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.
Just finished a build on Homero's ORS 87 PCB with what I believe is Khron's BOM and an LSK-189 (Neutrik transformer, 32mm trash cap pulled from an MXL) and it is definitely quieter than my other builds, so I think Khron is on to something with all the extra filtering in his component choices. Even though they're about $6 apiece, the LSK 189 feels like an upgrade worth the extra $$.
IMG_9905.jpg
 
Just finished a build on Homero's ORS 87 PCB with what I believe is Khron's BOM and an LSK-189 (Neutrik transformer, 32mm trash cap pulled from an MXL) and it is definitely quieter than my other builds, so I think Khron is on to something with all the extra filtering in his component choices. Even though they're about $6 apiece, the LSK 189 feels like an upgrade worth the extra $$.
View attachment 127960
I see you used a cheap, (smaller?) Ali body.What capsule did you use?
Did you buy the LSK189 from Mouser? I guess you bought several pieces to select and experiment.
What Idss and Vgs(off) measure in the ones from this lot and how much does the one used in ORS-87 have? How much did you BIAS jFET?
Other than the noise being lower as expected, how does the sound, tonal balance, compare to the 2N3819 versions?
 
Last edited:
Just finished a build on Homero's ORS 87 PCB with what I believe is Khron's BOM and an LSK-189 (Neutrik transformer, 32mm trash cap pulled from an MXL) and it is definitely quieter than my other builds, so I think Khron is on to something with all the extra filtering in his component choices. Even though they're about $6 apiece, the LSK 189 feels like an upgrade worth the extra $$.
Hello Marc. very nice the 1G resistor that goes together with Gain of the fet plug directly on signal. could you tell me if the microphone quiet like a Neumann microphone or just quiet :))? i'm also interested on how much volt should you have on jet fet bias voltage then i could work it out with my hears. Thank you! and congrats to you all!
 
I see you used a cheap, (smaller?) Ali body.What capsule did you use?
Did you buy the LSK189 from Mouser? I guess you bought several pieces to select and experiment.
What Idss and Vgs(off) measure in the ones from this lot and how much does the one used in ORS-87 have? How much did you BIAS jFET?
Other than the noise being lower as expected, how does the sound, tonal balance, compare to the 2N3819 versions?
Capsule is generic 32mm k67 pulled from a CAD or MXL mic years ago. LSK-189 was just pulled from the bag and installed, not measured beforehand. I bought 2 just to try. I biased it to 14v. Too soon to tell how it compares with the 2n38a9 builds with the original vintage 87-style BOM yet. Haven't really done more than a cursory A/B test. Capsules are different from my other builds so I'm reserving final judgement until I build a couple more, but the lower self-noise jumped out at me right away and it's not like the previous builds were noisy at all either.
 
Hello Marc. very nice the 1G resistor that goes together with Gain of the fet plug directly on signal. could you tell me if the microphone quiet like a Neumann microphone or just quiet :))? i'm also interested on how much volt should you have on jet fet bias voltage then i could work it out with my hears. Thank you! and congrats to you all!
Thanks! I just originally dialed in 14v for the FET bias and I like it there so far, but as I go through my process of examining the results that may change. It does seem to have very low self-noise, even better than my u87ai, which surprised me. I want to build one with Homero's board with a 2n3819 to see if there is a noticeable difference.
 
Quick comparison in the booth. Below, Green is u87ai, White is Homero ORS 87 build (with Neutrik NTE 10/1 and cheap MXL donor capsule). Both normalized to -23lufs.
I think a better capsule would give it more at 3k.
Self noise samples included. ORS 87 is about 1.1db quieter. I left the vent fan on in my booth to give a clearer reference.

The 87 is picking up more sub 70hz rumble even with 80hz low cut filter applied at the preamp. I initially thought the ORS87 felt low end light, but seeing this graph dissuades me from adding more low end on it.

View attachment U87ai v Homero ORS 87.wav

Screen Shot 2024-04-28 at 9.51.24 AM.png
 
Quick comparison in the booth. Below, Green is u87ai, White is Homero ORS 87 build (with Neutrik NTE 10/1 and cheap MXL donor capsule). Both normalized to -23lufs.
I think a better capsule would give it more at 3k.
Self noise samples included. ORS 87 is about 1.1db quieter. I left the vent fan on in my booth to give a clearer reference.

The 87 is picking up more sub 70hz rumble even with 80hz low cut filter applied at the preamp. I initially thought the ORS87 felt low end light, but seeing this graph dissuades me from adding more low end on it.

View attachment 127986

View attachment 127987
Great example of the issue with these MXL capsules. The problem is not excess high end, which can be fixed with EQ, but the resonance that creates huge null befor the HF boost. It's resonance coming from 180°.
 

Attachments

  • Screenshot_20240428_210624.jpg
    Screenshot_20240428_210624.jpg
    373.3 KB · Views: 1
Great example of the issue with these MXL capsules. The problem is not excess high end, which can be fixed with EQ, but the resonance that creates huge null befor the HF boost. It's resonance coming from 180°.
chamber issue maybe? i wonder if this could be fixed. It's been years since I've taken one of these apart.
 
Since everyone is so keen on PCBs, and because I wanted to be able to build a few of these for friends, I laid out a simple board to fit an HL-95 body and use "normal" parts. Nothing too special here, but it sounds great...about as close sounding to a U87 as anything I've ever built.

View attachment 128000
Sweeeeet!!!! Very nice.
That looks fantastic. I also got some boards done for myself. But I did not make them slick enough to fit the transformer.

