Marshall Mosfet Lead Output Stage Problem

[Lee_ranaldo]

Hello all! hope u're all doing fine.

I'm trying to bring my amp back to life. I bought this one used and the previous owner sent it to repair.

The "moron" tech did a full output stage change since the original MOSFETS are not available. He also re-winded the Power Transformer Dropping the Voltage from 50 VDC to 35-36 VDC.

I've Installed a MOSFET equivalent of the originals (2SK134 and 2SJ49) With
2SK1058 and 2SJ162, which have a higher voltage rate (about 10-20 volts more)

The thing is i've fired up the amp and it hums. Without Load the Power transformer starts to vibrate and hum too like they were in short circuit.

I've replaced every electrolitic cap in this amp and all the driver transistors are not shorted.

I'll include an schematic of it if someone may guide me.

Thanks in Advance!

[Lee_ranaldo]

i have to add that everything is stock. THe previous tech just added another driver board and attached the 2N3055 transistors to the heatsink.

So the driver board is as shown in the schematic with the only change of the mosfets already described

[Steve Conner]

Hi Lee

There's nothing special about this amp, it's just a solid-state thing and a simple one at that.

Now you say it hums as if it was shorted. Is it actually shorted? Ie, does it make no sound when you play a guitar through it? does it draw excessive current from the wall? do the fuses blow? do the MOSFETs get extremely hot? is there DC on the output? smoke? flames? Etc.

If so, follow the usual troubleshooting procedure for SS amps, with one caveat: MOSFET amps often suffer from parasitic oscillations, and the bigger the MOSFETs the worse they can be. If you changed the MOSFETs for more powerful ones, you may need to increase the values of the gate stopper resistors, or decrease the resistor in the Zobel network. Heck, I looked at the schematic and it doesn't even have a Zobel network, so maybe you should add one. (A 4.7 ohm resistor and 0.1uF capacitor in series, going from the junction of the MOSFET source resistors and choke, to ground.)

Parasitics usually just make the output transistors explode within a few seconds of applying power, but maybe if it's running off a lower voltage, you're lucky.



Or did the guy who rewound the PT simply screw it up? If he put the core back together poorly, it'll buzz loudly all the time, and may leak flux into the rest of the circuitry, making it hum through the speaker too.

I'd try disconnecting the PT from the rest of the circuit (except the AC input, of course!) and see if it still hums. If so you're looking for a new PT.

[MarkusBass]

I'm trying to bring my amp back to life. I bought this one used and the previous owner sent it to repair.

I've build similar amp for a bass guitar (with two pairs of output transistors - 200W) and there were no problems.

The "moron" tech did a full output stage change since the original MOSFETS are not available. He also re-winded the Power Transformer Dropping the Voltage from 50 VDC to 35-36 VDC.

It's true that original MOSFETs are not available but you could use BUZ901/BUZ906 or similar without any changes. The rewiding of the power transformer was not needed at all - the guy was a "moron".

I've Installed a MOSFET equivalent of the originals (2SK134 and 2SJ49) With
2SK1058 and 2SJ162, which have a higher voltage rate (about 10-20 volts more)

They are similar but not equivalents. They are in a different package and you have to make sure to connect drain and source correctly. Otherwise you may get results sounding like shorted power amp. Isn't it the case that you have? The transistors also need higher voltage on gates. You may achieve this by increasing R14. I don't remember the value but it was something like 1K instead of 100 Ohms. I can check it if you want.

The thing is i've fired up the amp and it hums. Without Load the Power transformer starts to vibrate and hum too like they were in short circuit.

I've replaced every electrolitic cap in this amp and all the driver transistors are not shorted.

This was not needed. The amp with 1058/162 is prone to oscillations. To avoid they you need to add a small capacitor between gates of the transistors and (as far as I can remember) the power supply rails. I can also check it if you want.
So you need some corrections in the power amp (but not electrolitic caps) and an old transformer.

PS: I didn't get the story with 2N3055. This is completely not needed.
I could fix the amp for you but I guess you are not in Europe.

Mark

[J M Fahey]

Hi Lee Ranaldo.
There's quite a bit of contradiction on what you post here.

i have to add that everything is stock. THe previous tech just added another driver board and attached the 2N3055 transistors to the heatsink.

So the driver board is as shown in the schematic with the only change of the mosfets already described

1) Stock? Starting with a different voltage power transformer I guess not so.
2) That added "driver board" certainly is *different* from the original one.
3) 2N3055 are not MosFets
4) You post the *original* schematic which from what you say is not connected or functional any more.
5) My guess is: that amp died, replacements were difficult to impossible to get, to avoid sending it straight to the thrash bin somebody did a non-moron decision to replace the power amp board with something available, and had to re-wind the power transformer to a tamer voltage, +/- 35V being an approppriate value for 2N3055.
I think that new board has died now.
I also think your power fuse does not blow at once, allowing all that buzzing humming stress signs to continue because it surely is grossly overrated.
So please take some pictures of that added driver board, try to identify the transistors there (post their labels) and I'm sure you'll make it work.
The other option is to try to get it back to original spec, but you will have to re-rewind the transformer; I'd start by repairing it as-is.
Good luck.

