Wednesday 4 January 2012

Making a Super-Modded Oktava MK-012 microphone!


Introducing: The Oktava MK-012

OK, so I've had one of these lovely Russian Oktava MK-012 phantom powered capsule condenser microphones for a few years now, and they really are very very good, and for the money, unbeatable. In film and TV audio, they are a real go-to mic for interiors for anybody that cares but doesn't own a Schoeps CMC641 (£1500!). 
Being Russian and originally made under the "employment irrespective of demand" scheme, they are not, however, perfect, and can be improved by substituting a few parts. Other than some of the parts in the pre-amp being less than superlative, the mics have a lot of handling noise and need a good suspension and a good wind foil. I recommend the Rycote Baby Ball Gag and Rycote INV-7 suspension mount though be warned that it's all a tight fit to mount the baby ball gag and still fit on a suspension system. Without a good wind suppressor and suspension mount though, forget it! Mods like this will do nothing for you.

The Standard Mod:

Scott Dorsey detailed a modification list and instructions for a few parts substitutes that make the mic considerably better.

Here's his schematic, and then another, almost identical one from t'internet, with the mods in place. Scott Dorsey's mods have now become generally accepted as standard practice for serious Oktava enthusiasts:




I did these mods to my mic, based on this kit of parts that I bought from the USA about 5 years ago. 


Here's a guide to replacing those parts and doing the standard Dorsey mods.



One Step Beyond...

My mic is pretty great, and over the last few years I've not only used this mic for a few video and audio projects, but also learnt a lot about electronics by doing quite a few audio projects, and so last week, whilst looking at the schematic, I realised that I had in my parts drawer most of the parts to build a new pre-amp for this mic - basically all of the parts in the body, between the capsule and the XLR output, excepting a few cheap resistor values and a dirt cheap transistor.

The more I looked, the more I saw, and the more I saw the more I looked, and then I realised what I really wanted to do - to make a super-modded Oktava MK012, with a hard-wired hypercardioid capsule. My parts draw wasn't going to be good enough. The parts in this pre-amp are dirt cheap, and so I was only going to use the best, most appropriate parts I could find, as the extra cost is negligible (£20 tops!). The problem is, as noted by Scott Dorsey, the best won't fit in the small tube. The tube would have to be bigger. I needed to make a completely new body to house the best MK012 preamp I could make.

I realised that I had a spare aluminium tube from a broken old Braun Super8 microphone. The tube is anodised black,more than twice as long as the Oktava, twice as thick (1.5mm wall, not .5mm like the Oktava) which should massively reduce the handling noise the Oktava is often criticised for, and 19mm in diameter, similar to the 20mm diameter Oktava, so a fairly close fit for the 20mm Oktava capsule. The tricky bit here is joining the two together nicely.

Here's the mic the tube came from: 


                                                                                      









It's a telescopic car aerial in a tube with a cardioid capsule on the end. 

                                                  

I have also realised that the -10dB capsule I have never used, will function beautifully as threads and connector for the hypercrdioid capsule, with only the simple minor modification of snipping of a shunt capacitor. Easy-peasy! It also adds a lot of weight and so needs chopping in half first. 

It's never going to look quite as slick as a factory made microphone, but I will make something that looks perfectly respectable wearing a Rycote baby ball gag, and sounds excellent.


Oktava MK12 "Super-modification" pre-amp build Parts List

Transistors
1x Toshiba 2SK170 BL transistor - already own several of this gem.
1x Toshiba A1015 Y £2.70 for x5

Caps
2x 68uF 63v Panny FC is 56p from Farnell (or try Silmic II if you prefer)
1x 47uF 63v Panny FC is 28p from Farnell (or try Silmic II if you prefer)
Vishay 1uF 160v MKP1839 - 70p from Farnell
1x 820pF Silver mica input DC blocking cap - already have (or try 1nF polystyrene)
3x 10nF 63v Wima Polyester MKS0 - already have

Resistors
4x 1Gohm = 2x 2Gohm TE £4.39 for x5 from RS Components (300% nominal value)
1x 33M0 Vishay VR37 £1.35 for x10 from Farnell (110% nominal value)
1x 84K5 Welwyn RC55Y 89p from Farnell (100.5% nominal value)
1x 39K2 Welwyn RC55Y 119p from Farnell (100.5% nominal value)
1x 35K7 Welwyn RC55Y 119p from Farnell (nominal value)
2x 13K0 Welwyn RC55Y 178p from Farnell (nominal value)
1x 7K5 Welwyn RC55Y 119p from Farnell (nominal value)
1x 51R1 Welwyn RC55Y 57p from Farnell (nominal value)
1x 39R0 Welwyn MFR5 128p for x10 from Farnell (nominal value)

So, my total costs for parts is:
£7.13 at Farnell, £2.70 for transistors from eBay and £4.39 from RS
= £14.22. The cost if you have no parts in your drawer is still under £20 in total.

Making a Microphone!

