Chinese mic mod fun!

[gemini86]

Hi Insomniaclown,
2SK170 is not quite good choice for KM84 circuit due to very high reverse capacitanceof this FET. It's better to try 2N3819 or BF245 - both FETs available at mouser.

isn't 2n3819 the orignal part anyway?

 

[Rybow]

Hey Guys. Sorry, another case of Ryan's speak as my friends affectionately call it. I sometimes write down only 70% of my thoughts, and then think I made sense. Working on it. I really need to learn how to formulate my thoughts.

I also wish MJ sold his capsules separately, but then I am sure that would cut into his modding business! I wasn't trying to say that the RK 47 was a direct comparison to MJ's, but the "unsmoothed" part caught my eye on MJ's site. Looking at rossi's measurement, it's very smooth. So, what part of the mic's circuit is responsible for the smoothing? Is there a way to tame this ripple in the circuit? That's more what I meant. Sorry for that. The harder question is whether or not the ripple is actually there, or if it's just a bad measurement. For the money, I wouldn't doubt that that ripple is actually there.

The 2sk170's are for the current circuit in the mic which already employs the 2sk170. I am not going to use them in the km 84 circuit. Just looking to learn a little more about FET biasing, and trying to squeeze as much performance out of the stock circuit before moving on to other modding options. Sorry for the confusion.

Hoping to get the rest of the work done this weekend, and then post some pics and more samples. I was also thinking about picking up another 435 to compare it with, and then mod the crap out of. Although, I don't know what I'd do if my modded mic sounded worse than the stock mic.

 

[Rossi]

The ripple is either in the capsule (+ headbasket) or in the measurement (or both). It's not in the electronic circuit, and you can't erase or reduce it with the mic circuit.

Without a proper measurement of both capsules using the same setup (and headbasket!) it is impossible to say whether Joly's casule has a smoother curve than the microphone-parts capsule.

BTW: Whenever you look at a plot you have to keep an eye on the Y-axis scaling.

Anyway, the frequency curve is just an indicator. Quite often you can see bad design and locate pproblems in the frequency response. You do have to listen with your ears and compare to other mics in order to ascertain the actual sound quality. Keep in mind for instance that a moving coil capsule can have a similar frequency curve as a condenser mic. But it will still sound different.

 

[jordan s]

I'd be really interested in building a mic from the KM84 circuit with those microphone-parts.com capsules. This would be a pretty good starting point:
http://aurycle.com/aurycle-a460-large-diaphragm-fet-condenser-diy-microphone.html

 

[Rybow]

Sweet! That looks like a great option as well. I have looked into those Aurycle kits as well. So cheap! I like that the body has spaces for pad and lo cut to be put externally. My apex has an internal lo cut switch which is a PITA. If you do end up building it, I would love to hear your thoughts on it! At the rate I am going, I won't get to it until I am 50 years old.

Thanks for the clarification on the measuring stuff Rossi.

 

[Rossi]

I'd be really interested in building a mic from the KM84 circuit with those microphone-parts.com capsules. This would be a pretty good starting point:
http://aurycle.com/aurycle-a460-large-diaphragm-fet-condenser-diy-microphone.html

Just had a look at the FET kit: I doubt the transformer is really 10:1. The tranny in that Chinese circuit is usually much lower ratio. IIRC it was 1.87:1.

 

[abechap024]

Is the spec'ed -6db pad really useful?

 

[Rybow]

Looking over the Aurycle stuff again, I would probably just buy the A460 that's built up, and then I would gut it completely. That way you don't have to worry about painting the body, and you get a flight case and shock mount for an extra $20. Same for the a620 although no flight case or shock mount, but I like the look of the mic.

I don't know how useful a 6 db pad would be. I think 10 or 15db is a little more functional than 6db. It would be easy enough to adjust.

I had a question about the mk 319 circuit. The transformer looks to be a centre tapped one. I can't seem to find much info on the transformer aside from the ratio being 2:1. The schematic I have also doesn't list the transistor used. Anyway, I'll do some more scouring of the Meta, but if anyone has any info, please let me know!

