Oktava MC/MK-012 mod -DIY

Middleman

Middleman

Professional Amateur
Anyone see the upgrade mods to the the MK/MC-012 mics in Recording magazine this month? Also a lot of details about the variances in parts found in this mic.

The leaky capacitor issue is diagrammed with photos, pretty informative.
 
Middleman said:
Anyone see the upgrade mods to the the MK/MC-012 mics in Recording magazine this month? Also a lot of details about the variances in parts found in this mic.

The leaky capacitor issue is diagrammed with photos, pretty informative.
Click to expand...
Yep. I’ve already put the mod on mine. My MK-012 came from GC about 2 months ago and the ‘leaky’ caps you see in the picture had already been changed to decent electrolytics. I only had to replace these parts: the FET with a 2SK170; the 680Ms with two 1Gs in series (to make a 2G resistor); the 680pF (C1) with a 820pF, NPO cap.

I tracked a ‘before’ and ‘after’ take on my acoustic guitar, flat-picked, of course. As far as I can tell, it resulted in less boomy bass (Yea!) and the entire sound became more accurate. At first I thought the original version was a little “warmer”, but after careful listening I’ve concluded that it was just more flabby and exaggerated. Some may even think the mod’ed version sounds ‘sterile’, but in reality it’s just doing a better job of reproducing the source.

I’ll post links to the clips as soon as I get a chance to sit down and do it.
 
I was following this in r.a.p and got a few more details on the parts. The FET is a blue version of the 2SK170 - they come in different "colors" that indicate different specs, presumably, and are not common, likewise the resistors.

I've got a couple of major electronics distributors here that may have this stuff locally. If so I ought to get a fist full of both components and sell what I don't need.
 
bongolation said:
How did you clean and conformally-coat? Any tips?
Click to expand...
When removing the parts, I have better luck with the braided solder wic instead of a vacuum solder puller. If you use a puller, you run the risk of removing traces. To use the wic, place it over the connection and apply the iron, pressing the wic into the joint. The solder should melt and suck up into the wic. The trick to doing this is that you need to wet the wic or the connection with flux before heating it. If you don't have any flux, the next best thing is to melt a little new solder onto the connection before you remove it. Even though you're adding solder to the connection, you're also adding flux which makes it take to the wic easier. BTW, I’d use a small wattage iron, maybe 20-25 watts.

After the parts are removed, clean the area (just where the parts came from) with alcohol to remove the left-over flux. When you put the new parts back on, clean the same area again - the main thing is to clean the area around the FET. I use solder with water soluble rosin, so for my second clean-up, I just used water. When cleaning, it's best to use one of those cheap "horse hair" type brushes with the metal handle (looks like a cheap paint brush about 1/2" wide). I always cut the bristles down to about 1/2"-5/8" in length. It seems to clean better that way. I have not applied conformal coating yet. I may do it later.
 
bongolation said:
...The FET is a blue version of the 2SK170 - they come in different "colors" that indicate different specs, presumably, and are not common, likewise the resistors.
Click to expand...
It isn't actually a 'color', it just has "BL" at the end of the part number to indicate it's "drain" current. BL is 6-12mA. There's also a "GR", 2.5-6.5mA, and a "V",10-20mA. This particular circuit needs the 6-12mA version.
 
Flatpicker said:
It isn't actually a 'color', it just has "BL" at the end of the part number to indicate it's "drain" current. BL is 6-12mA. There's also a "GR", 2.5-6.5mA, and a "V",10-20mA. This particular circuit needs the 6-12mA version.
Click to expand...
Yes, but they're apparently color-coded: B/BL=blue, G/GR=green and V=violet. Per a post from Scott in a FET discussion on r.a.p.
 
Last edited:
Flatpicker said:
I have not applied conformal coating yet. I may do it later.
Click to expand...
Supposedly, this is the most critical and difficult part of the operation, the defluxing/decontamination and conformal coating of the surfaces which are at least momentarily perfectly clean and bone-dry. Without the coating, almost any contamination or condensation will cause short across the Hi-Z circuit. Almost anything is a short compared to a 1G-3G resistor!:)

Apparently, cleaning the existing flux is no picnic, if it came with grunge from Tula. Scott's told me in the past that this was pretty tedious. All the components must be seriously sanitary. This is the one reason I haven't done any upgrading on my Oktavas.
 
Any of you tech geniuses interested in modding my Octavas for a modest fee. I am solder-iron-challenged, so reading about this stuff isn't going to do me any good. I'd be willing to buy all the parts and pay a modest fee to anyone interesed in modding mine. If you want, you can send me a PM.:)
 
Rev E said:
Any of you tech geniuses interested in modding my Octavas for a modest fee. I am solder-iron-challenged, so reading about this stuff isn't going to do me any good. I'd be willing to buy all the parts and pay a modest fee to anyone interesed in modding mine. If you want, you can send me a PM.:)
Click to expand...
I'm laughing at the "tech geniuses" part, ;) but yeah, I'm interested. Feel free to PM me about it.
 
bongolation said:
Supposedly, this is the most critical and difficult part of the operation, the defluxing/decontamination and conformal coating of the surfaces which are at least momentarily perfectly clean and bone-dry. Without the coating, almost any contamination or condensation will cause short across the Hi-Z circuit. Almost anything is a short compared to a 1G-3G resistor!:)

