Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[James Lugo]

I've had an '86 U87Ai and just acquired a '95 U87Ai. They sound pretty close, the '95 is a hair brighter and has a hair less low end. I ran them both through an A Designs 'Hammer', left the new one alone and boosted about 1.5db at 10k and cut 1db at 200Hz on the old one and ran them tip to tip on a few sources and they sounded almost identical.

Ultimately I want the newer one slightly darkened and maybe give it a slight low end boost. I do not want to touch the '86 if possible.

Can a tech match them? Is there someone in LA that does this?

[Klaus Heyne]

I would start by swapping the heads, and observe whether the specific characteristics you noticed about these mics follow the heads (my guess.)

Then you need to find someone with good ears who can manipulate the capsule performance of the newer K870, and possibly also adjust the older mic's capsule a tad.

[James Lugo]

Thanks Klaus.

[Tim Campbell]

This sounds like it's most likely the capsules.
From the way you describe the difference it would be easy for someone who builds or repairs capsules (not an average mic tech) to match their sound.

[James Lugo]

Thanks Tim. I'm gonna make some calls. I'd love to find someone local in LA. Maybe David Bock does this?

[Fletcher]

I don't believe David is a capsule guy... but Tony Merrill is if you're dead set on a guy from LA.

[J.J. Blair]

Be careful who works on it.  I would personally recommend against having somebody reskin the capsules to achieve a matching sound.  You will likely lose the correct U87 sound all together if that happens.  I would get two capsules from the same production run, and start by using those, then, make whatever adjustments to the electronics of the mic.  

When I rebuilt my M49s to become a matched pair, I got new K47s with consecutive numbers.  I was able to find a side of each capsule that were matched enough that I could use them with each other in cardioid, and consider it a "matched pair."  I also rebuilt the amplifiers, as well.  The mics were two serial numbers apart, and that was what I had to do to achieve that.  I suspect with U87Ais built 9 years apart, there could be a lot of factors that lead to audible differences.  But it will start with the capsules.  

[Arf! Mastering]

As Klaus said, first determine if the difference follows the amp or the capsule.  If the amp is the cause, it can be easily repaired.  If the difference is because of the capusle, this sort of perfectionistic quest can result in the loss of the authentic Neumann sound.  You've got two mics, each with slightly different personalities, but that can easily be made to work as a stereo pair with a little eq.  IMO, leave well enough alone.  A dB or two difference in frequency response is well within the specs of the capsule.

[Silvertone]

While on this subject... does anybody's two ears hear the same?

Not trying to be a wise ass but I've done stereo recordings with two different types of mics at times and have really dug the results (talking drums, not classical work here).

I find because of the way we (most engineers) create stereo mixes that one side always seems to be "the bright side" and one side always seems to be "the dark" side depending on what is panned where.

With all these frequencies floating in the stereo spectrum rarely do the drums sound "the same" (if you will) on each side... so I can tell you as a matter of fact no one has ever said to me, "did you use different mics on the drum overheads" while listening to the stereo mix.  Now, if you solo the overheads that's a different story    but again, sometimes I really like what the two different mics are capturing.

So I tend to agree with Alan here and say use a little EQ and call it a day.

Personally when doing critical stereo recording I always use these U87's modded by this guy named Klaus Heyne... if you can get him to do the work, you'd be all set. (Sometimes it was a modded pair of U67's but 9 times out of 10 I preferred the U87's.)

I'd also recommend Klaus' old right hand man, James Gangwer up in the good old Bay Area...

Good luck James.

[Dino Ziogas]

In the end, there's a standard little test you can do to determine in a practical way if two mics are acceptable as a stereo pair. You can ask for it after you've determined the cause of the discrepancies like the guys suggested.

[Klaus Heyne]

Silvertone wrote on Tue, 01 June 2010 04:13

I'd also recommend Klaus' old right hand man, James Gangwer up in the good old Bay Area...

Just to clarify:
James Gangwer never worked for me, and we have never worked together in any capacity.