IMG_9769.jpg
 
Is that a T-13 transformer from AliExpress?
Do you like it?
Yes, one of the $18 ones from Ali.
I don't hate it. I don't love it. It seems less colored than the 3u GZT 87 or the Neutrik 10/3 and I can't tell if it's a bit low end light or not. I have two in separate builds and one I really like better than the other, so I'm still trying to determine what other factors are involved (probably capsule) or if it could be inconsistency??? Lately, I've been spending more time listening to the Neutrik since it's much easier and quicker to source and cheaper through Mouser.
Maybe ask me again in a couple of weeks.
 
So, this one really surprised me. CAD GXL 2200 body, which I had laying around and only used it because I ran out of "better" bodies. Among my parts, I found a WGT Center-purchased K87, which I had tried with a Fuchs build a few years ago and didn't like at all. Transformer is Neutrik NTE 10/3, which I really like... All with Homero's board with Khron's BOM and an 2n3819 from my small stash.
When I plugged it in I couldn't believe how close it is to my u87ai.
It's a two-sided capsule. Maybe I only had previously heard the other side. The capsule lead for the side I'm using looked like it hadn't been soldered before.
I know WGT Center's website is gone, but these are AYM caps, right? Maybe I'll try the single-sided one JLI sells.

IMG_9906.jpg
View attachment 87ai vs ORS 87 CAD Body - WGT 87 Neutrik 10:3.wavCan you tell which one is the real 87?
 
So, this one really surprised me. CAD GXL 2200 body, which I had laying around and only used it because I ran out of "better" bodies. Among my parts, I found a WGT Center-purchased K87, which I had tried with a Fuchs build a few years ago and didn't like at all. Transformer is Neutrik NTE 10/3, which I really like... All with Homero's board with Khron's BOM and an 2n3819 from my small stash.
When I plugged it in I couldn't believe how close it is to my u87ai.
It's a two-sided capsule. Maybe I only had previously heard the other side. The capsule lead for the side I'm using looked like it hadn't been soldered before.
I know WGT Center's website is gone, but these are AYM caps, right? Maybe I'll try the single-sided one JLI sells.

View attachment 128040
View attachment 128041Can you tell which one is the real 87?
I like the sound of the first one. The second one has quite a bit more high end content and the first one sounds warmer. Is the first the real 87?
 
So, this one really surprised me. CAD GXL 2200 body, which I had laying around and only used it because I ran out of "better" bodies. Among my parts, I found a WGT Center-purchased K87, which I had tried with a Fuchs build a few years ago and didn't like at all. Transformer is Neutrik NTE 10/3, which I really like... All with Homero's board with Khron's BOM and an 2n3819 from my small stash.
When I plugged it in I couldn't believe how close it is to my u87ai.
It's a two-sided capsule. Maybe I only had previously heard the other side. The capsule lead for the side I'm using looked like it hadn't been soldered before.
I know WGT Center's website is gone, but these are AYM caps, right? Maybe I'll try the single-sided one JLI sells.

View attachment 128040
View attachment 128041Can you tell which one is the real 87?
The second has a slightly larger volume, they must be matched in daw. It's more sibilant, increasing the de-emphasis capacitor a bit will make it sound better. And the lower middles are poorer.
The first sounds rounder, more balanced, more natural, so it seems to be Neumann.
Was the recording done simultaneously?
How much was the BIAS of the jFET (U drain)?
Was the noise slightly higher with the 2N3819 than with the LSK189?
 
I found a WGT Center-purchased K87, which I had tried with a Fuchs build a few years ago and didn't like at all.
It's a two-sided capsule. Maybe I only had previously heard the other side. The capsule lead for the side I'm using looked like it hadn't been soldered before.
Maybe I'll try the single-sided one JLI sells.
A personal note:
✨Even in a Cardioid-only build, I like to use two-sided capsules.
You can connect the back diaphragm to GND, to the backplate, you can leave it free or you can apply a small polarization voltage (positive or negative), to get super-cardioid, hyper-cardioid, subcardioid,a more accurate cardioid, etc., for a natural sound, better rear rejection or wide sound dimension.
You can fine tune the sound to your taste.
✨In addition, you can choose your favorite front side right from the start.
✨When it gets dirty, damaged, etc., you have a quick backup, you connect the back-side.
Just my 2 cents.😁
 
Last edited:
It's more sibilant, increasing the de-emphasis capacitor a bit will make it sound better.
I agree that it’s more sibilant, but the deemphasis cap won’t help in this case, since it’s just a low pass filter. Usually, sibilance issues are lower in frequency (closer to 3-7kHz), and you’d have to filter out way too much top end to affect frequencies that low. The treble deemphasis cap is really only useful in taming the rise above 10kHz that is inherent to the K67/87 capsule.

The only way to make these mics less sibilant (without making the mic incredibly dark) is to use a different/better capsule. These are simple circuits, and most of what we’re hearing is the capsule anyway.
 
I agree that it’s more sibilant, but the deemphasis cap won’t help in this case, since it’s just a low pass filter. Usually, sibilance issues are lower in frequency (closer to 3-7kHz), and you’d have to filter out way too much top end to affect frequencies that low. The treble deemphasis cap is really only useful in taming the rise above 10kHz that is inherent to the K67/87 capsule.

The only way to make these mics less sibilant (without making the mic incredibly dark) is to use a different/better capsule. These are simple circuits, and most of what we’re hearing is the capsule anyway.
You are perfectly right.
I have not fully described the frequency response.
It is more sibilant (so more mid-highs) and in addition it has more highs (which bother me) specific to low quality Chinese capsules.
A larger de-emphasis capacitor will not reduce the sibilance, which depending on the voice appears in the mid-high range, but it will reduce the exaggerated highs
and it will sound a bit better(for my ears)
In this particular case, you de-ess or use a better capsule.You can also look for a tilted/rotated position and different distance from the microphone to further reduce the sibilance.You use another technique to speak into the microphone, a field in which @Wordsushi is a professional
 
Last edited:
Back
Top