Apparently, cleaning the existing flux is no picnic, if it came with grunge from Tula. Scott's told me in the past that this was pretty tedious. All the components must be seriously sanitary. This is the one reason I haven't done any upgrading on my Oktavas.
Click to expand...
I don't know diddly, but I'd think that the only critical part of the circuit, impedance-wise, would be from the capsule to the FET, which is probably a wire, a resistor and the gate leg of the FET and associated circuit board traces, so that's the only part that needs to be defluxed and clean. After that it's a low impedance circuit.
 
bongolation said:
...Without the coating, almost any contamination or condensation will cause short across the Hi-Z circuit. Almost anything is a short compared to a 1G-3G resistor!:) ...Apparently, cleaning the existing flux is no picnic, if it came with grunge from Tula...
Click to expand...
The coating is helpful in humid environments as far as self noise is concerned. But unless you’re in a damp basement or something it may not be that critical. Here in my room the humidity is 53% and the mic is exhibiting a very low S/N ratio, probably lower than any other mic I have.

As for cleaning the original flux, it really depends on the type they use. It may take something stronger than alcohol to clean it well. In the old days we used Trichlorethane (sp?) to clean boards, but it’s illegal now due to the ozone depletion thingy. Then they got smart and started making flux that is water and alcohol soluble. BTW, if you want to see a nasty, flux-ridden circuit, pop the cover off a MXL 603s. I can’t see, for the life of me, how they get away without cleaning those boards any better then that.

Here’s something I’d like to point out – notice how the high-Z stuff (the FETs gate, the G-ohm resisters, the capacitor and the capsule connection) are all mounted on post at opposite sides of the pcb. In order for flux or moisture to break down this high-Z area, it would have to make continuous contact from post-to-post or from post to another pad/trace which is way more difficult than just making contact across a couple of close traces as with other mic circuits. This is pretty smart on Oktava’s part, as it eliminates a lot of error potential.
 
I've read that simply the oil from your hands on the body of the resistor will cause a relative short across the leads in the case of 1G-3G resistor, and that is nearly impossible to overstate the need for both meticulous cleanliness in the original installation and sealing the surfaces once this is attained.

Apparently, one of my older Oktavas (I have a few) is not adequately sealed from the factory and it demonstrates what Scott described to me as symptoms of humidity/condensation bridging in colder times of the year in the studio. This with just the original 680M resistors.

The real hassle is when the flux and/or contamination has been sealed in place from the factory with conformal coating over it, as happens in some of the older Oktavas. :eek:
 
Last edited:
bongolation said:
I've read that simply the oil from your hands on the body of the resistor will cause a relative short across the leads in the case of 1G-3G resistor, and that is nearly impossible to overstate the need for both meticulous cleanliness in the original installation and sealing the surfaces once this is attained.
Click to expand...
Good point. I have an DVM that measures up to 2G. When I get a minute, I'll open one up and see what it reads. Probably wouldn't hurt to do some cleaning with alcohol afterwards.

The real hassle is when the flux and/or contamination has been sealed in place from the factory with conformal coating over it, as happens in some of the older Oktavas. :eek:
Click to expand...
Exactly, which is why I'm waiting until winter (~20% humid) to coat mine. :D
 
OK, now you guys are scaring the hell out of me. I was gonna do this but your making me paranoid about cleanliness and flux and now human oil.

It is only a $50 mic but is this really somebody wants to jump into? I have built radio kits and modified my fender amp but never done a mic mod. This sounds like critical care is necessary to get the mic to work right. Flatpicker, what's your take on this? Can the average hobbyist do this with out screwing it up?
 
The circuit mods are nothing. They are childishly simple. The prep, cleaning and sealing are about 95% of the work on these Oktava projects.

If you stop to consider that we're talking 3,000,000,000 ohms here, you can see that almost anything is like a copper wire compared to that. That's about as high a resistance as I can imagine this side of a 1" air gap in dry air.
 
Middleman said:
OK, now you guys are scaring the hell out of me...
Click to expand...
You can do it! Really, it's not brain surgery. :p Buy some solder wic, a cheap cleaning brush and take your time. Follow the instructions in the magazine and the ones I wrote above. I guarantee you it'll end up much cleaner then it was to begin with. :D

If you or anyone else does this, feel free to pm me if you have any questions. I'm glad to help. The tricky part is soldering two resistors together so that they'll fit in the place of one. You have to bend them in a "s" shape. You'll see what I mean when you do it.

Two more things:
1) The original FET in mine was labeled "AB M7" and was in backwards form the magazine picture. Put the new FET in just like it's shown in the magazine.

2) Once you've finished and cleaned it, LET IT DRY OVERNIGHT. If you try to dry it off and plug it in right away, even though you 'think' it's dry, it's not and it won't work (been there, done that :p ).
 
One of the things I'm wondering about is to what extent changing the FETs (and other components) might unbalance a match between a closely-matched pair of microphones. As I said in another message about my tests, I lucked out and got a pretty well-balanced random pair and as a consequence am leaning toward the "ain't broke, don't fix it" position.

From what I gather, there is quite a bit of variation in these FETs.

On the other hand, it may be that the capsule is the main variant and the circuitry has a negligible effect, I dunno.:confused:
 
You must log in or register to reply here.
Back
Top