Can a Germanium Transistor Be Replaced With a Silicium One?

My second Google hit was rfparts.com and they have them in stock for $9.95. That would seem like the easiest way to go.

Substituting transistors is not that straightforward, even catalog substitutes are not guaranteed to work in all cases, it depends on what is exactly doing in the circuit (switches, amplifies linearly, has to match with something, etc). But with some gain tweaking, a TO-3 like MJ15004 should work in your case, assuming it's used to control exciter current, if everything else in the circuit is verified OK. Type "germanium" should not be critical since the MJ has relatively low saturation voltage even for chopper use and as a newer design and higher power, should be more robust. S.M.

The MJ15004 works fine!! I've done some drilling and soldering at the module and checked the generator today with the altered AVR. It gives steady 220 VAC, no problems at all. Tested only with a grinder and a drill (approximately 1KW each).

I will check it again with a more demanding load during the week and inform about the results. I intend to hook the welder and see if it works.

thank you all for helping me

Mishel

SM,

Good point. Usually the AVR circuitry limits the drive so that there is a point called the "Ceiling Voltage" beyond which the current in the field won't go any higher, but with all the part substitutions that point may have shifted. I suspect those wirewound power resistors may also play a part in that as well, and it's not the power transistor that's at risk, it's the field and stator windings plus all the iron that needs the protection from the excess heating from overexcitation. Poor power factor loads like some welding machines can wreak havoc on small generator sets like this one.

RAM

Beautiful description. Wish I was that good! Good Answer.

Being overrated is bad for the character, but thanks anyway. (LOL) S.M.

I agree with Yusef1's comment Being able to give a savvy and constructive comment like yours is not being overated!

Your comment just demonstrated / showed a person's experience /knowledge on the subject matter!

Yep, I will be careful and will increase the loads with the eye glued to the instruments (it has both Voltmeter and Ampermeter on it). I am calm because this generator has an electronic circuit that "shuts of" the output voltage if there is a ground problem or over-voltage regime. I guess it is designed to protect from surges and eventual shocks if not grounded properly. That's why I want to repair and keep this one !

And, it is not a small at all, I mean , it should have at least 5KW, fair enough for the welder.

regards

Mishel

I did a quick Google search, and it looks like the 2N174A, as others have said, is still available. Lots of times power transistors are in a push-pull circuit with a matching transistor of the other polarity (PNP - NPN) and you wouldn't want to substitute. Best to replace it with the same type.

Thank you all for your inputs,

here are some more details regarding your posts....

- I also found some offers of the original one and I am aware that it is always better to use the original transistor. I may sound cheap but having in mind the price, shipping cost and the possibility that I might need more than a one transistor () I am considering the "swap" option. It should not be critical since it is not Push-Pull or other circuit design that needs good pair (match). The original transistor is used as a regulator of the rotor field (excitement) and it has an adjuster in the circuit so, as far as I understand, the working point of the new (close match) can be altered in some range so it is possible to have automatic feeding of the rotor, more or less. I have never attempted any similar steps so I thought it would be better to consultant brighter brains for electronics (before I f... something again)

-the unit works with the manual regulation of the field excitement so I would say it is not very likely that part of the unit has any problem. I don't like the manual regulation because when I use inductive loads (electric arch welder) the output voltage of the generator drops significantly and the arch is not good. Manual adjustment provide a usable performance for constant loads (pure resistive) but poor for more demanding inductive loads. .

-Yes, I think I found the reason, at least one of them. It is overheated resistance 100 Ohm/10W (open - no reading at all) and I change it (the one with the adjuster). The unit might have some hidden problem that will bring me back to the first paragraph.

-will ask tomorrow for MJ15004 and try it! Several other options have been suggested by some friends in the field (BU208, BUT11) but I've decided to ask experts for advice.

-I guess we can say it is vintage, it's made in the early 70' for the needs of the Swedish Army. The engine is JLO L230, two stroker with mechanical centrifugal speed regulator (good compression and generally well maintained unit) and the generator is 220V AC and 12V DC, brushes collector/automatic voltage regulator (the module with the burned transistor). This is the Genset before the restoration started, it is all better now except the AVR

thank you all again

greetings from the rainy Macedonia

Mishel

No easy swap out from Germanium to Silicon.

The "turn on voltage" for a Germanium is .3volts and silicon is .7 volts(approx).

Best to bite the bullet and get an original spec. replacement.

How about simply replacing the whole AVR with a modern fixable unit......you might even find a cheap 2nd hand one in a unit where the prime mover is already destroyed for some reason, maybe on ebay for example....

Fixing an old pre-silicon unit means that even if it works eventually, the same problem could happen again......or an even worse one! It sounds like something that you should not be investing in.....

Identify the type and size of alternator you have exactly and go look......or sell it and buy new/secondhand.....