More updates! I just picked up a 797 audio k67 capsule out of the black market. So now I have a new plan of attack. I think I'll end up combining the 797 capsule with my upgraded stock circuit along with some HF de emphasis for super cheap mic mod #1. Mic #2 will have the rk 47 capsule with either the km 84 or mk 319 circuit, for mid priced but still cheap mod. Mic #3 is a maybe right now depending on finances, but royer tube mod with Peluso something or other. That should give me 3 different but useful condenser mics on a budget.

 

[Rossi]

No, the Oktava  MK-219/319 transformer is roughly 8:1. Someone once posted measurings of a couple of those, maybe you can find that thread.

The FET is some russian type and changes from time to time. If you don't have excess noise, just keep it as the source resistors are specificially selected for each mic (their values are penciled on the transformer cover). FETs vary a good deal, even within the same parts number. If you replace that FET for whatever reason, you have to determine the correct values for that exact FET. As the Oktava circuit uses two source resistors (one of which bypassed with a cap), this is more difficult than in the otherwise similar KM84 circuit.

 

[Rybow]

Thanks for that Rossi. My sights are back on the KM schematics. Just read an old post of yours Rossi, and my sights are set on the KM 86. The RK 47 being double sided, I would love to get some multi patterns happening.

Edit: Km 86? Nope. Just spent some time pouring over the schematic. Due to the shared back plate of the RK 47, the KM 86 and u87i options are out. I know Omni is easy enough to implement, but it's that figure 8 that is tricky. I have looked at the u87a schematic. Might look in to doing that down the line. I also looked into a CMOS based DC supply, and I read a great document about that on this site. Still investigating. The other option is to just build an external power supply, but then I might as well build a tube circuit and call it a day.

So in short, multi pattern mic with a double sided/shared back plate capsule = no easy answer. KM 84 wired with omni should suit me just fine. Just talking with David Geren at Cinemag about some transformer options.

 

[MatthisD]

You can use the DC board for 60V polarization voltage (60V is the usual value for most capsules); you can't use it for the FET drain voltage as the current capabilities of a typical dc converter are way too low.

So I'd recommend building the full KM84 circuit, and when you get it to work properly you can use your DC converter board for the polarization voltage.

Hi, I have been putting a km84 circuit together and I am planning to house it in t.bone SDC body and use its capsule. In relation to the above; will the 40 something volts in the neumann circuit be inadequate for the capsule which normally runs off 60v, will doing so create a noisy signal?
Matt

 

[Rossi]

Yes, it will work. At 44-46V you lose about 2 dB in sensitivity and s/n compared to 60V. That's no problem with most SD capsules. Which t.bone SD are you talking about?

 

[MatthisD]

sc140, 2db is not a big deal then. I have tested it and it passed a signal but with a temporary fet which was not biased, thats my next step, thanks Rossi
Matt

 

[Rybow]

Cool Matthis! Me want pic!

 

[MatthisD]

I need to do a new board layout and decide on a way to attach the transformer, for now its very scrappy and I used the Neumann schematic and not the stripped down version you can find in the forum. After biasing its working well, requiring the same preamp gain as the sc140 to get similar levels.

 

[Rybow]

Sounds good! I am actually working on an LDC with the KM 84 circuit (full schematic). Still sourcing parts for the body. What transformer did you go with?

Good to hear that after biasing the gain is good. I think that's my issue with my MK 012 circuits that I built. I picked up a cheap 24VDC wall wart and some breadboard jumpers to do biasing, but I have yet to put it into practice.

 

[MatthisD]

BV-107, the way the board is positioned in the mic barrel means the transformer sits slightly above the centre so its a tight squeez! I'm very new to all this so I can't offer any advice but I biased my fet in circuit, with phantom power connected, a 1k signal fed into the gate and a scope on the drain to tell me whats happening.
cheers
Matt

 

[Rybow]

Cool man! Hopefully it all fits in nicely. I'm new to mic building as well. I am actually building a "junk mic" right now as practice. I am guessing you got the BV 107 on ebay? I saw a few of those on there. I ordered a cinemag cm 2480, but then decided on a Tab funkenwerk t49 for this mic. All is good though. I'll have another mic transformer to build/mod/upgrade another mic one day.

I recently decided to replace the circuit in my Apex 180's. I hear you on the tight squeeze aspect.

I think I have the biasing thing figured out. Gus wrote a good post about it that I have saved somewhere. Just have to try it.

Good luck with the build! I can't wait to hear your impressions of the sound.