KH and SPA u87 etc adjustments

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Gus

Well-known member
Joined
Jun 3, 2004
Messages
5,249
Location
n
Has anyone traced what KH and SPA change in the u87 circuit and other microphone circuits? I am not asking about the capsules.

The reason I ask is because I have not had a chance to look at the insides of one and I am curious what the adjustments are about.
What pictures I can find on the web seem to show tants and electrolyic caps changed to film or different electrolytic cap and the transformer sometimes.
The JFET might be changed and the version with the subboard changed to a JFET and missing parts added to the main board that are missing with the sub board and of course rebiased
And maybe the EQ is adjusted.

So what is the big deal about the circuit changes?
 
I think the big part of what KH did was tuning the mics for each customer to meet their expectations. Which is the best way to go when it comes to mic mods imho.

Sometimes slight change in feedback cap value in u87, or removal, sometimes the whole capsule. Sometimes the whole circuit, so that Whitney Houston could have elam tone in the pocket using phantom power.
Over the time fairytale aspect of it got blown out of proportions. I believe he had a lot to do with that, because capacitor type hype might serve as a cover for simple cap value change.

EVH did such tricks all the time in order to prevent people from copying him. 
 
Imagine a world where, for the most part, music was recorded in very expensive studios, where expensive microphones mostly resided. Where most people would never have an opportunity to even step into such a place, let alone allowed to even touch said microphones. Where those that did have permission, had the audacity to have a peak inside. Where eventually some people were able to buy their own mics from old broadcast outfits, giving them an opportunity to learn how to repair them themselves. Where some that did, tried experimenting to improve things and were blasted for committing such atrocities. Where after 20-25 years of contributing to the improvement of the industry, the internet comes along and everyone is an expert. Where 40-50 years after you started your work (where it was the dark-ages if you weren’t part of the 2 big European manufacturers), most people don’t respect what you contributed and take it all for granted.
 
I would like to expand on the first post. Also note I am not posting about plate/anode, coupling cap, transformer out tube microphones.

I have worked on a few older 87s and most of the "sound" followed the heads. Worked on a batch of 6 one time from the 70s.
One set of electronics maybe sounded "better" however it was related to the JFET parameters and biasing.

So I get the change of capsules being an adjustment not so much on the u87 circuit

I also understand the EQ change if that is something that was done. I will adjust the network for different capsules however as others have noted at this forum the adjustment is HPF and LPF and can only do so much.

Something that I have not noticed much off is addressing the capsule environment however, to do that in a scientific manner you will need an anechoic chamber(higher frequency as a minimum) or software and good hardware to measure the changes and that costs money and takes up space. Also polar charts vs frequency should be done.

Look for MS vienna's VM1 thread and look at the KHE picture you can find at recording hacks.
Or even look for KHs U67 reissue pictures and note the capsule mount change
 
Gus said:
Something that I have not noticed much off is addressing the capsule environment however, to do that in a scientific manner you will need an anechoic chamber(higher frequency as a minimum) or software and good hardware to measure the changes and that costs money and takes up space. Also polar charts vs frequency should be done.

You would be surprised how much can be done by lowering sample window to 10ms, or even 1ms with REW in a regular vocal booth using good, flat CARDIOID reference mic. ;)
 
I've never looked inside a KH U87 but, besides the changes you mentioned, he himself has said that feedback is (sometimes?) removed. 
Also, when talking about the older U87i (non A), the capsule to gate capacitor is removed, with the gate resistor value changed to accommodate.  I assume it's halved in value from 1G but?

 
I don't think it is a problem to remove the capsule coupling capacitor and one of the 1 G.ohm resistors.
(No idea how much this will improve the microphone...)
 

Attachments

  • U87mod.PNG
    U87mod.PNG
    38.4 KB · Views: 48
RuudNL said:
I don't think it is a problem to remove the capsule coupling capacitor and one of the 1 G.ohm resistors.
(No idea how much this will improve the microphone...)

Yep, I don't think there'd be a problem either.  Like you, I also don't know what improvement there'd be.

My only related reference is with a new U87Ai of mine.  I thought that there was a slight difference in the front and back capsules, with the front side being slightly better.  I did wonder if the additional cap connecting the rear side was part of the equation so I switched the connections inside the headbasket to see if I could tell anything at all.
When in cardioid mode,  the rear side now being the source,  I then thought that side was now slightly better. 

A very non-scientific test, and I may easily have been fooling myself  8) 
 
Be careful if you are checking this with headphones on.
Because in the figure-8 position, the polarity ('phase') of the rear capsule is opposite the polarity of the front capsule.
And what you hear is a  combination of the sound 'in your head' and the sound coming out of the headphones.
This may give misleading results.
 
RuudNL said:
Be careful if you are checking this with headphones on.
Because in the figure-8 position, the polarity ('phase') of the rear capsule is opposite the polarity of the front capsule.
And what you hear is a  combination of the sound 'in your head' and the sound coming out of the headphones.
This may give misleading results.

Yep.  I was listening on headphones and it was far from scientific.

When listening to both capsules, one after the other, I used Omni.  I also checked cardioid and figure eight but knowing the polarity might be swaying my judgement. To combat that I had to take the head basket off, switch connections inside etc. so there was a time gap and my memory was only approximate.