It simply requires the silicon be biased correctly. Germaniums have approximately 0.2V from emitter to base, where silicon has approximately 0.7V, so typically you increase the resistor value from the base to the supply. But do you know how it is biased? Do you have a schematic showing the voltage on the collector? Since it is bad now, you can't measure it.

Yes, there is a schematic at the unit but without values. I will see the values of the resistors at the actual module and let you know. I intend to try the MJ15004, I saw the specs and have ordered it and think it can't damage anything...if it does not saturate-no worries-no excitement-no output voltage. This would be my first step, simple swap. The second would be to change the value of the base resistor and see if that opens the transistor and initiate the excitement.

GA

I have been known to forget that too......

Good note

I just drilled two new holes at the module and hooked the MJ15004, it is easy to follow terminals since there is a schematic at the actual board. I was told by local radio technician to add one resistor in serial before the base, approximately between 50 and 80 Ohms...should I? Sounds logical to me to bias the silicon transistor down since is more sensitive than the original one??

thank you for helping me

Normally the main body of the device serves as the collector terminal and and typically they are considered hot, not insulated. Be sure to insert a plastic or Mylar insulation between the device and the aluminum heatsink! I don't see the schematic and therefore can't tell if logical or not? I don't know what you mean about being more sensitive than the other? I guess it will be a learning process for you! Good luck!

Ok, this is the schematic of the module. The value of the R12 (the base resistor at the transistor V3-2N174A) is 3.9KOhms.

As I said before, I was told to add 50-100 Ohm in serial to the factory base resistor if I use MJ15004. I put a small adjustable 100 Ohm and intend to experiment and see how it works with a different values.

I also add a heat sink and of course, I electrically separated the collector from the chassis (Mylar insulator). This is how it looks, ready to try, hope won't fry :)

Any other thought before I put it in a slot and start the engine

regards

Mishel

Considering it's an emitter follower in a darlington arrangement, I would not change anything. It should work as long as all other parts are good. The bias is controlled by V1. I hope you tested V1 and V2.

Yes, I have tested all the diodes and transistors at the board (disconnected) and they are all OK. I am going to bypass the resistor now and test it tomorrow how it works.

Thank you

Mishel

Hi everyone,

I tested the generator during the weekend and I am confused with the results. Now, without any load connected at all, the generator instrument shows 10A of current, the output voltage rises slowly (by itself) up to a point where the automatic circuit for protection shuts it off. And that is it! I turn off the engine, start again - the same story repeats, the voltage rises slowly, (it takes aprox 5 seconds to reach the 250 VAC) and at that point the protection shuts off the excitation. The generator runs nicely with the manual excitation. It runs properly both manual and automatic at 12 VDC. I noticed that the resistor that I had to change (the original was burned) is hot. I did not connect any load or did anything else.

I know it's maybe close to be "off-topic" but any suggestion would be highly appreciated. I pulled out the module with the avr and check the output transistor (MJ15004) and it is ok. I am going to re-check other active components and try to figure up what draws those 10A

thanks for reading

Mishel

A seriously leaking C1, dried-up C3 if I can see well on the scematic, leaking junctions on the other transistors, zener V7 or burned R15 could be the fault. System IS made to have slow response, so don't worry about its timing, only for driving to protection at no load. I assume that with some loading, this does not happen. About the 10A, something does not fit here. Where is it measured? On 220V(!), 12V (charging a starting battery) or on exciter coil? If on exciter, It's propubly already half-burned. S.M.

how can this be "over rated"?? :)

10Amps showed at the main ampermeter at the generator at 220V AC. I will check all the things you mentioned and will inform you.

thank you so much

Mishel

10A on 220V AC is 2.2 KVA(!). Real power or complex, what kind of no-load is that? No way. Your ammeter's (possibly its internal shunt resistor) is dying/dead/zombie. S.M.

well, I don't know the reason but it happened twice in a row. I did not want to go any further before I check the MJ15004. I hope it's just the instrument. I am going to the local shop and see if they have new capacitors on stock.

regards

Mishel

need assistance (again)

Zener V7 is OK, R15 is OK, will recheck all the junctions as you advised. The problem is with the values of C1 (8μF) and C3 (5μF)! I can't find the exact values anywhere.

Can I put 10μF for C1 and 4.7μF for C3? The voltage is 40V for both.

thank you

Mishel

Absolute values are not critical, caps filter the rectified AC to be used as negative feedback when it goes over the zener preset level, also determine the system time response that MUST be slow. Just use higher voltage and temp rating caps. But forget all that untill you troubleshoot and get an explanation of the high "no-load" current you see, else it will be useless. Check for example for rusty grounding points on the gen. Not wize trying to fix possible multiple faults at random order. It will be more difficult, expensive and take longer. S.M.

Will do all these in a week. I shall be away and the genset is at the cottage while the module is here at home. Hopefully will find the cause ,

Thanks a lot!!

Mishel