[Lee_ranaldo]

Now you say it hums as if it was shorted. Is it actually shorted? Ie, does it make no sound when you play a guitar through it? does it draw excessive current from the wall? do the fuses blow? do the MOSFETs get extremely hot? is there DC on the output? smoke? flames? Etc.

Actually it fired up, sounded like either it was draining too much current from the PT like a shorted amp, or an transistor AMP without voltage on the base. (it happened to me on a hartke system amp i was fixing and thanks to this forum i've done it). No overheating, No Flames or any transistor in the driver board got hot.

Without the Mosfets Soldered i got a DC voltage output close to Zero (0,295 Volts DC)

If so, follow the usual troubleshooting procedure for SS amps, with one caveat: MOSFET amps often suffer from parasitic oscillations, and the bigger the MOSFETs the worse they can be. If you changed the MOSFETs for more powerful ones, you may need to increase the values of the gate stopper resistors, or decrease the resistor in the Zobel network. Heck, I looked at the schematic and it doesn't even have a Zobel network, so maybe you should add one. (A 4.7 ohm resistor and 0.1uF capacitor in series, going from the junction of the MOSFET source resistors and choke, to ground.)

Parasitics usually just make the output transistors explode within a few seconds of applying power, but maybe if it's running off a lower voltage, you're lucky.

I'm gonna Simulate that on the computer since i'm not familiar with MOSFET power amp designs to be honest any info that u may share would be appreciated

Or did the guy who rewound the PT simply screw it up? If he put the core back together poorly, it'll buzz loudly all the time, and may leak flux into the rest of the circuitry, making it hum through the speaker too.

I'd try disconnecting the PT from the rest of the circuit (except the AC input, of course!) and see if it still hums. If so you're looking for a new PT.

The transformer by itself doesn't hum or got problems but with the previous power amp stage there was an annoying hum in the background and was the reason why i decided to restore things back to original.

I'm gonna post a modified schem with the voltage readings i'm getting.
Thanks so much for your info and reply , i'm gonna do my google time and read about what you've suggested.

[Lee_ranaldo]

I've build similar amp for a bass guitar (with two pairs of output transistors - 200W) and there were no problems.
It's true that original MOSFETs are not available but you could use BUZ901/BUZ906 or similar without any changes. The rewiding of the power transformer was not needed at all - the guy was a "moron".
They are similar but not equivalents. They are in a different package and you have to make sure to connect drain and source correctly. Otherwise you may get results sounding like shorted power amp. Isn't it the case that you have? The transistors also need higher voltage on gates. You may achieve this by increasing R14. I don't remember the value but it was something like 1K instead of 100 Ohms. I can check it if you want.
This was not needed. The amp with 1058/162 is prone to oscillations. To avoid they you need to add a small capacitor between gates of the transistors and (as far as I can remember) the power supply rails. I can also check it if you want.
So you need some corrections in the power amp (but not electrolitic caps) and an old transformer.

PS: I didn't get the story with 2N3055. This is completely not needed.
I could fix the amp for you but I guess you are not in Europe.

Mark

Hey thanks mark for your reply.

Yes the case of these new mosfets are completely different. I've installed them correctly with the proper insulation and thermal compound. I've followed the datasheet to connect them properly. In the original design the original mosfets are attached to the heatsink connected by wires coming from the PCB , so it's easy to install another type of transistors since there is plenty of room inside the chassis.

I know the problems of parasitic oscilations in tube amps and the results that i may get if i don't fix it quickly so any help is apreciated since i'm not familiar with mosfet designs.

my goal is to make it work with the current PT and then send this one to be re-winded to the original specs.

[Lee_ranaldo]

Hi Lee Ranaldo.
There's quite a bit of contradiction on what you post here.

1) Stock? Starting with a different voltage power transformer I guess not so.
2) That added "driver board" certainly is *different* from the original one.
3) 2N3055 are not MosFets
4) You post the *original* schematic which from what you say is not connected or functional any more.
5) My guess is: that amp died, replacements were difficult to impossible to get, to avoid sending it straight to the thrash bin somebody did a non-moron decision to replace the power amp board with something available, and had to re-wind the power transformer to a tamer voltage, +/- 35V being an approppriate value for 2N3055.
I think that new board has died now.
I also think your power fuse does not blow at once, allowing all that buzzing humming stress signs to continue because it surely is grossly overrated.
So please take some pictures of that added driver board, try to identify the transistors there (post their labels) and I'm sure you'll make it work.
The other option is to try to get it back to original spec, but you will have to re-rewind the transformer; I'd start by repairing it as-is.
Good luck.

The "driver board" is completely stock. no mods in it. that's what i meant.
the 2N3055 transistors are not mosfet of course and they were installed into a "new" output stage with an added extra driver board that i've removed from the chassis.

