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.
What is "K1" near the capsule?
That's playing the role of the capacitance of the capsule itself.
What is the "R*" 4.7k resistor? On the OG schematic, there's a 25k trimpot on the source leg of the 2N3819 JFET. Is the "R*" just a standard 4.7k resistor?
You'll probably want that to (at least temporarily) be a trimpot indeed. 4.7k just happened to be the optimal value that was appropriate for the simulation (i believe..?).
The D1 Zener Diode is noted specifically as a CDZV36B. Will any decent 36V Zener Diode do the job in its place, or is there something else I should be aware of?
Once again, a simulation thing. Any 36V zener will do, indeed.
What are "PR1" and "PR2" on this schematic?
Probably standing for "Phantom Resistor 1/2" - those are (by standard) inside your preamp. They're in the simulation schematic in order to have the same conditions for the powering as you'd get in real life (to get the appropriate voltage readings etc).
And lastly, are pins 2 and 3 from the XLR connector joined here?
Yes, XLR2/3 are at the node between PR1/2 and the respective 2.2k resistors.
I have begun putting together a list of components on RS, but I'm in doubt of which cap types to use
You proooooobably don't need 250V rated caps for the 33nF's - those will end up being on the huge side. Just plain old 63V types are plenty fine, and they're tiny by comparison.
As pointed out a few posts ago, you can replace that single 10nF with a(nother) 33nF, but once again, no need to go for ridiculously high voltage ratings, this is not a tube mic 😁
For the 120pF cap, you'll want to make sure the material is C0G / NP0. Can't see an exact part number in your parts list, but i recently stocked up on some of those tiny blue TDK ceramic caps. Colour is no guarantee of construction though, so you'll want to check that in the datasheet.
Or should I just go ahead and build the ORS 87 Modern PCB and do the biasing straight on the board itself with a trimpot of some variety in place of R*?
You might consider splitting up the R9 into... I don't know, 6.8k + 3.3k, or 7.5k + 2.2k, something in that ballpark - see the original U87 schematic. There, R9 is actually a 6.8k + 560 resistor, and one can apply signal from a generator (a line / headphone output, in your case) there, and be able to monitor the output of the mic connected to a preamp input (on your interface, which would also provide the phantom power).

Trying to get a signal connected "in place of" the capsule gets complicated and delicate, if you also want to have the effect of the tone-shaping feedback network.
 
That's playing the role of the capacitance of the capsule itself.

You'll probably want that to (at least temporarily) be a trimpot indeed. 4.7k just happened to be the optimal value that was appropriate for the simulation (i believe..?).

Once again, a simulation thing. Any 36V zener will do, indeed.

Probably standing for "Phantom Resistor 1/2" - those are (by standard) inside your preamp. They're in the simulation schematic in order to have the same conditions for the powering as you'd get in real life (to get the appropriate voltage readings etc).

Yes, XLR2/3 are at the node between PR1/2 and the respective 2.2k resistors.

You proooooobably don't need 250V rated caps for the 33nF's - those will end up being on the huge side. Just plain old 63V types are plenty fine, and they're tiny by comparison.
As pointed out a few posts ago, you can replace that single 10nF with a(nother) 33nF, but once again, no need to go for ridiculously high voltage ratings, this is not a tube mic 😁
For the 120pF cap, you'll want to make sure the material is C0G / NP0. Can't see an exact part number in your parts list, but i recently stocked up on some of those tiny blue TDK ceramic caps. Colour is no guarantee of construction though, so you'll want to check that in the datasheet.

You might consider splitting up the R9 into... I don't know, 6.8k + 3.3k, or 7.5k + 2.2k, something in that ballpark - see the original U87 schematic. There, R9 is actually a 6.8k + 560 resistor, and one can apply signal from a generator (a line / headphone output, in your case) there, and be able to monitor the output of the mic connected to a preamp input (on your interface, which would also provide the phantom power).

