Topic: Transound TSB-120A microphones

[Chuck]

As usual, I'm late to this party...

I've read the discussions here and on MicBuilders but still have some questions.

I found I really like the sound of the Transound TSB-120A capsules. They fit on the Audix M12xx bodies, but the Audix pre-amps draw too much current.

I like Chris Church's Audix cable, but I want a slightly longer body that I can use to mount on a stand for open taping. So, I bought some of the capsules with FET to experiment with. (first image)

I know the FET is junk, and the resistor needs to be changed to 4.7k. But, what FET is best to replace the stock 2SK596 with?

I'm also half thinking about using a Naiant PAF to power this and also have the ability to use a Church STC-9000.

So, I want to make sure I got the electronics part right. The original circuit is a three wires per mic. Can I use the two wire 4.7k mod circuit pictured to make it two wires per mic?

Are there any changes I need to make or can I just swap the 2.2k resistor with the 4.7k find a better drop in replacement FET and be good to go? (second image)

I'd appreciate any insight. This is just a project to keep me busy... and it looks like an easy one unless I made some bad assumptions on the way.

[SmokinJoe]

You know more about this than I do...

I'm just thinking if it were my project, I would try them in the 3 wire setup as suggested before I jumped into the resister mod 2 wire approach.  Try to establish a baseline.  If you want little bodies, my first idea would be to cut up a couple of mini-maglights (AA battery type), use that for the body and holder for AA battery (or 2). The battery goes between +V and ground, the RC gets tucked in the body there somewhere, and Signal+ and ground on Left/Right come down to a mini-plug TRS.  You plug that into your recorder or ST9000.  The cap in the RC blocks plug in power from the recorder of ST9000.  You remove the battery in the mod-flashlight between uses.

The caps aren't that expensive, right?  I think I would do one pair this way, then another pair with the 2 wire resister mod, and see how they compare.

[Chuck]

The extra length on the body I'm going to have to fiddle with. Since I know I have to seal the back of that FET case, I has thinking about using epoxy to mount something that I can easily grab to hold in a stereo mic bar. I'm going to figure out how best to extend the body after I get the circuit figured out. If I need more room for components in the mic, that'll dictate how long the body will have to be. AA or AAA batteries won't work, as I'm powering it with 9v. Plus batteries are heavy.

I have thought about keeping them as three wire mics, as I already have some Belden cable that would work perfectly.

I bought several of the capsules ans FET's, so I might end up with a couple different versions as you suggested. I'd also like to try a different FET than Chris uses. There's no sense in re-inventing it. Since Chris did a great job figuring out how to do it already. I just want my own that is a little longer and runs on phantom power if needed.

The whole circuit board pulls out of the bottom of the M1290's. So, I could even use those bodies and make my own 48v to 9v converter and replace that with the original circuit. I even have the exact same mini-xlr jacks that screw into the bottom of the M1290's.

[Chuck]

Some guy used to sell matched 2SK118-0's on eBay. Then, I could use the Nak 300 circuit. That would solve the FET matching problem too.
I know this design is workable.

[Chuck]

That may just be the perfect solution!
But, not much for me to do as far as a project... 
But it will solve my problem...  cheaply...

I can use perfboard to go from the mini-xlr jack to the capsule end and keep the M1290 brass bodies.

I haven't received the capsule assemblies yet. So, I don't know for sure, but I should be able to mount the TSB-120A spring steel loaded FET section on that perfboard with epoxy to mate with the capsules.

Man, if I could find some m1245 brass bodies (much shorter than the M1290's) it would be perfect!

Thanks Jon! I am worried about that FET though. Everything I've read says it's not that great. But, of course I defer to your knowledge and judgement about that. The guy I bought the TSB-120A assemblies from said he matches them +/- 1db.

[MIQ]

Interesting thread. 

I've been trying to learn more about low noise preamp design lately too.  A couple of good articles have appeared in the latest issue (Vol. 3) of Linear Audio http://www.linearaudio.net/ that you guys might find interesting.  The second part of Scott Wurcer's article on DYI Low Noise Mic Preamplifiers and Samuel Groner's article on A Low Noise Laboratory Grade Measurement Preamplifier.  The Groner article has a lot of good info on how to minimize noise in common source JFET front ends.  Jon mentioned the reduction in noise for higher currents in the JFET.  This is covered and it's only true up to a point.  Higher currents will require lower drain resistances (for a fixed PIP voltage) which will increase the noise of the JFET stage and higher currents can lead to increased dissipation = temperature of the JFET and this will also increase the noise of the JFET stage.  He's included a graph of "Measured vs Calculated Equivalent Input Noise Density for a Typical JFET" and it shows the the reduction in voltage noise density stops falling at close to the "calculated" rate at about 2mA of drain current.  It's still going down but at a much smaller rate of change.  The Groner preamp design ends up having an equivalent input noise density of 390pV/root Hz  down to 100Hz.  That's nuts.

The BF862 is a low cost and easily sourced low noise JFET that many people have been using lately.  Unfortunately the input capacitance is 10pF, so like Jon mentioned, you would loose about 3dB.  Finding a low input capacitance JFET with better noise performance may be a real challenge.  If you find some, it might be very hard to buy only a handful.

-MIQ