During the Brauner Klaus Heyne Edition days, I appointed and recommended him as warranty station, because I had known him for decades as a conscientious individual (AND craftsman.)

James has deep knowledge of high quality professional audio components, including microphones, from his many years as technician for Dan Alexander with whom he is no longer affiliated.

[David Bock]

Quote:

While on this subject... does anybody's two ears hear the same?

Do you mean between the two ears of a person or sets of ears compared between people? Of course they're all identical!
This job is an example of where measurement can be really useful.

Quote:

Not trying to be a wise ass but I've done stereo recordings with two different types of mics at times and have really dug the results (talking drums, not classical work here).

Agreed. Precision matching can be over-rated, and often in the field, mics that are assumed to be matched or stereo are far from it.

[John Monforte]

Don't underestimate stereo matching!

I often used a demo in teaching that had a mono source treated very gently in one channel by any of a number of changes (level, complementary EQ, delay, etc.)

No matter what, the ear latches onto any interchannel difference and tries to ascribe some sort of position information out of it. That's fine when that's what you are after, but if preserving stereo image is important, matching is essential throughout the chain.

As far as these mics are concerned, the fact that two U87a's are made in different years is not an immediate red flag for me. While some might think sequential serial numbers is good enough for matching (I don't), I think there is enough variability in capsules that even an SM69 is not above scrutiny.

Put the mics as close as possible to each other in the far field of a loudspeaker making pink noise. Flip the phase on one and try to cancel the signal out. What is left is what doesn't match. Since the mics are not in the exact same point in space, there will always be a little HF in there. Even in good mics, you will be able to identify some variations. It is worse if they have been out and about for a few years.

Good luck with it!

[Mark Lemaire]

Simply shifting capsule heads from one mic to the other (as Klaus advises) is an excellent start to your project. You'll find how much of the differences are the capsule, how much is the amp. IMO- do NOT go buying new capsules, have your cap reskinned, or any other crazy thing. If you want a REAL stereo pair, you are almost there. If the differences are already subtle, it's detail work you need. You need a capsule guy, as they say.

[Jim Williams]

You may be able to match the mics, but so what? You will never be able to match the acoustics. Those spacing differences will create far larger response variations than any mic to mic differences.

Much ado about nothing. Record some music and don't get so hung up about stuff no one cared about 30 years ago. You know, when the gear was crap but the music was good!

[Mark Lemaire]

Quote:

You may be able to match the mics, but so what? You will never be able to match the acoustics. Those spacing differences will create far larger response variations than any mic to mic differences.

If you record orchestras and choirs (what I do for a living), a slightly un-matched pair is not very useful except to use as spots on different things- that is, I use them seperately, NOT as a pair. The distance from the orch makes the stereo pair's exact matching quite relevant.

If, on the other hand, your typical use might be close miking an acoustic guitar, miking two singers doing duets, etc, etc, then I agree that the differences between the different mikes are dwarfed by the different sounds that come off an instrument at close range.

[Stephen Andrew Bright]

It is my understanding that you will never get a closely matched pair using mics made 10 years apart. Aside from the age difference in the capsules,  the parts in the mics are likely to be different as well. My suggestion is to sell the one you don't like as much and look for another one made in the same year as yours. At least then you will be working with the same components.

I have actually built matched pairs of vintage Schoeps mics on eBay several times by finding consecutively serial'd mics from different sellers on different continents.

Stephen

[Mark Lemaire]

With all due respect to Stephen and other folks, this has not been my experience.

I have three beautifully matched stereo pairs of Neumanns: a pair of M269s, U67s, and KM86s. All are used daily in critical orchestral applications where exact matching is a necessity.

With the exception of the M269 pair, none of them were made in the same year. All have pretty different histories. The U67s are separated by some thousands of serial numbers. They were all stereo matched and modded by the guy who runs this forum.

I've no idea if Klaus changed out components that did not 'match', and have no in-depth knowledge of his methods. I do know that the mics function admirably.