Trying to get a signal connected "in place of" the capsule gets complicated and delicate, if you also want to have the effect of the tone-shaping feedback network.

Thank you so much for the input. I didn't consider the physical size of the components, so the large voltage ratings were a by-product of wanting to get specific brands of caps (my dad is a long-time nichicon and rubycon fan) and not having to buy a roll of 5000 components haha :D

Are there any brands that people are particularly fond of? I know Panasonic and Wima also enjoy good reputations among this crowd, but if there are any other standouts I should consider, by all means let me know.

The only c0g ceramic cap I could find on RS is this one: https://dk.rs-online.com/web/p/keramiske-flerlags-kondensatorer-mlcc/1940472
Unfortunately, it's only rated for 50V. Guess I might try my luck on mouser. I presume whether it is multi-layered or single-layered is of little importance.

As for the R* trimpot, I presume just about anything will do. I might get a 10k one like this: https://dk.rs-online.com/web/p/trimmepotentiometre/8967140
Anything wrong with that?

I will revise my component list according to your recommendations. Once again, thanks for the help.
 
my dad is a long-time nichicon and rubycon fan
I can relate, although i'd expand that to the other Japanese brands too (Panasonic, Elna and United Chemi-Con as well).

Are there any brands that people are particularly fond of?
Film caps? I don't care. Overall, any film caps are better than random ceramics (apart from C0G/NP0, but finding those in larger values than 10nF gets physically large and/or really pricey).

The large distributors, though, largely only carry reputable brands anyway, so less risk of duds.

The only c0g ceramic cap I could find on RS is this one: https://dk.rs-online.com/web/p/keramiske-flerlags-kondensatorer-mlcc/1940472
Unfortunately, it's only rated for 50V.
That looks perfectly fine; why would 50v not be enough? Phantom power's (by standard) 48v, and that particular cap won't be exposed (on its own) to that full voltage (which will get pulled down by the consumption of the rest of the circuit anyway - remember the phantom power gets fed through that pair of 6.8k resistors, and Ohm's law always applies).

As for the R* trimpot, I presume just about anything will do. I might get a 10k one like this: https://dk.rs-online.com/web/p/trimmepotentiometre/8967140
Anything wrong with that?
Might wanna go for a 20k / 22k / 25k, "just in case".
 
Yeah, everything @Khron said is right. A lot of what's in the schematic is for simulation purposes.

You could probably get away with 50V for all of the caps (except maybe C11/C13), since the rest of them will likely only see up to ~47V, but it's best practice to build in some buffer. Generally, you want the expected voltage to be less than ~80% of the capacitor rated voltage to ensure long life of the capacitors and prevent failure.

And lastly, are pins 2 and 3 from the XLR connector joined here?
One point of clarification, just to be sure - pins 2 and 3 are not connected directly. They are connected to the 2K2 resistors (which do connect together), and to the secondary inputs of the transformer. If they are shorted together, you won't get any signal.
 
That looks perfectly fine; why would 50v not be enough? Phantom power's (by standard) 48v, and that particular cap won't be exposed (on its own) to that full voltage (which will get pulled down by the consumption of the rest of the circuit anyway - remember the phantom power gets fed through that pair of 6.8k resistors, and Ohm's law always applies).

You're absolutely right, that's me having a brain fart.

Might wanna go for a 20k / 22k / 25k, "just in case".

Roger that, I'll do so.

One point of clarification, just to be sure - pins 2 and 3 are not connected directly. They are connected to the 2K2 resistors (which do connect together), and to the secondary inputs of the transformer. If they are shorted together, you won't get any signal.

Right, I couldn't quite make sense of it. Thank you for the clarification, it is very much needed, as I have no idea what I'm doing.
 
Tried the uber-affordable Neutrik NTE 10/3 transformer in the build and I really like it. Sounds great. They're like $13 at Mouser and tiny.

