U87 circuit, and HF roll-off curve

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

blue_luke

Well-known member
Joined
Dec 13, 2008
Messages
174
Location
Montréal, Québec
Hello all! :)
I am getting the point between K-47 and K-67(87?) differences in frequency response, and the need to adjust the impedance converter to compensate for the curve (too much boost in the 10khz region).
But I have two 'beef' (beefes?)
I don't have a reliable U87 FET schematic.
The one I can find on the web are quite blurred, a copy of the copy of the copy.... And mostly, I don't understand what components makes the equalisation as it is.
My instect would be that the EQ should be a bandpass , a bit broad bandpass, but I fail to 'see' this EQ in the circuit diagram.
Somebody can shed some light on this detail?
I have a project that will use two Microphone parts RK-47 capsules that I have, which in principle do not need this EQ, and I also have a bunch of K-87 caps that I removed from a few Apex 460 mods and the like...
I would love to experiment with these in a proper circuit.
Any ideas?
Thanks, Luc
 
The most important correction with the U87 is the peak around 10 KHz. In traditional designs this was always compensated by an HF rolloff caused by negative feedback. The disadvantage is that this also causes the real top end to disappear (the specification of the U87 is 16 KHz.)
That is why I have sometimes thought of a solution that only removes the wide 10 KHz peak. In that case, the highest frequencies should not be affected. I remember that forum member Kingkorg once experimented with an L-C filter. I think something like this could also work for the U87, perhaps supplemented with a little bit of HF roll off.
I want to experiment with this again.
 
The disadvantage is that this also causes the real top end to disappear (the specification of the U87 is 16 KHz.)
That is why I have sometimes thought of a solution that only removes the wide 10 KHz peak. In that case, the highest frequencies should not be affected.

I've though about that many times,
I don't like the 10khz boost (more or less), but I really like the Air Frequencies above 12Khz until 20khz,
a filter that removes the 10Khz peak but leaves the higher frequencies untouched makes complete sense and would be really useful.
 
Last edited:
I have read all the replies and thank you all!
I will look at the links suggested and this morning I spent some time searching for how to design a so called "band stop" circuit, which is the opposite of a bandpass filter
Such a circuit would remove about 6db in any band desired if well calculated/simulated.
Check here for a very good description of such a filter.
https://www.electronics-tutorials.ws/filter/band-stop-filter.html
 
Also there is a series of VDO's here with the actual guy who designed a lot of the RODE microphones and he explains quite well his rationale and his decision making process.
I own a few RODE mic and find them very well built, fairly priced and of really good value.

He starts at the bases, but he gets in the actual innards in a very well presented fashion.
A must watch for all DIYer that want to further his knowledge.

 
Also there is a series of VDO's here with the actual guy who designed a lot of the RODE microphones and he explains quite well his rationale and his decision making process.

I own a few RODE mic and find them very well built, fairly priced and of really good value.

Nice guy for sure and very knowledgeable, but then it all fails when Rode releases a bad sounding mic after the other.
Never heard any Rode microphone that sounded good in my opinion, and there’s none that I would want to use in one of my recording sessions.

Rode has great engineering, a great factory, very good production technology but then they make microphones that are voiced wrongly, they re highly sibilant have a bad sounding and agressive high end, it seems along the process they focussed only on circuit design and theory and forgot to listen to opinions of professional engineers and musicians.
Sometimes the sound most people love is not what looks good on paper or in circuit theories.

Fellow mate Kingkorg disagrees with me, but I know that my opinion on Rode mics as a professional engineer is shared with the majority of my colleagues.
 
Last edited:
Fellow mate Kingkorg disagrees with me, but I know that my opinion on Rode mics as a professional engineer in shared with the majority of my colleagues.
Tja, kinda do, kinda don't. I know exactly what you are talking about, and what you don't like. In all honesty i do like Røde mics, i learned how to adress their quirks, but they are not the first ones i reach for. I see them as great utility mics, i don't think they are bottleneck in a recording if used. I guess jack of all trades, master of none. Where they do shine is tremendous dynamic range with very low noise, especially the ones with pad. Also great capsule mod platform, stick whatever capsule you want, and the circuit won't get in the way. Kind of a mobile phone camera image with no filter applied. Never going to have wide aperture, large lense, shallow depth of field, with beautiful blurry background. But if you can use Photoshop, and know a thing or two about photography, you can make things happen.

