EMT 262 Amp noisy after recap

[salomonander]

Hey there,
im a bit lost…. Maybe someone has an idea. I did recap a pair of emt 262 amplifiers. Supposedly they were a bit noisy. Changed all tantalum caps and the one electrolytic thats in there. I then received a call, that the unit is now way more noisy than before. Both channels behave the same. At first i assumed that maybe i put in a wrong value in one position and therefore shifted a lowpass filter. But i now have both units back and after triple checking, i can not find an error. I did everything correctly.

The units are constructed in an unusual way. One has to desolder two transistors in order to even open it up. I had to extend the transistors legs afterwards to reassemble - so maybe T105 and T106 did get some mechanical fault - or the big resistor R161 next to them.Could any of these three parts be responsible for increased noise? I doubt it but better ask.…

If anyone has an idea - any input is appreciate. Im lost at this point. Maybe its an user error… i wasnt there while testing.

Service manual

if anyone has a different schematic, please let me know. Maybe some revision is screwing with me

 

[user 130681]

If you replaced component for component with the same type and value, it won't be that.
As both channels are the same, it could be an input issue.

 

[Bo Deadly]

It's generally not possible to deduce noise problems without doing some kind of empirical analysis.

Whenever I work on something I always test it with a USB audio interface and software for capturing a spectrum including test tones and noise floor. This should be standard procedure for anyone doing work for hire for sure. Then if the job comes back to you, you can show them the test results.

It is very easy for the customer to conclude that something is wrong without any actual evidence. It could easily be that the noise is simply the result of using the gear in a different way, a different microphone, a different room, a different mixer, it's not grounded as well, the metal cover was off, levels are different, .... etc.

At this point, the only thing you can do is capture spectrums and at least be able to quantify the noise. Spring reverbs are not know for being quiet. But maybe there's something off and you'll see it in the spectrum. Or do you at least have a scope?

 

[pucho812]

Whenever there is talk of noise, one has to describe the noise. Noise is vague but saying things like It has a hum or it has a hiss is more specific and can help point you in the right direction as to how to proceed as far as a repair. Sometimes the noises described are result of other sources and not the unit itself. For example many times I have had to talk a customer down from my unit is all distorted and the reality is they were going from their preamp into the mic pre input on their universal audio interface. Their thought is xlr - xlr not knowing the unit has line level on a TRS connection.
What compounds the issue here is that you were the last to tech it, so no matter what it was doing before to the customer it's not right, what did you do? It's easy enough to get into this trap. The only way to avoid it is not do any work. But it does happen.
I found a schematic on the sowter website but I am sure it's the same from your service manual.
https://www.sowter.co.uk/schematics/EMT 262 reverb.jpg
Good luck...

 

[salomonander]

Thanks guys. I really start to think its a user error. There is no such thing as a noisy batch of tantalum capacitors right?
Guess ill have to build a test setup.

 

[salomonander]

Ps: i told the customer to simply short the in- and output xlrs. Assuming that there should be silence if the error was in the unit.
And it was silent supposedly.…

 

[JohnRoberts]

Thanks guys. I really start to think its a user error. There is no such thing as a noisy batch of tantalum capacitors right?
Guess ill have to build a test setup.

Nothing is impossible but the fact that both are behaving the same is a good sign...

Ps: i told the customer to simply short the in- and output xlrs. Assuming that there should be silence if the error was in the unit.
And it was silent supposedly.…

Customer may have altered his gain structure from before and after listening...

JR

 

[salomonander]

im starting to fear that the amps were somehow modded and therefore the schematic doesnt help. Or that there were different revisions. Proof for that is C221. In the schematic its 1uf. In the units its 22uf. I did change it to the 1uf in the schematic and the amps went nuts, only playing back the transients without any sustain. So there is a difference between these schematics and the actual units. At least for that capacitor.

Is anyone engineer enough to tell me what capacitor is responsible for the lowpass filter? Or if there is any position where a wrong value capacitor would lift the high end or cause noise in any way? That would really help.

 

[Bo Deadly]

That circuit is potentially sensitive in a number of ways. The bias trims have to be right or it looks like something could go "nuts".

There are some high value parts that might create problems with leakage current if the board is dirty (e.g. the 10M R250 and the 2.2M R251). You could try cleaning the board with 99% isopropyl alcohol. Although cleaning this kind of PCB requires great care. You could easily just spread the residue around. There are certain areas that will need to be effectively polished clean.

Check to see that the voltages match up with the schem (in the rectangular boxes).

 

[salomonander]

Thanks Bo,
i appreciate the input. That‘s a great lead. Although i doubt that i managed to touch these pcb parts with say greasy fingers - and did so perfectly identical on both units. Still, ill give that part of the pcb a good cleaning. Oh and I did not touch any trim pots either.

 

[salomonander]

One more idea. After spending more time with it last night i actually learned how to open this up properly. And i realized that during reassembling i might not have had r161 properly attached to its heatsink (chassis). Say it got too hot…. could that account for extra noise?

 

[user 37518]

One more idea. After spending more time with it last night i actually learned how to open this up properly. And i realized that during reassembling i might not have had r161 properly attached to its heatsink (chassis). Say it got too hot…. could that account for extra noise?

Resistor noise is dependent on the square root of the temperature in Kelvin, so yes, if you heat up a resistor it will produce more noise. Whether that is the source of your problem it is hard to tell. Is the resistor in the signal path? is it being amplified a lot by the following stage? those are questions you should be looking for. I can't seem to find resistor 161 in your schematic. What is the resistor for? sounds to me that it is just a power related resistor, or is it in the audio path?

 

[salomonander]

Thanks user 37518,
looks to me like its not in the signal path but part of the +-10V supply. BTW i did receive the correct schematic now. But it only differs for that one capacitor (1uf/22uf) So all caps i desoldered were replaced correctly….

 

[jokeramik]

Hello,

did you adjusted the bias with the new capacitors?
Please note that tantalum capacitors have different leakage currents and frequency response. The ESR ist totally different to new low-ESR-Al-electrolytics. Sometimes, under these circumstances, the operational amplifiers can oscillate. Have you checked the signals with an oscilloscope? Noise can also be a high-frequency oscillation.

Best regards!
jokeramik

 

[salomonander]

thanks jokeramik,
yes thats correct. I did just receive an email from the very kind EMT tech and the entire unit needs recalibration after recap. That should solve the puzzle.