Oktava MK-012 high pitched tone. Need help.

[stason99]

Hi all. I have two old MK-012s which were modified by Michael Joly quite some time ago. Recently one of them developed this problem. There's random low level whine. Attached is the recording of it, normalized to - 3 dB.

https://1drv.ms/u/s!AtSsRNK67O3dkzr_vo53TiXHd0KV

I can't reproduce the problem. Sometimes it's there, other times it's not. Sometimes my hand touching the metal XLR connectors at the mic makes the noise lowder, other times there's no noise no matter how I touch it.

The other MK-012 never behaves like that. Swapping capsules makes no difference. The problem remains with different cables. Other microphones connected via same cables never behave like that. It's this one microphone, and it happens randomly. Proximity of my phone to the mic makes no difference. I cleaned all connectors with Dioxit. I've also inspected all soldering etc under a microscope and there are absolutely no issues.

I'm using RME Babyface Pro.

My knowledge in circuitry is somewhat limited (but my soldering skills are great, and I have good soldering equipment at my day job). And, of course, Michael Joly quit any microphone related projects long ago...

So I need your help in finding what is causing this. Any idea?

Here are the pictures:
https://1drv.ms/f/s!AtSsRNK67O3dkzuzD6pTE2cWIhaV

 

[stason99]

Anyone, please?

 

[kingkorg]

Try with another fet. Clean the board and re solder all the contacts. I guess Jolly "upgraded" the fet. Might be the source of the problem.

 

[stason99]

Try with another fet. Clean the board and re solder all the contacts. I guess Jolly "upgraded" the fet. Might be the source of the problem.

Before your advice on changing the Fet I was suspecting the electrolyte caps.

What are the brands of these 68mcF 63V electrolytes? Are they good? Can they be particularly dryed out? Can bad electrolyte caps give the symptoms I described ?

 

[Gus]

I don't like all the flux on the PCB

a blue film cap in place of the tant or electro "between" the JFET and PNP?

Silver Mica instead of a poly____ or C0G?

Are the red caps cases melted in spots?

Compare what you trace against a 012 schematic you can find on the web

 

[stason99]

I don't like all the flux on the PCB

I'm not sure if that's leftover from soldering or intentionally coated with some enamel, I couldn't tell for sure.

Are the red caps cases melted in spots?

Yes, but those don't seem to be very deep melts.

I will try to trace it.

 

[stason99]

What FET is that? Is that an original Russian one? It reads on the package: "3BF4". Need to know what to replace it with if that's the culprit.

Pictures:
https://1drv.ms/f/s!AtSsRNK67O3dkzuzD6pTE2cWIhaV

(Can't upload a picture here)

 

[stason99]

Here's what I traced. 2 caps seem to have been replaced (brown and blue), 2 more added (red). It's unclear whether two 10nF yellow caps were replaced as well. Everything else seems intact. PNP is definitely Russian but I can't find out what the FET is.

Any idea on what might be causing the problem and what components should I replace for better ones?

 

[dbelousov]

I guess Jolly "upgraded" the fet. Might be the source of the problem.

He did not

What FET is that? Is that an original Russian one? It reads on the package: "3BF4". Need to know what to replace it with if that's the culprit.

99,9% chances it's the КП303, a Russian/Soviet RF JFET for very high impedance circuits. Ignore the marking, it's special private codes for factories, thus one can't find any information about their meaning. The BF245 is a reasonable and straightforward replacement.

I don't like all the flux on the PCB

All Oktava PCBs what I've seen were coated after soldering. This coating turns into brown mess after resoldering, so it's hard to say what we see here.

a blue film cap in place of the tant or electro "between" the JFET and PNP?

Yep. It's not a bad idea, actually. I use the same kind of TDK Mylar caps there.

Silver Mica instead of a poly____ or C0G?

They were ceramic in that era of production (not c0g though), they are Mylar now. Mica is a bad idea for their replacement, some PP or Styroflex should be there.

Here's what I traced. 2 caps seem to have been replaced (brown and blue), 2 more added (red).

You are right.

It's unclear whether two 10nF yellow caps were replaced as well.

They are original.

PNP is definitely Russian.

Any "Western" PNP like BC5xx will do the job.

Any idea on what might be causing the problem and what components should I replace for better ones?

Replacing the lytics in old equipment is always a good idea.

 

[Gus]

The red film bypass don't make sense to me I would do a meter ohm check across them and compare the readings with the other microphone that works. I don't like to see melted spots in film caps and check them if I do.

The stock 68uf electros are fine in the ones I have or looked at.
The stock JFETS should be fine
I have posted in the past about using film caps like the blue one in the 012

Do a search for  Oktava 012, MK-012 etc at this forum.
Look for the posts with voltage readings

The first thing I would do is take voltage readings at the PS and drain node, source, collector, emitter and voltage divider nodes.

