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

[AdamFrisen]

As this project inspired me to build microphones...

I wanted to replicate U87AI original schematics with capacitance multiplier and 60V for LDC in a separate board (starting with only cardioid).

I used KiCad 8.0.

The boards should fit inside the smaller U87-style body like this one:

https://www.aliexpress.com/item/100....order_list.order_list_main.59.985a1802Qlm5WZ

...and still have enough space for UTM-transformer.

Four capacitors are meant to be soldered on the backside (C1, C8, C9 and C12).

I added an optional NFB capacitor (C13) just the U87 classic has (cause I liked the sound of it when added).
The DC-board using LTC3290, I realized MSOP-10 might be too small to hand solder. OPIC 60V should be easier.

I will replicate U87AI dual DC for LDC and see if that will fit later.

Modifying to get a cardioid version with 60V should be just to remove few connectors for switches and remove some traces.

The big 1uF output polyprop. (MKP1349_1uF_160V) and Bourn trimpot (PV36W) footprints are inside the zip (custom_lib.pretty-folder).

All the footprints are from components I found in Mouser.

I'll post some pictures when I get the PCB from JLCPCB/PCBWay and have them working (or not) and installed inside the mic body.

My intention is to share the U87AI schematics in a complete CAD project which you can modify yourself and order PCB from JLCPCB, PCBWay etc.

I removed silkscreen from the input area, you can bring them back by deleting the components and sync from schematics or some better method in KiCad. These are the three first projects ever I made in KiCad so consider me as a noob learning.

NOTE: XLR 123 is not correct. It should be XLR 231 unless you fix it in the schematics and rewire in PCB layout.

 

[Khron]

To whom it may concern:

ORS87-Plus updated to v1.1 - added the "calibration" input & associated 560 ohm resistor; PCBway project, Mouser / Digikey BOMs and toner-transfer / DIY-etch files updated accordingly

 

[AdamFrisen]

Here is the schematics and PCB for U87AI Hartley oscillator DC voltage for the capsule.

KiCad 8.

I used 220uH shielded transformers (oversized). The footprint is in the zip file.

Generally, I think for audio frequencies, ground plane might do more harm than good as ground return for signals are spread out in the plane. But it is so easy to pour ground and test the PCBs how it gets for $7-9 was cheap enough for me.

But now you have the schematics so anyone interested can experiment with the layout.

 

[Khron]

Here is the schematics and PCB for U87AI Hartley oscillator DC voltage for the capsule.

Why bother adding the two diodes, two resistors and four capacitors used only in the U89? Arguably pointless use/waste of board space & components...

 

[z11111]

To whom it may concern:

ORS87-Plus updated to v1.1 - added the "calibration" input & associated 560 ohm resistor; PCBway project, Mouser / Digikey BOMs and toner-transfer / DIY-etch files updated accordingly

Hey Khron, do you know if your board will fit a Sterling ST51? I’ve had one laying around that I got for $free.99 about 10+ years ago but never, ever use it (was my talkback mic for a while lol…)

 

[Khron]

Hey Khron, do you know if your board will fit a Sterling ST51? I’ve had one laying around that I got for $free.99 about 10+ years ago but never, ever use it (was my talkback mic for a while lol…)

It's a possibility - what are the distances between the centers of the mounting holes, in both directions?

 

[untamedfrontier]

Was wondering if someone could help clear up something for me, as I'm trying to figure out if I'm doing this right.

I'm using homero's 1.1 board, but I'm trying to use a UTM transformer and am trying to make sure that I'm doing everything right.

As this is a center tapped transformer, I'm assuming the correct action is to:
* Yellow (center tap) wire goes to TXC
* Jumper wire on J1-J2

Would I completely omit R18 & R19 in this case, put a jumper, or keep R18 & R19 installed?

I searched prior posts where it referenced center tapped transformers, but I'm confused. I currently have the center tap disconnected, and R18 & R19 installed, but I'm not quite yet happy with this build, and am trying to debug a few things, as it's much noisier than my prior 2 builds. The most drastic changes in this current build are Transformer and Capsule, both of which I expected to perform much better than what I used in the first 2 builds.