It's all in the skills of the microphone tech. Your mics are so closely matched already that many posters here suggest you simply call it 'close enough'.

It may seem intuitive to suggest that you sell one of your U87s to get one that is closer to the other in manufacture date, but in my experience, that is not necessary if the mic tech has the skills. Not all do.

[Phil Mayor]

My 1968 U87 looks like a completely different microphone to my 1998 U87 internally. I'm assuming it would be difficult to match those, not that I would want to. I record rock, pop etc and I've never felt the need for an exactly matched pair of microphones.

[Klaus Heyne]

Your 1968 model is a U87. Your 1998 version is the successor model, U87A, introduced in 1986.

While fairly closely related in timbre, the processing of polarization voltage, gain, output, s/n, and other parameters are different from the earlier model and a bit harder to match between them than if both were the same U87 models.

[Jim Williams]

Stephen Andrew Bright wrote on Mon, 07 June 2010 09:14

. I have actually built matched pairs of vintage Schoeps mics on eBay several times by finding consecutively serial'd mics from different sellers on different continents. Stephen

Anyone with the time, fortitude and chops can select components and pre-match them. You need a quality DVM with enough digits to match resistors up to 10k ohms within 1 ohm. That insures a .01% match. Capacitors can be matched with a quality capacitor bridge. Transistors can be matched using a curve tracer. JFETS must be individually tested and selected as they have greater variation in specs than bipolar transistors.

That effort also allows one to choose the brand of components without the bean counters selecting them for you.

[David Satz]

> I have actually built matched pairs of vintage Schoeps mics on eBay several times by finding consecutively serial'd mics from different sellers on different continents.

I would certainly not put it that way. For one thing the terms "matched pair" and "vintage" are rather like oil and water to begin with, since capsules age and change their characteristics over decades even if they're safely locked away in a drawer somewhere. Instead, however, they are subjected to great variations in their conditions of use--and eBay sellers in my experience are notably bad at knowing or telling the full truth about that.

In addition, manufacturing tolerances were considerably less tight in the years when any microphone was made that deserves to be called "vintage" today. Unless you can control the detailed characteristics of your capsules, you can't make a matched pair of any microphones, new or old. Microphone amplifiers are purely electronic, rather than spanning the realms of acoustics and electronics as capsules do.

So (barring accident or incompetence) the amplifiers within any one microphone series are always more uniform in their characteristics than the capsules are. The more significant manufacturing variations among condenser microphones--both in dB and in their audible consequences--are always in the capsules. Yet it is the amplifiers that carry the serial number for the microphone as a whole, which tends to make "consecutive serial numbers" mainly a vanity concern.

--best regards

[Stephen Andrew Bright]

Hi David:

I should have added that I also send them back to Schoeps for service, and while that still may not make for a perfectly "matched pair" according to Schoeps' definition, they come back sounding indistinguishable to me (and with lower self-noise).

Actually, it was Schoeps that suggested I get amps that were made at the same time in order to get a good "closely matched pair." So while it doesn't speak for the capsules' condition, at least it gets you in the ballpark with the same amp components.

I mentioned this as an alternative to getting capsule adjustments (from the 2 or 3 guys with a 2-year waiting list) to compensate for a 10-year age spread in a pair of mics, and because these days it really is possible to find a 30-year-mic made at the same time as yours.

Best,
Stephen

[Klaus Heyne]

This thread is getting a bit frazzled. Back to what I know, and what I think matters to the initial inquiry:

1. Neumann's FET mic amps of the same series/version are typically within about 1/2 dB or less across the frequency range. If more, then there is usually a defect.

2. Neumann's Mylar capsules of the same type can spread from piece to piece as far as 4dB across the range when new (thus the

[DavidSpearritt]

Mark Lemaire wrote on Tue, 08 June 2010 06:53

I have three beautifully matched stereo pairs of Neumanns: a pair of M269s, U67s, and KM86s. All are used daily in critical orchestral applications where exact matching is a necessity.