Next I would test the power supply bypass electrolytic cap. 

Do you have or have access to a cap checker and scope?

Another thing thing you need to be mindful of in some SD microphone is the distance from the parts to the metal body tube. You don't want the parts touching the metal body tube.

 

[stason99]

The BF245 is a reasonable and straightforward replacement.

What about Toshiba 2SK170 that everybody is taking about? Which one is better?

I actually like how the mics sound besides the tone issue in this mic, self noise is quite low, almost as low as in my Sennheiser MKH50. I'd like to replace some components in these MK-012s if that will improve the sound or longevity even in theory, but first I need to fix this one, of course. And if replacing a component responsible for the issue, then I obviously need to install a better/same quality one.

All Oktava PCBs what I've seen were coated after soldering. This coating turns into brown mess after resoldering, so it's hard to say what we see here.

Cleaning that residue did not fix the problem for me. I suspected it would not.

They were ceramic in that era of production (not c0g though), they are Mylar now. An SM is a bad idea for their replacement, some PP or Styroflex should be there.

Can you please post a link to best components you recommend replacing in my MK-012s?

In this old thread (last post #33 on page 1) he lists components. Please see if you agree with those:
https://groupdiy.com/index.php?topic=57114.0

But I don't understand why he talks about Panasonic electrolytes but links are for Vishay.

The red film bypass don't make sense to me I would do a meter ohm check across them and compare the readings with the other microphone that works. I don't like to see melted spots in film caps and check them if I do.

I did that. They are the same between two mics.

The first thing I would do is take voltage readings at the PS and drain node, source, collector, emitter and voltage divider nodes.

Next I would test the power supply bypass electrolytic cap. 

Do you have or have access to a cap checker and scope?

Yes, I haven't checked voltages yet. I wasn't sure if that can be done without capsule connected. Can it ???
Otherwise I have to make some clamp jumpers to connect that capsule while mic is disassembled...

Another thing thing you need to be mindful of in some SD microphone is the distance from the parts to the metal body tube. You don't want the parts touching the metal body tube.

The distance is fine, I checked at first inspection. There was a strip of electrical tape over the components but the distance between components and case is sufficient.

Do you have or have access to a cap checker and scope?

No, unfortunately I don't.

 

[stason99]

The 10nF caps in the good mic are blue, but they are yellow in the bad mic (see the picture). The yellow ones seem to be Russian. Are the blue ones Russian as well?

Can this difference be responsible for the issue?

 

[dbelousov]

Axial caps are rare and pricy nowadays, so radial ones, like Panasonic FR, will be fine. Even the factory workers installed radial lytics in these mics sometimes, I suppose when they were out of axials.

There is nothing wrong with the original JFET. 2SK170 has the lowest noise on the market at the price of high capacitance, so I, personally, stick to low capacitance RF parts for this duty. The BF245 family has lower noise and Cis than the original JFET, but the overall improvement won't be significant, if any.

The orange and blue ceramics are RF filters. You can remove them for a test. Like completely. I doubt that they cause the problem, but who knows. The noise sounds like radio interference, though.

Unscrew the PCB from the connector part and clean the pads around the screws, if you haven't done it yet. 

 

[stason99]

Thank you.
Can voltages be checked without a capsule connected?

 

[stason99]

Checked. All voltages are fine according to the schematics and same as on the other mic :-\....

 

[Gus]

Just got the sound file to play

Check and clean all the connections

XLR pins

Look at the solder connections under magnification

It sounds like you might be rectifying RF caused by oxide diode junctions. 
dbelousov posted about the RF before me

 

[stason99]

Just got the sound file to play

Check and clean all the connections

XLR pins

Look at the solder connections under magnification

It sounds like you might be rectifying RF caused by oxide diode junctions. 
dbelousov posted about the RF before me

I checked all connections under a good stereo microscope. I also soldered casing of the XLR connector dead straight to PCB ground trace. I also tried grounding the capsule with a jumper straight to ground trace, thinking that they might be bad contact between the tube and XLR - no difference whatsoever. Seems like it's not grounding or shielding related at all. I'm puzzled...  :-\

 

[stason99]

Also, the tone is 1kHz with harmonics every 1 kHz up (2k, 3k etc). There's no extra noise when the tone is there. What the hell is that?...

 

[Gus]

Can you scope the 13ks and 47uf node
if you don't have access to a scope  try a different 47uf power supply cap.

 

[dbelousov]

Without a scope, the only option is to replace components one by one. As Gus said, the 47uF cap is the first option.

BTW, there is a photo of factory-installed radial lytics.