9/12/16

Oktava MK-319 Modifications Pt. 3: Circuit Modifications

Now that we have the 2SK170BL biased, let's finish the circuit modifications.

The circuit is the same as the MK-219, so all of the changes to the capacitors are the same.

On the left is an illustration I did to show what is where on the MK-319 board. The most critical changes are to R1 and R2, to C2 and C8. C2 is the input cap, and C8 is the signal cap to the output transformer. I put an Elna Silmic II in this position, as I did on the MK-219.

And if you're wondering, here's a side-by-side comparison of the MK-219 and MK-319 boards.

Same stuff, different locations for some components.

One of the bad changes on the MK-319 compared to the 219 was the relocation of the pad/bass cut circuit. Most of the components were moved to a separate pcb with the switches on it.

That relocation meant that there are very long leads going to and from that board, which degrade the signal from the capsule.

To rewire it, we need to remove the capsule mount from the main pcb.

There are two sets of leads - one from the capsule to the switch board and back, and one for the bass cut which runs from the main pcb and back.

I'm not using the bass cut, and in fact I removed the switches entirely. So I'm removing all of the factory leads and putting new, much shorter leads in from the capsule to the main board.

On the picture with the ruler, you can see the length of the factory signal lead which runs from the capsule, to the pad switch and back. It's about 175mm! That puts a lot of additional capacitance on the input which degrades the signal. A really bad idea.

The same lead on an MK-219 is about 20-25mm. That's what we'll wind up with.

I cut open some Mogami 2893 'high definition' twisted-pair coax microphone cable. The cable inside is oxygen-free low capacitance, so it's great for rewiring out Oktava.

I got really crazy and decided to use the Mogami wire for the output leads also.

The XLR connector is held on to the main pcb with two screws. Undo them, and you can access the solder connections for the output connector.

Here we have our new wiring in place.

Maybe a bit overkill, but easy to do.

I marked the pcb with the numbers of the output connections to make sure I rewired them correctly.

You can also see those two screws I mentioned that hold the XLR to the board.

Then we put new, shorter leads onto the capsule.

Handle the capsule with care - if it gets punctured, it's ruined.

The rewired and reassembled capsule.

Note the red wire - that's only about 20mm long, instead of 175mm stock. It connects right at the top side of the new cap to the right of the mount.

Since I had the capsule mount off the board, I decided to put a piece of closed-cell foam between the mount and the board. Hopefully it will absorb vibration.

Here's the completed MK-319.

The mods I made are:
  • The top screen and body are dampened with CLR material, and the inner screen is removed
  • The JFET is now a Toshiba 2SK170BL biased correctly
  • The capacitors are upgraded
  • The capsule and output wiring has been upgraded with Mogami wire
I still need to do more tests, but the mic sounds more 'open' and the bass is tightened up a lot.

UPDATE 13 September 2016: Did some recording (acoustic guitar) with it. Compared to the MK-219, it's much more open, has the same bottom end, but sounds really clean, airy, spacious and sparkly on the high end. The MK-219 sounds a bit 'boxy' in comparison. I really like it.
at

3 comments:

  1. timOctober 13, 2016 at 2:25 AM

    nice.. i just did a Bill Sitler mod on an mk-319.. it sounds great, a big improvement.. i used adhesive cork to dampen but i think i'll try the material you used. i do notice that i have to use more gain on these mics with my preamp than i do on my other condensers like a langevin cr3a.. is that basically what you solved with your biasing? more output? sensitivity? how much for one of your mods if i send you a mic? just the electronics mod. i can handle the damping.
    cheers,
    tim

    ReplyDelete
  2. timOctober 13, 2016 at 2:26 AM

    This comment has been removed by a blog administrator.

    ReplyDelete
  3. Yr Fthfl BlggrOctober 31, 2016 at 8:19 PM

    I could do a mod for you, but I don't think the way I biased the transistor is any different than what Bill does.

    I presume that he does it the same way - with a scope and trim pots to determine the resistor values. In his kit, he supplies the two bias resistors, so he has to have a way of determining the proper bias. All I did differently was to leave the trim pots in my mic.

    As to the gain, remember that when you change the transistor, you're changing the device, meaning that the gain will likely be different. No one seems to know the specs on the stock Russian JFET, so unless one has a way of measuring its gain, its spec just isn't known.

    We do know the Toshiba transistor is one of the quietest ever made, which is why it's a popular swap into the Oktavas. I *have* noticed that the gain of the Toshiba does seem to be a bit less than the stock JFET, and that could be what you're experiencing. That's just by seat of the pants evaluation (or by turning a gain control :-) ).

    Hope this helps.

    ReplyDelete