It's a U87Ai so I can't try direct connection of the capsule to J-Fet or I would.  So who knows?  :D 
 
A test in omni position would indeed be a more 'honest' comparison.
I have the impression that the influence of a coupling capacitor is in general overrated.
(The same as 'the sound of capacitors'. Maybe it has to do with my ears, but I never heard much difference...)
 
RuudNL said:
I have the impression that the influence of a coupling capacitor is in general overrated.
(The same as 'the sound of capacitors'. Maybe it has to do with my ears, but I never heard much difference...)

You've played with WAY more mics than I have, but applying my experience in other equipment, I agree.  The caveat being that they are good capacitors in the first place.  I find that a tantalum in the signal path will be evident, and easily measurable.


Anyway:  On U87 type mics, I think a mod worthy of trying for DIY'ers might be to use an expernal 60V supply.  Use that to feed the backplate(s) and two separate amplifiers fed from each capsule.  There's room on the inside of a body for that.

6 pin XLR on the bottom and you have two identical (hopefully! )  signal paths and the ability to have whatever pattern you wish. 

Just an idea.     
 
I've looked at Klaus' U87 mod from a distance. Never had the opportunity to check one out myself. But I have learned:

1. changes FET and re biases to get a bit more gain
2. changes transformer to a custom made AMI
3. changes output coupling cap (I'm guessing polystyrene)
4. chooses best sounding side of the capsule for the front side. He's quite adamant about this.
5. Disconnects the pad and lo cut switches and repurposes one to switch the rear diaphragm off for "true cardioid" a la U47. Gains sensitivity and output in single diaphragm mode.
6. makes changes to feedback. I haven't been able to find any info on exactly what is done.
7. I have also seen some discussion of "capsule adjustments" as discussed here.

All I ever heard about SPA when I working in LA was the thin diaphragm capsule mods, though it makes total sense they would do electronic mods as well.
 
Just a remark in relation with the 'audible' difference between capacitors:
Yesterday someone mentioned that he could hear the difference between types of solder...
IMHO that must be an extreme 'cork sniffer'...  ;D
 
AusTex64 said:
I've looked at Klaus' U87 mod from a distance. Never had the opportunity to check one out myself. But I have learned:

1. changes FET and re biases to get a bit more gain
2. changes transformer to a custom made AMI
3. changes output coupling cap (I'm guessing polystyrene)
4. chooses best sounding side of the capsule for the front side. He's quite adamant about this.
5. Disconnects the pad and lo cut switches and repurposes one to switch the rear diaphragm off for "true cardioid" a la U47. Gains sensitivity and output in single diaphragm mode.
6. makes changes to feedback. I haven't been able to find any info on exactly what is done.
7. I have also seen some discussion of "capsule adjustments" as discussed here.

All I ever heard about SPA when I working in LA was the thin diaphragm capsule mods, though it makes total sense they would do electronic mods as well.

Assuming what you posted is correct

If it is the same process and brand JFET I would like to see the measurement for more gain not just say it has more gain.

EQ changes are easy to do but you need to be aware of what does what

At lower voltage differences cap changes do not matter as much. At the input where there is a greater voltage difference it does.

FWIW in some current test builds I am using tants on purpose at certain places.

Big caps with long leads give you a large loop area and the body of the microphone is not mu metal.

Sometimes you want the high pass switch to work.

How is the AMI transformer "better" than the stock one?
 
Yes, indeed. Lots of questions...
I have experimented with different FETs. In my tests the difference was usually not more than 2 dB output change.
As far as I remember, the J305 had in general the highest output when properly biased.

I am interested in the opinions about tantalum capacitors.
In the past I have had contact with Geoff Tanner (ex. Neve).
He always said: "Do NOT change the tantalum capacitors for modern aluminum capacitors, because that will degrade the sound quality. Those tants are there with a reason."
 
RuudNL said:
I am interested in the opinions about tantalum capacitors.
In the past I have had contact with Geoff Tanner (ex. Neve).
He always said: "Do NOT change the tantalum capacitors for modern aluminum capacitors, because that will degrade the sound quality. Those tants are there with a reason."

I'm assuming that "reason" was never specified or detailed? ::)

I'm just speculating, but could it be because, at the time, they were better than the available electrolytics, and/or provided more capacity per volume? And it just "stuck", for legacy reasons? Because no one dared alter the "classic design"...

Call me cynical, but there sure is a looooot of "voodoo" floating around...  ???
 
I don't have a u87 type mic with tantalums at the time, but it would be a piece of cake to inject a whole song through a mic, swap the capacitors for EXACT SAME VALUE electrolytics and do it again, then do a null test.
 
Yes, I agree, there are a lot of 'myths' in the audio world.
But what to think about (as I wrote earlier), somebody who can hear the difference between types of solder, hahaha!
 
RuudNL said:
Yes, I agree, there are a lot of 'myths' in the audio world.
But what to think about (as I wrote earlier), somebody who can hear the difference between types of solder, hahaha!

I don’t know... I did an entire studio with WBT 4% silver solder (leaded) after reading that’s what Jim Williams uses. I’d bet after an entire song with 100 tracks and analog console mixing, it’d contribute the smallest fraction of a percent in difference. If all the gear was soldered-up with the same, I bet even more. It’s simply logical. The problem is when people want scientific evidence even when it’s simply not practical, so they pass it off as people throwing fairy-dust magic. It really isn’t that much more expensive to do, so maybe it’s worth rolling the fairy-dust magic dice in such a case, when someone very-well grounded in the real world uses it.
 

Latest posts

Back
Top