OPEN SOURCE DIY Mic Project - ORS 87 - Stripped Down u87

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Ok, here's a few questions on C7.
Does C7 affect the tone?
Does C7 affect transient response?
Is this a component I should spend a little more for?

Additional notes:
Found 220nf is way too much low end boost for C6 for my taste. Switched it to 33nf for the time being.
150pf may also be a bit too much high end boost, will probably try 100.
Does anybody even make 20uF's anymore? Not that they're not (at the very least) +/-20% tolerance to begin with, anyway... :)

Not trying to be snarky, but how else will anyone take notice and consider trying to cure themselves of "vintage blindness"?..
I was able to get them from Mouser, but they were $4.50 each, and only in axial format, and larger in size than I expected. So, I'm happy to be able to replace it with a wider choice of 22uf.
 
Quick vocal test. No EQ.
More tweaking to come, but I'll stick with 33nf at C5.
(Thanks @Khron for catching it) I'm not hating the brightness of this mic, yet, but I think the next build will get a higher value cap.

The whole circuit is less than $20 in components. The most expensive thing was the $39 3u GZT 87 transformer.
Cheap body which I had laying around was about $25 when I got it. Capsule is a rescue K87 from a BaiFeiLi V47, which will get swapped out for something better.

Questions I'm wondering about:
What is the absolute cheapest build of this mic possible that still sounds OK?
What is the project cost of going more upscale with a build. (Arienne K87, Moby Transformer, HL-95 body)?

This project will probably be my DIY obsession for the forseeable future.
View attachment OneRoomStudio-Basic87-Test1.wav
 
I found that it's useful to start with the stock values (i.e., C5=33nF, C6=220pF, C8=22uF), and see how your mic sounds with the capsule/body/transformer you have. With this simplified circuit, C6=220pF is probably going to be too dark, so something in the 120-180pf region will probably be better, depending on the brightness of the capsule used. I think the preference for C5 value has a lot to do with how you plan on using the mic. For up-close voice work (which is arguably what the U87 was designed for), I really like the stock value. I think it has just enough low end without being boomy. For distant-micing (i.e., drum overheads, orchestra, folk instruments, etc.), that can sometimes sound thin, and a larger cap (between 100n and 300n) might be better.

Just keep in mind that changes to the bass and treble don't happen in a vacuum. Cutting more bass can lead to a mic sounding subjectively brighter, because as you increase gain to make up for the lack of bass, you end up boosting what's left - including treble. Also, if you increase the bass and treble too much, you're effectively decreasing the mids (making the mic "flat" or even adding a slight "smiley face" curve), which takes away a lot of what the "U87 sound" is.

To each their own, of course, but personally, I like the stock values quite a bit (with slight tweaks to C6 to account for capsule brightness).
 
Talking about Neumann and 'Ali' (Express): today I received this message.
It seems they now really have a problem with Neumann!

Dear Customer,

We hope this message finds you well.

Your satisfaction is of the utmost importance to us, and we strive to ensure that the products available on our platform meet corresponding standards.

We have identified a potential non-compliance issue with a recent product you have purchased.
Product Name:"U87AI studio microphone U87AI U67 M149 TLM103 TLM107 professional condenser mic pc gaming recording micro, u87ai with logo"
Cause for the potential problem: Intellectual Property Infringements
We take your safety and satisfaction very seriously, and we want to bring this matter to your attention promptly to ensure your shopping experience with us is in compliance with applicable laws and your expectations.

You may contact the seller in order to find more information and in case that the product is actually illegal, negotiate a solution such as requesting the return address of the seller and returning the product to the seller.

If however there is actually no issue with the product, please disregard this message. If you have any concerns about the product or have any further related questions, you shall contact the seller directly. We regret any inconvenience this may cause.