It's useful to have a schematic printed out that you can scrawl on. Schematics can be misleading, however. Sometimes it's just not good enough to lay things out like this. In a very sensitive circuit like a mic pre-amp, the topology is sometimes very fussy, and so I referred to the original PCB to exactly determine the order of things. Earth, for example, should follow in the correct sequence if possible, not just be randomly thrown together as the schematic seems to suggest. That could cause all kinds of unexpected nasties.

So, I had pretty much everything I needed to get started now. I was just waiting on the A1015 transistor and the 4x 1G resistors. I cut the matrix board with a jeweller's hand-saw. It won't be going dead centre in the black tube, but is off-centred to accommodate the large capacitors on one side, and is just big enough for wires and the odd resistor underneath.

The new capacitors and the old Oktava circuit board. This is where the major improvements lie. Even the fully modded Oktava has some seriously compromised caps in there, on account of space.

I am using an 820pF silver mica input cap (the brown one). If the mic sounds overly bright and not balanced, I will likely experiment with a 1nF polystyrene cap I have knocking around in place of the silver mica one, but it may be fine like this - silver mica caps are sparkling clean sounding. Another option is polypropylene of course.

The dark blue cans are Panasonic FC electrolytic caps, which are wonderful for audio. Other options I would happily use are Sanyo Os-Con's (but they are all rated at too low a voltage for 48v phantom power!!) and more practically, Elna Silmic II, which are a sweet (and very expensive) high quality audio grade cap, but I've always had superb results with trusty old Panny FCs. Besides, there are Elna Silmic forgeries around everywhere...Panasonic FC is the sensible way to go.

The white cap is a 1uF polypropylene cap that I scoured Farnell to track down. This will be much nicer than the 10uF tantalum that the Dorsey mod relies on. Scott even noted that this was a compromise for space. Most 1uF film caps are either the square Wima types, or larger 400v rolled axial types like this one, but bigger. This smaller axial type is very rare, as it is only rated at 160v, but that's fine for our purposes, and I have read engineers argue that these rolled types are better than square ones. I would ideally have used a 2uF cap, but find me one that will fit!

The small red ones are Wima 10nF polyester shunt caps which replace the 10nF brown ones on the old circuit. I think it will be an improvement, if noticeable at all. 

Transistors. Toshiba 2SK170 (BL is the IDSS grade) and A1015 (Y grade) I have never used the A1015 before, but the 2SK170 is pretty much the quietest fet available, with excellent properties for pre-amps, and used in many, many top-end audio applications. My home made hi-fi has literally dozens of them. I worked out once that I'd soldered more than 50 in total for one hi-fi, in almost every power supply and every amplification stage! A key improvement over the stock mic, for sure.

The tube from the old Braun telescopic Super8 camera mic. It's amazing, but I've had this tube in a drawer for years wondering if there would ever be a use for it, and it is extraordinary how perfect it is for this project. Much better than the original Oktava tube at 3 times the wall thickness, made of aluminium, not brass like the Oktava. The Oktava tube rings like a bell when you take the gubbins out and strike it - VERY acoustic! This one thuds - completely dead. Handling noise on a boom should be vastly improved. Lovely!

The resistors. Those singles are a stupidly expensive way of buying resistors, but they are excellent quality, and with so many odd values I didn't have, it was cheaper this way, and will make a better mic no doubt. I resisted (boom boom) substituting values other than by the tiniest percentage, which I also recommend you do - with caps you can experiment a bit more, but not resistors on a circuit like this. 38R3 instead of 39R0 is ok, and 33M0 rather than 30M0 are about as far out as I'd like to take it. Obviously the 2G resistors are 300% the designer's value, probably due to availability more than any design principle.

It's very important, as I said earlier, to copy the sequence of the board, if not the layout, not just building from the schematic, so here's the original resistor positions, so that you can trace things easier.

Here are my 2x 2 Gigahertz resistors, made by connecting some 1G resistors in series and shrink wrapping them. Lovely! 2G's not strictly necessary - 1G is fine, but Scott Dorsey recommended 2G, so I'm trying it.

The board all soldered up. It took me about 5 hours of intensive soldering to knock this together - a long time! Mostly because I am point to pointing a circuit that I am trying to cram into as small a space as possible, just working from one end to the other without a clear plan to follow, which always takes longer, trying to devise the best way to lay things out as you go.

The underside looks a mess, perhaps, but it is very clean actually for point to point work, and not fragile. 

The whole point of the Oktava for film and TV use, the Hypercardioid capsule! The -10dB pad is a nonsense for this mic. It is surely never used by anyone as the mic is known to have a moderate output, and it merely contains a tiny shunt capacitor! The whole point of this heavy chunk of aluminium is to house that stupid tiny blue capacitor!! Putting Oktava's new high pass adapter in the multi-capsule kit would make so much more sense, to cut out low frequency handling noise, which this mic is very guilty of in its stock form! Well, it will serve me much better in a cut down form, donating threads and button receptor for the capsule(s) on my new mic body. I considered implementing a sorbothane donut between the black shaft and the threads, to isolate handling noise. I don't know what a real mic designer would say. In the end, I thought that it may cause some acoustic issues of it's own, so I cemented the cut down -10dB pad to my main body with JB weld, which is also conductive, giving me a fully earthed mic.