Relevant part of schematic:

 

[Khron]

Would I completely omit R18 & R19 in this case, put a jumper,

I'm not the one who should be answering this question (and I can't for the life of me see the point of even including that jumper), but yes - jumper connected, abd R18/19 not installed.

as it's much noisier than my prior 2 builds

Only the noise is increased, but not the signal? You 100000% sure the primary & secondary haven't been swapped around, if the transformer is also a new element?

 

[z11111]

It's a possibility - what are the distances between the centers of the mounting holes, in both directions?

mounting holes are roughly 25.5mm center-to-center wide, then 34.5mm center-to-center tall / long.

stock board is roughly 32.2mm wide x 56.5mm tall. roughly 11mm to 12mm ish of clearance from the top of the PCB to the headbasket / capsule chamber, not so much on the bottom of the board.

 

[untamedfrontier]

I'm not the one who should be answering this question (and I can't for the life of me see the point of even including that jumper), but yes - jumper connected, abd R18/19 not installed.



Only the noise is increased, but not the signal? You 100000% sure the primary & secondary haven't been swapped around, if the transformer is also a new element?

Thanks for taking the time to help, Khron. I'm probably not 100000% sure, but I did measure the UTM 0547f transformer on my DMM, Red & black had a value consistent with what I believe the primary should have (I just remeasured, and it's installed, but at 368 ohms), and blue/green had a much smaller level of resistance (~25 ohms).

I did just install the yellow wire into TXC, installed jumper for J1/J2 and removed R18/R19, and now I have no signal. Reinstalled jumpers into R18/R19 to see if that was the culprit, and still no signal.

Removed yellow wire for TXC, removed jumper on J1/J2, reinstalled 2k2 resistors in R18/R19, and am back to where I was before, which is fairly low output, and probably just as a result of having to crank the preamps to get a useable signal, more noise.

I've got another Neutrik NTE 10/3 in the mail, so I might just do a full fresh board with the modern BOM, and see if the different transformer has an impact. Based on how the yellow wire for TXC is performing, I may just have some sort of a dud when it comes to the transformer.

 

[Khron]

Thanks for taking the time to help, Khron. I'm probably not 100000% sure, but I did measure the UTM 0547f transformer on my DMM, Red & black had a value consistent with what I believe the primary should have (I just remeasured, and it's installed, but at 368 ohms), and blue/green had a much smaller level of resistance (~25 ohms).

I did just install the yellow wire into TXC

Have you also measured between blue and yellow, and/or green and yellow, to confirm that is indeed the center-tap for the secondary? Assumption is the mother of all f**k-ups...

While there's no wire colour description for the 0547f, the 0549 (similar "enough") does specify that

Red-Blk – Primary (Tube side)
Blu-Grn – Secondary (XLR side)
Yel – Shield

Soooo... no wonder you get no signal with the yellow wire soldered to TXC, and the resistors removed (no power -> no signal). Because that may very well NOT be the center-tap of the secondary, but the electrostatic shield between primary and secondary, which you'd need to connect to ground. That being said, wonder if that may affect noise in any way...

Photos of your build? And what sort of noise are you encountering? Is that with the mic body fully assembled?

 

[AdamFrisen]

Confirmed by the manufacturer, the yellow wire is the internal shield for UTM0547x, UTM0587.
I used the ground for R* as I don't use it.



Also could be good to check the wiring of the transformer itself so you get right phase. Some are not wired according to the documentation.

https://groupdiy.com/threads/open-source-diy-mic-project-ors-87-stripped-down-u87.86814/post-1162471

 

[untamedfrontier]

Have you also measured between blue and yellow, and/or green and yellow, to confirm that is indeed the center-tap for the secondary? Assumption is the mother of all f**k-ups...

While there's no wire colour description for the 0547f, the 0549 (similar "enough") does specify that

Soooo... no wonder you get no signal with the yellow wire soldered to TXC, and the resistors removed (no power -> no signal). Because that may very well NOT be the center-tap of the secondary, but the electrostatic shield between primary and secondary, which you'd need to connect to ground. That being said, wonder if that may affect noise in any way...

