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

[Wordsushi]

This project needs a name. I would like to suggest calling it the “ORS 87” or “ORS Stripped 87” as a nod to @OneRoomStudio for bringing us this killer iteration of this circuit.

Also, what are thoughts on declaring this an “open-source” project? It basically already is, but I believe formally labeling it as such (for search) could help inspire more tinkerers and DIY-minded folks to join the conversation and potentially expand the exploration of mods and options (and crazy ideas) using the ORS 87 as the foundation. I don't think I'd mind seeing someone develop this as a kit to expand the potential user base.

Maybe it could be used to help support GDIY's operating fund by suggesting a nominal donation to the site if you build the ORS 87?

 

[LevinGuitar]

Good to hear. It’s been my experience that most 2N3819’s can be used, as long as the source resistor is adjusted for the individual FET. Of the dozens I’ve tried in mic circuits over the years, I only had one actual “bad” one.

I had a lot of them giving me same voltage in source and drain no matter what source resistor adjustment made up to 10k. Also had some with the source resistor needed to be higher than 10k to give me 10v on drain, but otherwise, sounding good.

 

[OneRoomStudio]

This project needs a name. I would like to suggest calling it the “ORS 87” or “ORS Stripped 87” as a nod to @OneRoomStudio for bringing us this killer iteration of this circuit.

Also, what are thoughts on declaring this an “open-source” project? It basically already is, but I believe formally labeling it as such (for search) could help inspire more tinkerers and DIY-minded folks to join the conversation and potentially expand the exploration of mods and options (and crazy ideas) using the ORS 87 as the foundation. I don't think I'd mind seeing someone develop this as a kit to expand the potential user base.

Maybe it could be used to help support GDIY's operating fund by suggesting a nominal donation to the site if you build the ORS 87?

I appreciate the support, but all I did was delete some parts from the original U87i circuit. No need to name anything after me! Feel free to call it "open source" though, I always figured it was.

 

[OneRoomStudio]

I had a lot of them giving me same voltage in source and drain no matter what source resistor adjustment made up to 10k. Also had some with the source resistor needed to be higher than 10k to give me 10v on drain, but otherwise, sounding good.

With a 24V zener, an "average" 2N3819 (according to the data sheet) should require an 11K source resistor to put the drain at ½ supply.

 

[Wordsushi]

I appreciate the support, but all I did was delete some parts from the original U87i circuit. No need to name anything after me! Feel free to call it "open source" though, I always figured it was.

Well, unfortunately, that's not an acceptable refusal. Think of it like this, Lou Gehrig didn't invent Lou Gehrig's disease, ALS is actually only referred as such because he put it on the map. You, my savvy mic sherpa, took the initiative to make the changes and put this stripped down 87 project on the map regardless of it's provenance, so I decree from this day forth, this project will be referred to by me as the ORS 87.
And today...today... today.... we are the luckiest men... men... men... in the world... world... world.. (ok, that was a Lou Gehrig joke).

 

[Wordsushi]

Quick soundcheck of the 33v Zener build. Solid.
View attachment ORS87-33vZenerBuild-Test1b.wav

 

[Barry Hufker]

I understand your excitement about the project. There is much to like. However, I encourage you now to think about another capsule. Forgive me for saying but I'm sure you want an honest assessment -- there is a sibilant and very unpleasant quality to the sound that I think a better capsule would eliminate.

 

[Wordsushi]

I understand your excitement about the project. There is much to like. However, I encourage you now to think about another capsule. Forgive me for saying but I'm sure you want an honest assessment -- there is a sibilant and very unpleasant quality to the sound that I think a better capsule would eliminate.

Absolutely. Yes, please always on an honest assessment, and you're absolutely right that I need to swap out this cheapie test capsule for something more legitimate. I've just temporarily run out of better capsules and am at the moment stuck with my sad box of crap capsules pulled from cheap mics I've experimented upon over the years until I get off my lazy butt and order something better.

 

[micolas]

I did something like this about 2-3 years ago.
I reduced the U87, only Cardioid, no attenuation pad, no hi-pass, only recycled parts, no pcb, point to point.
Because I was very curious to study the influence of the polarization voltage of the capsule on the character of the sound, I made a minimal dc/dc converter, without zenner. I tested from a few volts to almost 100v. Stock capsule. The results were beyond expectations.

Thread 'U87ai clone from recycled parts' https://groupdiy.com/threads/u87ai-clone-from-recycled-parts.86367/

 

[ubxf]

Could you share your conclusions?

 

[Gus]

Here you go. Just a simplified U87i.

View attachment 124983

One note - I just simulated it, and due to the lack of the rear capsule components (including C1/C3), C6 has a little more pronounced of an effect on the top end. Changing it to around 120pF instead of the stock 220pF seems to match the original curve fairly well. Feel free to experiment.

IMO you can do better look at other 87 variants think and mix fragments. I mean this in a good way.

I sim circuits then build it then test and adjust things for the capsule used
I have been building designing single pattern only circuits for k67/87 variants for some time I don't sell any and I will not post them here.

IMO this type of circuit for the K67/87 capsule with one pattern sounds better than the 3 pattern circuits
kingkorg posted about multi vs single pattern microphones.

 

[Wordsushi]

IMO you can do better look at other 87 variants think and mix fragments. I mean this in a good way.

