DIY home made P2P Oktava Mk012.......circuit questions & observations

[tchgtr]

Geez Gus,
  Now you have truly inspired me. I have a lot of the same parts lying around (probably for similar reasons), and this would all fit just fine in an old 2001 body...

 

[gary o]

Heres my CF012  CF stands for Cat Food as it is setting in a cat food can like I said I got the 012 with the MK101 head basket so the can hold the electronics for now

Gus or anyone the 620 pf 820pf 1000pf at front end that may make the most audable difference .....whats a larger value gonna do ?  more bass ??

I have another question the 1 gig resistors infact any large value resistors, I noticed a wide price range I was lucky enough to get some for 1 pound each but seem some for silly money ....are the more expensive ones better for our mics ??

Since starting this thread Iv been waiting for parts to arrive from Farnell ...so I did some other experiments ....I made a Studio projects TB1 in another cat food can & a G7 in a chocolate wafers tin .....dont laugh.....oh ok I dont care...


The CF012..
.

The CFTB1



The Choctin G7

[URL=http://entertainment.webshots.com/photo/2645

 

[Gus]

Not value as much as type.

 

[pyjaman]

Hi, I've just built this circuit for two oktava SD cardioid capsules following Scott Dorsey's article in pieces of plumbing pipe. So far it works good, but not fully tested yet. I used a wima MKP 1000pf. I'll try to report back when I listen to it.

Laurent.

 

[tchgtr]

Heres my CF012  CF stands for Cat Food as it is setting in a cat food can like I said I got the 012 with the MK101 head basket so the can hold the electronics for now...
Since starting this thread Iv been waiting for parts to arrive from Farnell ...so I did some other experiments ....I made a Studio projects TB1 in another cat food can & a G7 in a chocolate wafers tin .....dont laugh.....oh ok I dont care...

I'm laughing, but I also think it's great. There's a new thread under which you should post those pics
What Gus said: the material the cap is made from seems to be more important than the value.
Just found some 2n3906...need some breadboard...

 

[gary o]

There's a new thread under which you should post those pics
What thread is that then.....

I now have The BT101, BT stands for biscuit tin.........same circuit but with J305 FET that some people are raving about in this mic......

First tests it sounds ok bass light compared to stock 012  but more detailed top id say....very un scientific tests tho....

Looking at my vocal I recorded on my Sonar DAW the waveform isnt symetrical ??? the lower waves are bigger ?? when Iv messed around with other projects before when I see this I usually find I have the output round the wrong way ?? I cant swap output leads in this case can I ?? cos of phantom power ??

Sadly after all PRR & Gus advice I still dont understand how to set up the the bias points or what they should be or where to take voltage readings from.....I have installed 100K trimmer & 50K trimmers to adjust it but not sure what im looking for and where ...so Im guessing its setup wrong for the J305 but its working

Im wondering why the lack of bass........circuit has same 820PF input cap as the Catfood 1....but 100UF caps instead of 68UF same big poly orange drop 1UF ...so it points to the J305 & or its biasing.....

 

[emrr]

lack of bass is easily dirt and solder flux between components. 

 

[gary o]

Ah that is interesting I knew its wa bad just didnt no what it might do......this thing is on vero board as well and some is point to point am I right to keep component leads short.....

Thanks

 

[tchgtr]

There's a new thread under which you should post those pics
What thread is that then.....

http://www.groupdiy.com/index.php?topic=42181.0

 

[gary o]

Tchgtr I like that thread  I better not gate crash tho.....

Board cleaned wire shortened some unused vero board tracks cut......its sounds full.....similar to stock 012 got capcitors to try and decide on....need to A & B the catfood with this the J305 version....aslo ordered a ADK Vienna to play with......

 

[Gus]

A mod

 

[tchgtr]

Interesting...I just ordered some 2N3819s from Digi-key, and I have a 2001 with the original MK-012 circuit in it that I did P2P. It's bright, but incredibly clean.
Might have to give this a try, especially considering the source...
Thank you.

 

[bruce0]

A mod

Update:  Sorry, I see. My mistake. C10 makes it a Lowpass, I was trying to do a highpass... Got it. Thanks.
So what is the knee on your lowpass?  Is it just C10 and R2 or is there more to it?  I calculated the knee at 162 HZ or 208Hz... so I must be wrong (that's a lot of low highs to roll off!).

Added some pics of my rebuild

bb




Gus: I like your mod is much better than what I did!  Thanks!

All I did was change C1 in hopes that I could use the resistance of the base of the PNP as part of the RC filter, but I guess r2 and r3 are involved, and also i am not sure of the impedance of the base.  

Anyway, your approach is more direct, easier to measure.  Is the RC filter just involving R2 (so the knee is at 163Hz?) or is R3 and r4 involved to the VCC (so the knee is at 218Hz)?

Is there any way to put a filter in the pad screw adapter (make it not a useless pad... make it instead a nice low cut filter)? Can I just put an RC filter in there to the case ground?  Or is the impedance so high that any resistor will add noise?




Moved from the other thread:

I had some Oktava Mc-012's and I did try the recording magazine article mods (Dorsey)  But I had more significant results when I built a circuit board for an MC-012 that let me fit in larger leak resistors and bypass a few caps (still fit's in the tube, but tight... not a cat food can).  I can send the layout if you want.  Tightened up the bass a bit I think, and makes strings sound great  (but I kept those circuit boards in case I wanted to go back).