The Added PCB was made soo porly i wouldn't trust the quality of it and removed it completely.

The "stock" board is working, i've tested every component and they're fine.

[J M Fahey]

extra driver board that i've removed from the chassis.

Ok, that simplifies it, I didn't get that detail.
Hope you can get the power board back to original, then.
I guess your only two problems will be:
1) parasitic oscillations, which can me tamed/killed by, say 1K series resistors, soldered straight on the power Fet Gate legs themselves and
2) biasing, which can be adjusted by a gate to gate trimmer, maybe 1k, which you will adjust while measuring voltage drop across emitter resistors.
3) Using the lower voltage power transformer as an interim solution is not *that* bad, you will still have ample volume if you use decent speakers and load it with 4 ohms.
You'll get close to 80W .
Good luck.

[MarkusBass]

Hey thanks mark for your reply.

Yes the case of these new mosfets are completely different. I've installed them correctly with the proper insulation and thermal compound. I've followed the datasheet to connect them properly. In the original design the original mosfets are attached to the heatsink connected by wires coming from the PCB , so it's easy to install another type of transistors since there is plenty of room inside the chassis.

I know the problems of parasitic oscilations in tube amps and the results that i may get if i don't fix it quickly so any help is apreciated since i'm not familiar with mosfet designs.

my goal is to make it work with the current PT and then send this one to be re-winded to the original specs.

Hmm, so it looks like there are no problems. You have MOSFETs soldered correctly, power transformer with slightly lower voltages. The only things that you need to do is to check oscillations (and if they are present add capacitors), and bias the output transistors. As J M Fahey writes, you can do it with a trimmer in place of R14. I will check the value (I have such an amp at home - modified in the same way). It was somewhere between 200 Ohms and 1.1k.
If you have any equpment like e.g. oscilloscope, you could check whether you get correct signal on the output - without MOSFETs - also without a speaker. Then add MOSFETs and bias them correctly. This is very simple.

Mark

[Lee_ranaldo]

thanks guys for ur knowledge. !

I don't have an osciloscope but i thnk i could use my soundcard with a proper signal pad to check it.

I'm gonna re-wire those power mosfets again (i always double check before turning on the amp) and install a trimmer in R14.

How i should set this trimmer correctly?, without an oscilloscope?. is like a few amps (Crate, Ampeg) that tells u to measure the voltage in the emiter resistors (0,33 ohms in this case) for a certain value?

Another question.

I'm getting around +/- 34 volts on the gates of both mosfets
is that normal?

I'll post the schem with the reading values i'm getting without the mosfets installed yet. Don't wanna fry those since they were a bit expensive.(12 US dollars each)

[MarkusBass]

thanks guys for ur knowledge. !

I don't have an osciloscope but i thnk i could use my soundcard with a proper signal pad to check it.

The "proper signal pad" term is very important here. Otherwise, you may need help with your soundcard . You know that soundcards are very sensitive to high voltage on the input?

I'm gonna re-wire those power mosfets again (i always double check before turning on the amp) and install a trimmer in R14.
How i should set this trimmer correctly?, without an oscilloscope?. is like a few amps (Crate, Ampeg) that tells u to measure the voltage in the emiter resistors (0,33 ohms in this case) for a certain value?

This is true for every amp. Do you think that Marshall or Eden are an exception? The rule applies to all amps. You set the trimer in the way that you get reasonable current through the resistor (10-20mA).

Another question.

I'm getting around +/- 34 volts on the gates of both mosfets
is that normal?

No. It should be just several volts (2-3 Volts). I think that you should tripple-check the components on the board and the wiring. How many transistors have you got there? 4, 5? Measure the voltages and especially check the current source (D3, D4 diodes and TR6 transistor).

Don't wanna fry those since they were a bit expensive.(12 US dollars each)

I usually buy those transistors for 4$ - check another shop.

Mark

[J M Fahey]

I'm getting around +/- 34 volts on the gates of both mosfets
is that normal?

There's NO WAY you can get +/- 34V there.
1) Gates would puncture with anything higher than 20V
2) Output MosFets would turn-on so hard (if they were still alive) as to place a dead short across power rails.
3) R14 would blow like a dynamite stick since it would dissipate 46W, being probably a 1/4W resistor.

[MarkusBass]

There's NO WAY you can get +/- 34V there.

As I wrote before; the voltage should be 2-3 Volts (can be calculated for the current source current and the resistor value). Either you are not measuring the voltages correctly, or the R14 resistor is open, or driver transistors are shorted. Keep working on the amp .

Mark

[Lee_ranaldo]

hey thanks guys

i was thinking the same gonna go all techie on the driver board and post the voltage readings.

the funny thing is the transistors are not shorted but gonna replace them with new ones. it seems that the faulty driver board caused the original mosfets to blow and the tech didn't bother to repair it and installed a "poor custom made" output stage.

thank you guys for ur suggestions and i'll post the updated info here ASAP