**Contact the Seller:** You can do so by navigating to the "My Orders" section and selecting the specific order in question. From there, you can contact the seller to discuss the product's concerns. The seller may be able to provide more information. If you made this purchase for someone else, please notify the recipient of this issue. Information on the seller can be found on the listing page corresponding to this product.

Thank you for choosing AliExpress, we are here to assist you every step of the way.

Best regards,
AliExpress
Ruud,
Interesting.
Is this the first time you've received such a message from them concerning Neumann infringement on Ali? I wonder what, if anything they'll do about those listings?
And maybe also why the prices on the knock offs had dropped significantly even before the current Ali Anniversary sale.
 
I was able to get them from Mouser, but they were $4.50 each, and only in axial format, and larger in size than I expected. So, I'm happy to be able to replace it with a wider choice of 22uf.
amplifiedparts.com has 20uF's for cheap. But yeah, big and axial. Wordsushi, email me and I'll give you some (I have a ton because I use them in my Dumble clones).
 
@OneRoomStudio, question.
While showing off this project to a couple of other DIY people, the question has come up of whether the omitted C3, 10p NFB, is needed to keep from overloading the FET. On my build, it does seem to be a robust signal and though I haven't noticed it clipping when I speak normally in it, the thought was that a loud signal would clip it hard without C3.

And the other related question was, in the absence of C3, could the signal be made a little less hot by increasing R18/19 slightly?

I was just wondering if you had noticed any of this when you simulated the circuit. Overall, it sounds good to me how it is, but I get where their questions are coming from.
 
@OneRoomStudio, question.
While showing off this project to a couple of other DIY people, the question has come up of whether the omitted C3, 10p NFB, is needed to keep from overloading the FET. On my build, it does seem to be a robust signal and though I haven't noticed it clipping when I speak normally in it, the thought was that a loud signal would clip it hard without C3.

And the other related question was, in the absence of C3, could the signal be made a little less hot by increasing R18/19 slightly?

I was just wondering if you had noticed any of this when you simulated the circuit. Overall, it sounds good to me how it is, but I get where their questions are coming from.
If you’re using the microphone for voice over work, I can’t imagine that you would need C3. The U87i circuit already has more headroom than the current U87ai circuit (48V vs 60V on the capsule).

Feel free to try it out though if you think you need it.
 
Also worth noting, if you do have problems with overloading the FET, you could use a 33V zener to increase the FET headroom relative to the capsule voltage. You should have plenty of headroom for all but the loudest sources.
Thank you! I really appreciate the help. I have some 33v zeners and a bunch of 47k resistors so if that will open up the headroom even more, I'll do it. That'd be perfect. Little optimizations like this that help make it the purpose-driven tool I'm looking for are awesome.
 
A few other notes:

1) As @Khron would probably point out, the backplate connection (BP at the top) isn't actually a high impedance connection, so you don't really need a teflon insulated pin. I just personally find it easier to make the capsule connections to pins. Feel free to omit that one, but still use one for the front diaphragm connection (FD).

2) R7 doesn't have to be 1G if you can't find it. Anything down to a few hundred Meg is probably fine.

3) There's no rule saying C7 needs to be a big film cap. A tantalum or even electrolytic would work too.

4) This is a fairly forgiving circuit to experiment with. Don't concentrate too much on getting the exact values of any of these parts (other than closely matching R18 & R19).

5) Don't forget to bias up the FET by adjusting R10 for lowest distortion! I'm a fan of using a spectrum analyzer (or an EQ plugin with a spectral view), and dialing things in for the greatest fundamental (i.e., 1kHz if that's the signal you're injecting), and the lowest harmonics. For me, that works better than the 'oscilloscope technique,' since it can be hard to judge the symmetry of a sine wave. To each their own though.
Is a film cap the best sounding option for C7?
What difference would an electrolytic (or a ceramic) make at C7?
 
Is a film cap the best sounding option for C7?
What difference would an electrolytic (or a ceramic) make at C7?
Well that’s a rats nest of a topic, haha. I’ll say this - the value (capacitance) of the capacitor matters exponentially more than the type.