There are some outliers tho, like Classic 1 and 2, TFM50, some ribbons. Whole different animals, deserve respect.
 
Where they do shine is tremendous dynamic range with very low noise, especially the ones with pad.

I know I've said to you in the past that I have a Rode NT1000 mic,
it's very low noise like you said, dynamic range is great, a lot of headroom and I don't don't the technical details like you do, but to me the low end transients sound great, very fast transients in the low end and very clear, I don't hear distortion in the low end transients, they sound very punchy with great definition.
But saying this I really don't like the high end of it, it's very sibilant and harsh, so I use that mic for kick drum, Cello and Double bass, instruments were the harsh high end doesn't get in the way and I can take advantage of the nice response it has for transients in the low end.
I don't ever use it on vocals, flute, horns, cymbals or room mic, etc etc

I though about trying a different capsule but then I got afraid I would destroy what I like about it in the lower register, so I left it has is.

There are some outliers tho, like Classic 1 and 2, TFM50, some ribbons. Whole different animals, deserve respect.

Although I've tried or used many different Rode mics in the past, I think I never used any of those models
 
Nice guy for sure and knowledgeable, but then it all fails when Rode releases a bad sounding mic after the other.
Never heard any Rode microphone that sounded good in my opinion, and there’s none that I would want to use in one of my recording sessions.

Rode has great engineering, a great factory, very good production technology but then they make microphones that are voiced wrongly, they re highly sibilant have a band sounding and agressive high end, it seems along the process they focussed only on circuit design and theory and forgot to listen to opinions of professional engineers and musicians.
Sometimes the sound most people love is not what looks good on paper or in circuit theories.

Fellow mate Kingkorg disagrees with me, but I know that my opinion on Rode mics as a professional engineer is shared with the majority of my colleagues.

How much of this difference of opinion is just due to some people judging a mic without any EQ, and others judging it after EQing it to have the kind of overall frequency response curve they want?

I find it weird that many people think a mic should sound great straight, and that overall FR curve is part of the "character" of the mic that you should like, out of the box, so the first requirement when picking a mic is that its overall FR curve flatter the source.

That seems completely backwards to me, because the one and only thing that you CAN straightforwardly fix with EQ is the overall FR curve. You can't fix the emphasis or demphasis of reverberant sound or bleed relative to direct sound, due to polar pattern issues, and you can't fix weirdness of the reverberant sound or bleed due to differing on- and off-axis FR curve shapes.

That suggests to me that in some sense, overall FR curve is the last thing that mic designers ought to worry about. If a mic is simply too bright or sibilant, you can EQ that right out, as long as the off-axis sound is similarly bright or sibilant.

But sometimes maybe you WANT the off-axis sound to be different from the on-axis sound in a certain way, to complement the reverberant sound.

For example, if you don't like the low frequency reverberation in a poorly-treated room, due to room modes, you may want to mic very close to maximize both direct (vs. reverberant) sound and (especially) direct low-frequency sound vs reverberant low-frequency sound. Then you can EQ that out and dramatically reduce the reverberant LF in the process. (You're basically pre-emphasizing the direct bass and then de-emphasizing both direct and indirect bass.)
 
How much of this difference of opinion is just due to some people judging a mic without any EQ, and others judging it after EQing it to have the kind of overall frequency response curve they want?

EQ you can apply to any sound source and can apply to any mic,
but a good sound without EQ only some microphones are able to provide.

I find it weird that many people think a mic should sound great straight,

Yes, a microphone to me has to sound great without any EQ applied,
it has to sound correct for the source and give me the sound as close as I want it to be without any EQ applied.
I have, as most everyone has, an extensive collection of EQs at my disposal I can then use any EQ to taste during or recording or mixing stage to improve on that or make the sound fit the mix.
But the recorded sound straight from the mic has to be good with no EQ to me.