Recently a friend asked me to build him a low cut filter for his octava, for better handling wind noise in outdoor shoots.  I had an old busted MC-12 I got on ebay, and I tried this by changing the coupling cap between the JFET and the PNP to a lower value (I thought I might make a switch for this) but I am not sure of what the resistance of that circuit is, and the filter isn't doing what I expect.

If anyone has any ideas about that, love to hear it.

Also, I noticed that Oktava now ships a low cut filter as a screw-on like the PAD, and that is exactly what my buddy wanted me to do (turn his pad into a low cut). Does anyone know what is in that low cut?  Is it just a cap to lower the coupling capacitance and thus form a filter?

Thanks

 

[Gus]

The mod I posted is a low pass(high cut) C10 to roll off highs.

I simmed the circuit in LT spice and it biased up like the stock voltages.

I posted the schematic as a starting point

Something to try for a high pass(low cut) would be a resistor in parallel with R7 to lower the input resistance.

I am thinking about what Oktava's 012 highpass might be

 

[Gus]

I just simmed some values. Try a >= .01uf in series with a 50 meg.

Connect one side to the center and the other side to ground.

EDIT  I have not built the circuit with the low pass C10 in the schematic I drew.

The high pass circuit a guess as to what Oktava might be doing The top picture in the PDF .01uf and 50meg.  One can not just use a resistor from center to ground because that would change the capsule voltage.

 

[bruce0]

I just simmed some values. Try a >= .01uf in series with a 50 meg.

Connect one side to the center and the other side to ground.  

Gus:

I went inside the current PAD (-10DB) and all that is in there is a 136pf cap to ground.

I sketched the circuit I think you mean in the attached PDF, do you mean the top choice, or the bottom?

(is it an RC network  capacitor in series with the resistor to ground? or is it like a Zobel network (which I really don't understand very well) with both the capacitor and resistors shunting to ground?)

Sorry I don't understand this stuff.  I do have a copy of spice for the mac however, so if you want you can just shoot me the spice commands and I can work it out myself, I don't really use spice, but there is no better way to learn than when I can do something real.

Thanks for your help, I really appreciate it.

Bruce

 

[Rossi]

A mod

I suppose you mean this to be used with a non-Oktava K67-type capsule. The Oktavas don't need any low pass filtering.

But doesn't C10 present a high load for the FET? The FET output is basically shunted to ground, no?



re: Hi pass filter: I wouldn't be surprised if the Oktava filter module were just a resistor to ground, say 10-25 Meg. Together with the capsule capacitance, it would form a 1st order hi pass.

 

[Gus]

Rossi

Sim the circuit and adjust C10.  Yes for capsule with a high end rise one might want to reduce.

In a post of mine I wrote "One can not just use a resistor from center to ground because that would change the capsule voltage. "

If one wanted the high pass fixed, reduce both of the high value resistors try two 100megs.

 

[Rossi]

Sorry, I don't have time to do simulations right now, might do next week. But it appears to me a cap to ground as seen from the FET's perspective.


Yes, you're right, I forgot about the capsule polarization scheme. Well, it must be simple. If I was more interested, I'd try to figure out, but honestly I rarely use a hi pass on a microphone. What I would like to see, though, is a switchable 30 or 40 Hz hi pass that's barely perceptible but reduces rumble and room noise. That I would find useful.

 

[bruce0]

I just simmed some values. Try a >= .01uf in series with a 50 meg.

Connect one side to the center and the other side to ground.

EDIT  I have not built the circuit with the low pass C10 in the schematic I drew.

The high pass circuit a guess as to what Oktava might be doing The top picture in the PDF .01uf and 50meg.  One can not just use a resistor from center to ground because that would change the capsule voltage.

So I think I get it - If I understand it is like you use a big enough capacitor to appear as a short circuit to all frequencies, then you put a resistor in to work as an RC network High Pass filter with the capsule capacitance of 40PF.  is that right? Is that how it works?

So the knee you chose was 79Hz?

In any case, I tried it with 55Megaohms and a 470PF cap (I mis-read it..thinking it was .010, all wimas look the same) and I couldn't hear it.  And since I really want to use it for wind noise, I changed that and put two little parrallel WIMA .010uf polyprop caps and 22Megaohm (so .020uf and 22Megaohms), which is giving me a 180 Hz knee (down 3 db at 180 and 6DB per octave below that?).  I think the .010uf would have worked.  But 55Meg is too much (that means it is only down 9DB at 36HZ, and it doesn't seem like that is not enough to prevent overloading the head amp when wind hits the capsule).

It is very hard to measure but I can hear it, and it sounds right gentle and good, and really reduces pop clunk bam wind noises to just noise.. no overload.  I may need to lower the frequency of the Knee a little, it cuts the proximity effect bass a lot (I have a good assortment of small high value resistors (5, 10, 18,22,33,66Meg ) so I should be able to pick a good knee, once I figure out (or you confirm) how this filter is working.  

But it works great, cost me 2 bucks in parts, and is a great use for all these Oktava Mk12 pads, which I hardly ever use... I have 4, I have never used more than 1 at a time.  Maybe I will make a couple of different knee frequencies in different pads.  

Anyway thanks for the theory, I think you are right, this must be what they are doing for the new low cut.

It will be very useful in a zeplin.

My Oktava's use 2Gig leak and bias resistors, so that may effect the filter too, although I don't think so.  I screwed it on to an unmodified oktava I have in a drawer (like the worst case example...) and it worked on that too, sounded about the same (although my modded ones are much quieter.)