When it comes to types, there are a lot of opinions. Generally, many people don’t seem to like electrolytics, but it’s worth noting that those were what were used originally in many classic microphones. Some people like tantalum’s (wet or dry), some people like paper-in-oil. Some people like film, but only certain films. There are people who claim to hear the difference between different caps of the same dielectric but with different voltage ratings. Some people use bypass caps of various types. There’s a lot of snake oil out there.

My opinion is to not spend too much, try a few things out, and see what you like. I used a PET cap in mine because I had one in my parts box. Sounds fine to me.
 
@OneRoomStudio, question.
While showing off this project to a couple of other DIY people, the question has come up of whether the omitted C3, 10p NFB, is needed to keep from overloading the FET. On my build, it does seem to be a robust signal and though I haven't noticed it clipping when I speak normally in it, the thought was that a loud signal would clip it hard without C3.

And the other related question was, in the absence of C3, could the signal be made a little less hot by increasing R18/19 slightly?

I was just wondering if you had noticed any of this when you simulated the circuit. Overall, it sounds good to me how it is, but I get where their questions are coming from.
I'm starting to reconsider my stance on C3. While the headroom part doesn't matter too much to me, C3 does linearize the response of the circuit. If you're feeling like you still don't have enough top-end, even after making C6 smaller, try adding C3. You'd just have to put a 10pF (technically, 9.792pF since it's in series with C1) from the front diaphragm to "top" of C6/R8 (the side not connected to the backplate). You'll lose about 14dB of signal, which is a lot (and will result in more noise when you make up that 14dB), but you may prefer the sound. Worth experimenting with anyway...
 
I'm starting to reconsider my stance on C3. While the headroom part doesn't matter too much to me, C3 does linearize the response of the circuit. If you're feeling like you still don't have enough top-end, even after making C6 smaller, try adding C3. You'd just have to put a 10pF (technically, 9.792pF since it's in series with C1) from the front diaphragm to "top" of C6/R8 (the side not connected to the backplate). You'll lose about 14dB of signal, which is a lot (and will result in more noise when you make up that 14dB), but you may prefer the sound. Worth experimenting with anyway...
https://groupdiy.com/threads/how-does-condenser-mic-capsule-load-affect-thd.82796/post-1071424
 
I'm starting to reconsider my stance on C3. While the headroom part doesn't matter too much to me, C3 does linearize the response of the circuit. If you're feeling like you still don't have enough top-end, even after making C6 smaller, try adding C3. You'd just have to put a 10pF (technically, 9.792pF since it's in series with C1) from the front diaphragm to "top" of C6/R8 (the side not connected to the backplate). You'll lose about 14dB of signal, which is a lot (and will result in more noise when you make up that 14dB), but you may prefer the sound. Worth experimenting with anyway...
That's interesting. Something to definitely consider, albeit so far I'm not hating the sound of the mic without it. It kind of has a "full-throttle" feel to it, but since I haven't spent enough time testing it I can't tell how well it'll work in this original configuration yet. I was planning to transplant the working one I have into a good mic body with a good capsule to gauge if the signal is too hot or not. If it is, it may still have a place on stuff that isn't too loud because so far, it sounds pretty good on making quieter voice stuff feel bigger.
 
Simplest i can give you is if you want extra headroom reduce the polarisation voltage. Adding capacitance one or the other way comes with penalties.
100% agree that if your goal is simply to decrease sensitivity (increase headroom), then lowering the capsule polarization is the way to go. You'd have to lower it a lot to get the equivalent 14dB of reduction though (you'd have to go down to ~12V vs 60V to lose 14dB).

My point was that there might be other reasons to use the NFB cap - one would be to change the frequency response and add linearity to the circuit. This is different than using a capacitor in parallel with the capsule.
 
Back
Top