I'm always able to find a mic that works great for a specific source of sound with no EQ, while I never listened to any Rode mic that was not harsh or sibilant.
So I don't personally like how Rode voices their mics, and as I've said most of my colleagues find the same.

and that overall FR curve is part of the "character" of the mic that you should like, out of the box, so the first requirement when picking a mic is that its overall FR curve flatter the source.

I don't care if the FR is flat or not, I just care that it sound good for that instrument and for the sound we are trying to achieve.

Of course taste and usefulness is subjective,
there's probably many people that like the sound of Rode mics and have made good records with them, but let's not forget that Rode mics got popular because they're were cheap entry level microphones and most people that have them have home studios.
A fact is you don't find many, if any, Rode microphones in big professional high-end studios.
 
EQ you can apply to any sound source and can apply to any mic,
but a good sound without EQ only some microphones are able to provide.

Sure, I understand that. What I don't understand is WHY it's so important to so many people.

Why is it important that the EQ be set right (to your taste) in the mic circuitry by default, rather than straightforwardly fixing it in your mixer or DAW, which accomplishes the same thing.

I for one don't want to even bother with several times as many expensive mics (much less pay for them all) just so I can pick one that's bright for some sources and another that's scooped for other sources, and one that has low bass extension for some other sources and yet others that are flat for sources I actually like as-is... while simultaneously satisfying all my other, more demanding criteria.

Yes, a microphone to me has to sound great without any EQ applied,
it has to sound correct for the source and give me the sound as close as I want it to be without any EQ applied.
I have, as most everyone has, an extensive collection of EQs at my disposal I can then use any EQ to taste during or recording or mixing stage to improve on that or make the sound fit the mix.
But the recorded sound straight from the mic has to be good with no EQ to me.

Why?

I'm always able to find a mic that works great for a specific source of sound with no EQ, while I never listened to any Rode mic that was not harsh or sibilant.
So I don't personally like how Rode voices their mics, and as I've said most of my colleagues find the same.

That sounds like your self-imposed discipline of NOT being willing to lift a finger to simply EQ the mic to your liking is the problem, not the mic.

To me that's like picking a smartphone based on screen brightness as shipped by default from the factory, without bothering to adjust it to see if you can make a too-dim phone brighter or a too-bright phone dimmer. Of all the things to pick a smartphone based on, that should be low on the list because there's a trivial fix for that. Better to pick a phone based on what its OS can do, processor, DRAM size, etc.

I seem to recall kingkorg saying that at least some of the Rode mics are the way they are because of well-thought-out tradeoffs, like setting the diaphragm tension(?) to make it less sensitive to plosives. If that means you need a little EQ, that seems just fine to me, and worth it.

I don't care if the FR is flat or not, I just care that it sound good for that instrument and for the sound we are trying to achieve.

Of course taste and usefulness is subjective,

So is lazy pickiness :)

As a cheapskate and somebody who is lazy in different ways, I'd rather make a few measurements and twiddle a couple of knobs to get my desired FRs than mess with a whole zoo of additional mics to avoid it.

there's probably many people that like the sound of Rode mics and have made good records with them, but let's not forget that Rode mics got popular because they're were cheap entry level microphones and most people that have them have home studios.
A fact is you don't find many, if any, Rode microphones in big professional high-end studios.

I really don't care, unless I can figure out a GOOD reason why, and an unwillingness to simply roll off objectionable treble is not a good reason in my book, unless there's something important I'm missing. (And there may well be.)

I'm not a fan of arguments from authority. I really don't care whether high-end studios use Rode mics, if they can sound great to me and the people who listen to my music, with some minor EQing. And if they can't, I want to know WHY, in detail, because that will guide how I deal with whatever the issue actually is.

I also do find it entirely plausible that most of the pros have some biases that aren't justified. I'm all too familiar with gear snobbery about other things, and I don't have a lot of use for it.

I am also not a professional myself, with paying clients to impress with my gear, and I don't have any desire to spend thousands and thousands of dollars extra on more and more microphones, just so I can be picky about overall FR curves as shipped from the factory. I could do it, I suppose, but it seems wasteful; I'd rather donate the money to a good cause, or spoil myself in more gratifying ways.
 