I had originally planned to use a dongle to an XLR jack, but discovered to my amazement that an old style Neutrik XLR is 19mm, the same as my tube, and also fits like a tailored glove into my tube. It is so perfect it's astonishing. The XLR even has two grooves off-centred in precisely the same off centre position that my circuit board needs to sit, so I can secure it in the exact place it needs to go. It's almost like I had help from angelic forces this was so lucky. 

It looked so good that my aesthetic plans for the mic went up a notch and so I just had to get a black Neutrik, so add £5 to the £14 I've spent on this mic so far, and it's still pretty reasonable to say the least.

I fastened the board into the XLR housing with a bolt. It is secured into the to grooves of the XLR housing

Check it out! This fitting so well was so satisfying, you have no idea. I had to precision drill three holes and use spare screws to tap threads in the XLR socket, and a dead-headed bolt through the circuit board to link the board and the socket.

Connection to the capsule receptor centre pin. This is my "hot" wire, the return being the entire mic body. A bit like a car's electrics really.

At the other end, here is the cut-down -10dB pad, JB-Welded to the tube, which I also cut about an inch off. The tube is conductive from the capsule to the XLR GND pin, with only a tiny bit of resistance because I used silver conductive adhesive.  Aluminium solder will make a more solid joint but this aluminium has >3% magnesium and won't solder. Inside a "Rycote Baby Ball Gag" it will just look like a black stick anyway, and this capsule needs a BBG inside or out, to eliminate wind noise, pops and plosive sounds.

Oktava MC-012 and Lucas Adamson super-modded MK-012

And contrary to the original plan to hardwire the capsule, I kept it modular!

That's all well and good, but how does it sound?

Testing, testing...I will upload some samples with comparison to the original when I get a chance!

My alternative to 820pF Silver Mica - 1nF polystyrene cap. It's super-subtle stuff, judging a mic, and especially the effects of a single component switch to a similar electrical value. Trying to retain the memory of one "sound" through dismantling and re-soldering a component, and then finding an objective judgement of the differences heard is near impossible. What I can tell you is that I've stuck with the polystyrene 1nF, and that the mic sounds incredible.


The silver glue that holds the cut down -10dB pad to the main body of the mic failed and the mic came apart, and so I've had to reglue it with the same silver glue, maintaining the excellent electrical contact, but I've also reinforced it with copious JB weld and some silk rag. Next time I'd try and find a way to either weld threads on, on even cut them into the tube. The mic looks a lot less refined now, without the BBG...

..but it is still an elegant black stick when in use. This is the overpriced but excellent Baby Ball Gag, from Rycote, who also make the Lyre suspension that this mic is cradled in.



7 comments:

  1. Great stuff! Did you prefer the mica to polystyrene? Did you compare the 1uf (that was a 10uf tant) to anything else in that position?

    ReplyDelete
  2. Hi there. I prefered the polystyrene 1nF to the silver mica 820pF. That's the only substitution that I've made since building the mic.

    It sounds really good, and unlike the original has excellent handling noise, but I'd have to do comparisons to an original Oktava to judge the sound signature properly, and I no longer own one. I'd also be intrigued as to how it sounds compared to a Schoeps CMC6(MK41) and an Audix SCX1/HC, as they are the movie industry standard mics in this application (indoor dialogue).

    ReplyDelete
  3. Certainly the handling noise is reduced by a huge factor, so that's the biggest gripe with the Oktava gone. The wind noise inside a Rycote baby Ball Gag is also negligible. The sound signature is still very much that of the Oktava MK012 obviously. It sounds really excellent.

    ReplyDelete
  4. Hi Lucas, what a Mod! Thank you for posting your knowledge here. I own two Mk-012, is an outstanding mic for the price, but as you said the handling noise on a boompole is a big problem. I'm wondering whether you could add to an original mk012 a component so it solves the bass frequency problem with this mic. I already have and use the screw low cut they sell, but it really increase the noise floor... So I'm looking for a way to have the low cut inside the circuit and don't use the screw one. Or do you think you could change the components inside the low-cut in order to get no increase of the noise floor. http://www.oktava-shop.com/MK-012-100-Series-modular-system/pads/Low-cut-Filter.html Thanks in advance!!!

    ReplyDelete
  5. I have two of these and got the parts for my mod from Bill Sitler...he still sells these kits.
    I coated the interior of the tubes with a sound-deadening polymer that I got from work. These microphones are truly awesome...I did some capacitor-juggling to match them as best I could, so they are a pseudo-matched pair.

    ReplyDelete
  6. Not surprisingly that the top-end is better if you are using "2 Gigahertz resistors" :-)

    ReplyDelete
  7. Merkur Safety Razor - Yn
    Merkur Safety Razor - Merkur 700 Futur 제왕카지노 Adjustable Chrome Handle worrione - Available in Satin, Satin, Chrome Finish, Extra Long Handle, Gold 메리트 카지노 고객센터 Plated

    ReplyDelete