I sim circuits then build it then test and adjust things for the capsule used
I have been building designing single pattern only circuits for k67/87 variants for some time I don't sell any and I will not post them here.

IMO this type of circuit for the K67/87 capsule with one pattern sounds better than the 3 pattern circuits
kingkorg posted about multi vs single pattern microphones.

Suggestions welcome if you have some.

 

[OneRoomStudio]

If you were to leave behind the "classic" values/parts, there are a few things you could do. Off the top of my head:

-Use a 36V zener and change R16 to 10K. Why not? More headroom!
-Use a different JFET - LSK189B looks promising (very low noise)
-With that setup, the source resistor should be around 4.7K

Output impedance might be a bit off though.

 

[soliloqueen]

If you were to leave behind the "classic" values/parts, there are a few things you could do. Off the top of my head:

-Use a 36V zener and change R16 to 10K. Why not? More headroom!
-Use a different JFET - LSK189B looks promising (very low noise)
-With that setup, the source resistor should be around 4.7K

Output impedance might be a bit off though.

the major advantage of linear systems' jfets here is that they all have power supply specs fit for this circuit, so you only need to sort for idss (i believe?). this can be huge time saver, but i am not sure if it offsets the increased cost.

for others: the linear systems jfets are process 50 adjacent new production jfets. originally they were for maintaining part stock to repair medical and scientific equipment but they have found niche use in re-implementations of vintage audio circuits. the LS846 is a drop in replacement for the 3819 with better breakdown and the LSK189 is similar, but ultra low noise. both are great choices for a u87 headamp but ridiculously expensive.

 

[OneRoomStudio]

both are great choices for a u87 headamp but ridiculously expensive.

Expensive for a transistor, but cheap compared to a tube. $7 ain’t too bad:
https://www.mouser.com/ProductDetai...9-TO-92-3L-BK?qs=T%2BzbugeAwjjUrS49wFak%2Bw==

 

[OneRoomStudio]

This simulates pretty nicely, and according to my calculation, should have an output impedance of ~18K, which is right on the money (would give you around 200 ohms output with a 9.5:1 "87-style" output transformer). It pulls about twice the current of the original circuit, but we're still talking about 0.8mA, so no big concern there.

 

[Khron]

but cheap compared to a tube.

Depends on the tube...

 

[Wordsushi]

This simulates pretty nicely, and according to my calculation, should have an output impedance of ~18K, which is right on the money (would give you around 200 ohms output with a 9.5:1 "87-style" output transformer). It pulls about twice the current of the original circuit, but we're still talking about 0.8mA, so no big concern there.

View attachment 126682

I'd pay $7 for a FET if it made a noticeable difference. So, would a swap to an LSK198b necessitate the change to a 36v zener like you have in the schemo or would it just give it a bunch more headroom, while also gaining better self noise with the 198b?

 

[OneRoomStudio]

I'd pay $7 for a FET if it made a noticeable difference. So, would a swap to an LSK198b necessitate the change to a 36v zener like you have in the schemo or would it just give it a bunch more headroom, while also gaining better self noise with the 198b?

Technically, you could just stick the LSK189 in your circuit with the 33V zener and 47K drain resistor, and bias it up (source resistor would have to be around 5K), but the output impedance of the FET in that configuration is high - around 46K. With a 9.5:1 transformer, the mic would have an output impedance of over 500 ohms. Probably not too big a deal, especially since many modern preamps have higher input impedance, but some older transformer-coupled preamps are expecting to see mics with 150/200 ohm output impedance, so you might be loading the FET a bit.

Will that matter since there's already high frequency dampening due to the NFB deemphasis? I don't know and don't want to bother modeling it, haha.

The LSK189 would be quieter than a 2N3819, and it's generally a better behaved FET, but would it make a big audible difference? Not sure.

**If you do this, make sure you note the pin-out! LSK189's are the reverse pinout of 2N3819's, so you've have to install it "backwards." The gate is still in the middle though.

 

[Wordsushi]

Does anyone have any particular components they like for this build?
Is there any specific part, or brand for this BOM you'd recommend based upon it's quality, or price? Or both?

Here, for reference is the ORS 87 BOM, as originally posted by @OneRoomStudio .

R7: 1G
R8: 1M
R9: 7k3
R10: 25K Trimmer
R12: 47k
R13: 330k
R14: 10k
R15: 300k
R16: 56k (for 24V zener, try 47k for 33V zener)
R18 & R19: 2k2 (closely matched!)

C5: 33n (can go as big as 330n if you want more low-end)
C6: 220p (treble cap - smaller = more treble, bigger = darker)
C7: 1u (can tweak to taste. smaller = less bass)
C8: 20u 6v
C9: 470n
C10: 4u7 35V
C11: 4u7 63V
C12: 10u 63V
C13: 10n

Q1: 2N3819

GR1: 1N4749A (24V zener - can try 33V instead with smaller R16)

I'm currently thinking there are 3 ways to approach this build.

1. Use whatever you have on hand.
Make a mic out of scraps and whatever bits and pieces you have laying around, augment with a few specific parts you have to purchase. This probably applies to a lot of folks here who have a collection of spare parts.

2. Find the cheapest way
Purchase everything with the lowest amount of spend possible. I think you could easily put one together with under $100 in total cost, but don't ask me for any specifics on that yet.

3. Go big.
Premium parts. Good capsule. Recognized transformer. See how good you can make it sound.