Mark, how have you determined that these pairs are exactly matched?

[Mark Lemaire]

An interesting question. Here, for what it's worth, has been what I have done whenever I am comparing two mics that I am going to use as a pair.

When I got the pairs back from Klaus (each pair was done in different years), I put them up as close to each other as I could, facing the same direction, in my studio. Then I ran them through preamps set to the same level (I mostly use Millennia Media). I turned them up as hot as I could stand and listened for matching levels of self-noise.

After that, I spoke, sang, or played guitar into each mic (close and far) and then played back the results with the mics panned to center and switched back and forth between them, listening for differences in tone or character between the two.

If I need to replace a tube or some other thing happens that suggests to me that I need to double-check whether the mics still match well, I do the above again.

Of course in actual sessions I am not setting the mics in the same place, so there are differences between what each one picks up. But I like to know that the differences are more derived from the mic's locations than the differences in the sounds of the different mics themselves.

I am not a scientist. And I know that my voice or guitar will not, for instance, produce low bass tones for me to compare. Still, I use the term "beautifully matched" because these stereo pairs have created lovely recordings for me and my clients over the years, with no one questioning whether the mics 'match'.

Let me know, David, if I have not answered your question fully enough.

[Klaus Heyne]

Mark Lemaire wrote on Thu, 10 June 2010 23:55

I am not a scientist.

Science as applicable to sensory investigation (in this case, hearing)  is largely an empirical undertaking (lest we forget how Bell came up with the loudness dB!)

So, you did just fine, from a scientific standpoint. You used the scientific method as fully as available to inquiry into a sensory phenomenon- hearing- that is still largely unexplored and unquantifiable.

Frequency response similarities between two mics don't go very far (see also the recent thread on ribbon mIc pairings) and leave out an important aspect of a microphone's character: timbre. You could theoretically find a Chinese-made cheapo with similar frequency response curve as a Neumann U87, yet the two mics will hardly sound alike.

[Dino Ziogas]

Take into account that any differences between the mics will be more prominent in coincident stereo arrays. Due to their nature, spaced stereo techniques are less prone to microphone mismatch.

One way to determine suitability of two mics for use as a [coincident] stereo pair is the following:

Have the two mics placed one on top of the other with the capsules as close as possible and facing towards the same direction. Have someone talk to them from around 60-70cm away and towards the middle distance of their diaphragms to ensure even pickup [you can use a speaker instead of an assistant].

Have the mics routed in two channels and pan them full left and right. Turn monitoring in mono. Use a regular amount of preamp gain for one mic and while turning the other mic channel out of polarity ["out of phase"] adjust gain so you get the deepest null possible. Now, you've calibrated the signal chains as much as possible. Flip the second mic phase back to normal. Turn monitoring back to stereo.

Have the assistant walk in a perfect circle around the mics [tie him with a string, he he - if the image does not stay reasonably centered you've got mismatch.

As a final check, turn one mic left and one right [45 degrees each] and by use of the assistant still walking make sure you get a good tracking of the source.

[Mark Lemaire]

Dino- Thanks for the idea! Now, I must find a piece of string...

[DavidSpearritt]

Thanks Klaus and Mark. I am not convinced about the necessity for  "matched pairs" from the marketing of SDC's. I think Mark's approach, checking noise, gain and timbre is appropriate, especially for LDC's, where it's difficult to calibrate the capsules in their head baskets. DPA seem to "match" by checking FR magnitude and phase plots, noise and sensitivity to within a tolerance.
http://www.dpamicrophones.com/en/products.aspx?c=Item&ca tegory=186&item=24005

[John Monforte]

To those who advocate testing a match by voice, please be aware that a voice has a limited frequency spectrum and some elements like sibilance appear only briefly.

Besides mismatches in sensitivity, which any test can identify, mismatches will tend to be found at the frequency extremes. That is why I recommended the pink noise in the test I described. That test is quick thorough and does not require instrumentation or subjectivity.