Sure, I understand that. What I don't understand is WHY it's so important to so many people.
I am pretty sure this comes from the analog age where you were stuck for the most part with board EQ curves. Especially for live. There was no guarantee HF filter of your eq would have the exact curve of u87 de-emphasis. So u87, and a mic with k67+ board EQ HF roll-off would sound different. Also if a mic is plugged into analog gear before it hits eq in the mixing stage, the recording chain would behave differently. U67, and u87 have significant LF roll-off, so LF rich content would hit recording chain differently than a flat mic.

I agree it doesn't matter much nowdays, in the digital age, but there are still many insisting on recording the "old fashioned" way.

There's also the thing where we, who are obviously obsessed with mics, know the fact that largest part of u87/u67 sound is sad face EQ curve applied to k67. But how many even professional producers know that, and pick a cheap k67 based mic, and start by applying sad face curve EQ? Most just hear it doesn't sound right, and ditch the mic.

I don't even want to go into simple topology transformer coupled mics into transformer coupled pres, you never know what you can get using these. In the end they are all just EQ curve, THD, phase... variations, but some people just want to use their favorite combos, use intuition, and get "the sound" immediately.
 
That completely explains your posts.
There's many things that I wrote that you will not have enough experience or knowledge to understand.

Maybe you could try explaining it to kingkorg, Khron et al., then.

Why exactly is it good to match the mic to the source BEFORE applying ANY EQ?
 
I agree with Whoops. EQ's mess up the signal especially if you have to make narrow boost or cuts. All of my recordings with the best end result sounded great when tracking. I find the low mids especially critical. If you don't have clarity and energy there from the beginning it's very hard to create it afterwards. Thats why U67 is my favourite microphone.

When I started mixing I used a lot of quite narrow dips and boosts. The tracks perhaps sounded good in solo but I'm not proud of the end result and I wasn't even back then. I'm still not a mix-master but I'm much better and I have learned a lot from a highly professional friend of mine. He's very conservative with EQ and push tracks in the right direction with sophisticated chains of compression, distorsion, exciters, short delays, etc and of course eqs but with broad curves and selected for a specific sound rather than clinical correction. Every plugin (he mixes in the box) usually doing very small changes but the sum is stunning.

At one time I had recorded some classical piano music. He listened to it and said something like: "It sounds great. If you apply eq it will sound clearer but don't do it. You will loose the energy of the piano. Leave it as it is."
 
Last edited:
I agree with Whoops. EQ's mess up the signal especially if you have to make narrow boost or cuts. All of my recordings with the best end result sounded great when tracking. I find the low mids especially critical. If you don't have clarity and energy there from the beginning it's very hard to create it afterwards. Thats why U67 is my favourite microphone.

When I started mixing I used a lot of quite narrow dips and boosts. The tracks perhaps sounded good in mono but I'm not proud of the end result and I wasn't even back then. I'm still not a mix-master but I'm much better and I have learned a lot from a highly professional friend of mine. He's very conservative with EQ and push tracks in the right direction with sophisticated chains of compression, distorsion, exciters, short delays, etc and of course eqs but with broad curves and selected for a specific sound rather than clinical correction. Every plugin (he mixes in the box) usually doing very small changes but the sum is stunning.

At one time I had recorded some classical piano music. He listened to it and said something like: "It sounds great. If you apply eq it will sound clearer but don't do it. You will loose the energy of the piano. Leave it as it is."
Huff, this gets way too complex. U67 you like has very broad EQ built in it. There is no scientific explanation why it would sound better built in a mic, or applied afterwards. So no sharp cuts or boosts here. I hate these as well!

Then, if i took your favorite u67, and introduced couple more microns of spacing between backplates it would create huge notch cut at a specific frequency. I promise it would sound bad. What's weird, it would sound less sibilant, but still bad. I wouldn't use any eq, but i would get the same result neither of us like. So is it really about EQ?

Every capsule has it's equivalent electrical circuit schematic. It is literally an EQ circuit. They operate exactly the same way, and the calculations and formulas are used to make all these capsules.

This is definitely one of those left vs right brain discussions, I like to think it's a matter of preference.

To be clear, i'm not claiming one or the other, just putting out things that come to mind.
 
Last edited:
Back
Top