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

08/12/2015 4:34 PM

Hello,

I've got exactly the same generator as you have. It is a KIPOR KDE6700T3

This is the AVR you need: http://www.wiltec.de/spare-part-generator-avr-wjs50z.html

Mine was working fine untill a big soundsystem did something. I was thinking AVR is broken, because when putting 12 vDC on the rotor it gives 150 vAC on the dail.

So I ordered a standard 5kw 3phase AVR from China: didn't work. Now I ordered the official AVR: doesn't work. Tonight we've been discussing and measuring a lot and now I think it is the sampling winding, because the AVR doesn't do anything.

Tomorrow I will measure the voltage of the sampling winding to see if it gives 16 vAC.

I have got a lot of service manuals with data for you. Do you want it?

And can you help me with this: there is no wire coming from the control panel which turns on the AVR. My guess it must be getting power (to start) from the sampling winding? Am I right? But how is this possible, is there a permanent magnet in the rotor?

Maybe this is double posted, I cannot find my first post, so typed it again.

Best regards from the Netherlands.

Marten

Register to Reply
Participant

Join Date: Aug 2015
Posts: 3
#159

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

08/13/2015 3:35 AM

This might help also:

http://portable.generatorguide.net/avr.html

Register to Reply
Participant

Join Date: Feb 2016
Posts: 4
#165
In reply to #159

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

02/29/2016 12:38 PM

This might help

Register to Reply
Participant

Join Date: Sep 2019
Posts: 2
#175
In reply to #165

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

09/07/2019 11:53 AM

In the 1980s I had to design an AVR for a 25KVA three-phase (220V) generator set. I took advantage of pieces I had available.

It worked wonderfully under any load and never failed more than 30 years of use. As it is a "universal" circuit, it can be easily adapted to other generators with exciter. In my application the maximum voltage supported by the field coil was 108VDC. Tr4 prevents the voltage on the coil from exceeding this limit. Tr1 sets the three phase output voltage to 220VAC. V1 was a central zero-voltmeter that allowed you to see the quick three-phase voltage corrections as a function of load variations. The other components perform conventional functions. I hope I have contributed in some way.

Register to Reply
Participant

Join Date: Sep 2019
Posts: 2
#176
In reply to #175

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

09/07/2019 12:15 PM
Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#178
In reply to #176

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

09/07/2019 1:43 PM

OK, I get the diagram from i.pinimg.com.

I guess the 6A8 & 10A10 are diodes [I am used to cathode bar filled black in symbol or arrow with cross-bar at point].

3A & 5A are fuses [ I am used to symbol made with a simple rectangle with line straight down middle as "fuse wire"].

Text now visible with Fs1, Fs2... makes clear. But I found your symbol in my ancient ARRL handbook.

Puzzled over 0.1 uF (polarised??) until I realised it meant polyester. The other capacitors are electrolytic, which ought to have + & - added to circuit for correct connection polarity.

The power filter inductor [180VDC?] could be a problem to buy (no value/rating given & such items are esoteric nowadays - I can remember when they were standard in vacuum valve radio), I think - since 3 phase rectifiers have very little ripple - this could be replaced by a "short" or low value resistor. For a single phase gen, there would have to be some modification.

Transistor 2SD380 seems obsolete

67model

Register to Reply
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#177
In reply to #175

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

09/07/2019 12:29 PM

It looks an effective design.

Unfortunately, CR4 images seem to condense the file size and detail of pictures, if they are too complex.

If you try to copy & zoom into the picture; you will find detail like component ohm/mF & type & rating details are unreadable.

You would have to go to the trouble of breaking circuit into several pictures for it to be readable.

One puzzle I have, looking at circuit, is that main control is by bipolar transistor & the inductor on right is generator field? These require powerful DC, but I cannot pick out any rectifiers in the picture.

67model

Register to Reply
Active Contributor

Join Date: Mar 2013
Location: BATANG INDONESIA
Posts: 10
#179

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

06/25/2021 10:29 PM

hello teachers and friends, how are you?

I made an Avr generator with this scheme, it can work but the volt sensing response is too slow, so for the use of changing loads on the generator it is still not good. please guide me to modify the schematic so that the voltage detection response becomes faster.

Register to Reply Off Topic (Score 5)
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#180
In reply to #179

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

07/06/2021 6:10 PM

Dear Lare,

Still surviving the Covid Pan-panic, thank you!

The response would be set by the 470 μF/35V capacitor connected to B of the transistor. Reduce μF to speed response.

Regards,

67model

Register to Reply Off Topic (Score 5)
Guru

Join Date: Dec 2010
Posts: 1623
Good Answers: 111
#181
In reply to #179

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

07/06/2021 6:48 PM

I ought to point out that fitting the fuse in the neutral [N, earthed] side of supply to a load is contrary to any supply regulations I know. If the fuse blows, or is simply removed in the belief this makes the AVR isolated & safe, the whole board will be live at 220V [instead of being at earth/ground if fuse were in live U]. If the generator is the only source, then with fuse out there will only be the residual generator voltage caused by residual magnetism of the field [this cannot be considered safe unless it is 24V or less]. But there is a temptation to connect a house mains supply in place of U phase for testing/fault finding.....!!

Register to Reply Off Topic (Score 5)
Register to Reply Page 2 of 2: « First < Prev 1 2 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 (59); A.A.Khi (1); Anonymous Poster (1); baxterm (9); beyondhelp (1); cooling (1); Dotternetta (3); kintipsy (1); Lare Dusun (10); Mishel (72); nesubra (1); oldero (4); Phunny (1); RAMConsult (9); roymor (3); sude (3); Tiredmg (2)

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

Advertisement