And of course, since it's designed 1:10, you have to wire it backwards to make it 10:1 to work here, I wanted to just post the guide of how to connect it to this circuit.

Red = P-
Black = P+
Yellow = S-
White = S+

You don't need the blue wire.
 
I have to ask. Why does that eliminate the 2k2's? What purpose do the 2k2's serve?

They're there to tap the phantom power voltage present on both signal lines.

With a center-tap on the transformer secondary, one can use that as the power take-off point (the phantom power DC is present on both signal lines, relative to ground). Removing them also has the mic work into a lower load (higher equivalent input impedance - keep in mind that pair of 2k2's is in parallel with the input impedance of whatever preamp you connect the mic to), as well as allowing a teensy bit more voltage / current available to work with. And hey, two fewer components to buy / worry about :D
 
They're there to tap the phantom power voltage present on both signal lines.

With a center-tap on the transformer secondary, one can use that as the power take-off point (the phantom power DC is present on both signal lines, relative to ground). Removing them also has the mic work into a lower load (higher equivalent input impedance - keep in mind that pair of 2k2's is in parallel with the input impedance of whatever preamp you connect the mic to), as well as allowing a teensy bit more voltage / current available to work with. And hey, two fewer components to buy / worry about :D
Yup, all this. The 2k2's essentially create a "virtual center tap" on the secondary of the transformer for the DC (phantom power) to flow through. If the transformer already has a center tap, you don't need them. Another benefit is that the two halves of the transformer winding are usually better matched than most resisters. If you're using a transformer not expressly made for this though, you need to make sure the windings can take DC (not all can).
 
Yup, all this. The 2k2's essentially create a "virtual center tap" on the secondary of the transformer for the DC (phantom power) to flow through. If the transformer already has a center tap, you don't need them. Another benefit is that the two halves of the transformer winding are usually better matched than most resisters. If you're using a transformer not expressly made for this though, you need to make sure the windings can take DC (not all can).
That sounds an improvement. Which inspires the quesiton: Where does the center tap connection connect to the board?

I looked and Terrapin doesn't sell that transformer anymore, but I like the idea of trying a torroidal in there at some point as I really do like the way the Soyuz mics sound.

@Khron - Thanks for the explanation :)
 
That sounds an improvement. Which inspires the quesiton: Where does the center tap connection connect to the board?

I looked and Terrapin doesn't sell that transformer anymore, but I like the idea of trying a torroidal in there at some point as I really do like the way the Soyuz mics sound.

@Khron - Thanks for the explanation :)
Paging @Marik. Do you have any ~9:1 torroids with DC-capable center taps that could be available to the DIY community? :)
 
@igs
Hi Igor!
Would you be willing to make a UTM 0587 TC variant?
🔽
*To internally connect the secondaries in series but also to add externally a wire from the connection point, a Tap-Center,
or
*Leave the secondaries unconnected like Neumann, we connect them externally.
✨
We could thus obtain the supply voltage for the circuits, without using the 2.2k resistors.
You said you can customize different configurations.
You could permanently include it in the offer, I know that many would like such an option.
Thank you for UTM transformers!
 
Looks right based on a quick glance. Depending on the body used, sometimes you have to be careful about placing tall components (i.e., some caps) near the edges due to the body tube limiting the space.

Also important to note that no one NEEDs a PCB for this project. It is handy, but you can breadboard this circuit for very little money and effort, which was my original point in posting the idea.
 
My apologies. I just could not resist.

Any questions, comments or suggestions???

Regards!

No concerns for the high-impedance area of the circuit? 😬 And for longevity's / reliability's sake, it's customary to connect the wiper of a trimpot to either end (so things don't completely "blow up" if the wiper happens to go open-circuit / intermittent).

All in all, a pretty... convoluted(?) layout.

I'll leave the questions about the 1KV-rated output cap for a later time, though...
 
Last edited:
Back
Top