Photos of your build? And what sort of noise are you encountering? Is that with the mic body fully assembled?

Yeah, I believe it's the same wiring as the 0549, as I can see the yellow wire is going to both sides of the transformer, so I definitely misunderstood the 5th wire's purpose.

So the noise I'm seeing is not ground hum, it's probably most accurately described as noise floor/hiss because I have to nearly max out the preamp to get a nearly usable signal (which it's still not because of the noise floor). I am attempting to use a flat 47 capsule in this, and as such, just omitted the low frequency filter C5 (which in your ORS plus board would be C12). I did leave in the 10k R9 (which would be R15 on ORS-plus) because I added in the 560 ohm resistor for the measuring input (which I've done the same on 2 previous builds and confirmed it worked as expected).

This poor board is quickly becoming the PCB of theseus based on how many components I've swapped so far, but I'll get there eventually I'm sure. I'm definitely not thrilled with my current solution for grounding the body, but once it sounds good, I won't care how ugly it looks

May also consider a different circuit for this capsule, but hoping to get something that sounds at least as good as the first 2 87 builds.

 

[untamedfrontier]

Confirmed by the manufacturer, the yellow wire is the internal shield for UTM0547x, UTM0587.
I used the ground for R* as I don't use it.

Also could be good to check the wiring of the transformer itself so you get right phase. Some are not wired according to the documentation.

https://groupdiy.com/threads/open-source-diy-mic-project-ors-87-stripped-down-u87.86814/post-1162471

Thanks for this, and good idea on using R*, I might do that as well. I think my wiring is also flipped, when compared with a good build, it was definitely phase flipped (which I can solve in my DAW, of course, but who wants to do that every time?)

 

[Khron]

just omitted the low frequency filter C5 (which in your ORS plus board would be C12).

Sooooooooo one end of your capsule is floating, then? No wonder you barely get any signal but lots of noise...

 

[untamedfrontier]

Sooooooooo one end of your capsule is floating, then? No wonder you barely get any signal but lots of noise...

Oh, on the contrary, I actually have a lot of wonder about this, but (quite obviously) zero electrical engineering background to provide an answer to that or see why that would be obvious. I thought all the discussions related the the 847'ing of this circuit and comparing the schematics pointed towards making that change, but apparently not

I suppose it's back to trial and error time with socketing in some different values in C5. Thanks for pointing out something to try.

 

[Khron]

I suppose it's back to trial and error time with socketing in some different values in C5. Thanks for pointing out something to try.

Actually, i might've been talking out of my a*s just there - not populating that capacitor shouldn't affect much, or anything. (as in, assuming all is well, things should still function). Might nullify the high-frequency attenuation as well, but I can't comment more on that.

All those sockets, while handy for testing, are a liability. How certain are you that the JFET is still 100% ok? And that the trimpot is making reliable contact (it seems to be socketed too)..?

 

[untamedfrontier]

Actually, i might've been talking out of my a*s just there - not populating that capacitor shouldn't affect much, or anything. (as in, assuming all is well, things should still function). Might nullify the high-frequency attenuation as well, but I can't comment more on that.

All those sockets, while handy for testing, are a liability. How certain are you that the JFET is still 100% ok? And that the trimpot is making reliable contact (it seems to be socketed too)..?

I gave it a go and popped in a 220n film cap (as an attempt to compensate for the low end de-emphasis, I remember seeing earlier this cap can go up to 330n), and the signal is full and it sounds nice.

Once I make some decisions on component values, I'll toss some solder onto the sockets to lower their chance of failure.

I've got a few more values of the HF cap coming tomorrow, so I might tweak that now because I think the 100p cap there is a touch too harshly bright.

So all in all, thanks for the assist

 

[Khron]

I gave it a go and popped in a 220n film cap (as an attempt to compensate for the low end de-emphasis, I remember seeing earlier this cap can go up to 330n), and the signal is full and it sounds nice.

There ya go, then!

I'm all for modding and trying out things, but I'm not a big fan of the "i couldn't leave good enough alone, now help me" shtick