R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: [1]   Go Down

Author Topic: DIY Mic Pre Kits?  (Read 16744 times)

onekid

  • Newbie
  • *
  • Offline Offline
  • Posts: 39
DIY Mic Pre Kits?
« on: May 01, 2004, 04:26:04 pm »

Does anyone have any experience with any DIY Mic Pre kits out there? I am particularly interested in the Seventh Circle Studios Neve, Jensen Twin Servo, and API kits they have. The price seems very reasonable for someone handy with a soldering iron.
Logged
Jeffrey Lonigro

spiral

  • Newbie
  • *
  • Offline Offline
  • Posts: 4
Re: DIY Mic Pre Kits?
« Reply #1 on: May 01, 2004, 07:04:51 pm »

The Hamptone JFET kit is very nicely put together and i think it sounds amazing. The price is crazy too for 2 nice channels of tons w/ gain and transformers on each channel. www.hamptone.com
Logged

John Klett

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 475
Re: DIY Mic Pre Kits?
« Reply #2 on: May 02, 2004, 04:07:33 pm »

First off - I have not built or heard these kits.  They look pretty good though.

Comments - based on looking over what is available on the web site.

J99

The J99 looks to be a direct implementation of the Jensen Twin Servo architecture that has been published in product handbooks and on the Jensen site.  People were encouraged to make copies because Jensen sold 990's and transformers that way.  The 990-derived mic preamp circuits are,therefore, public domain.  

John Hardy makes 990 based preamps, he makes his own 990 opamps (those are public domain as well) and he uses Jensen transformers.  His M1 is uses a single 990 and his Jensen Twin Servo is what it is - a Twin Servo.  He has the M-2 which is an M1 with a precision gain switch.  The Twin Servo is certainly a very good microphone preamp and worth having.

http://www.johnhardyco.com/  

it looks like Seventh Circle is offering a faithful reproduction of the Twin Servo architecture in Kit form and as an assembled product.  In a quick look at the schematics and Bill Of Materials (BOM) I am confused.  Hardy 990C's are specified in the main part of the list but down at the bottom there is a separate BOM for an opamp. So...  it may be that you have a choice of 990C's which are expensive or, maybe, you can stuff an opamp in there that is not any derivation of a 990.

The difference between the opamp indicated in the separate BOM and a 990 is that the one in the separate BOM is simply an Analog Devices low noise AD797 IC opamp with a BUF634 high current buffer to boost output current.  That amplifier is going to have a lower output impedance than a 990 and will drive the transformer differently.  It will not sound the same as a 990 based Twin Servo.

If my supposition is correct, that you are getting a choice between real 990's and the AD797/BUF634, you should figure out if you want a genuine Twin Servo or one that is based on some other amplifier.  They probably both sound good.  My experience in messing around with the twin servo architecture is that the gain split and servoed twin amplifier setup works well with lots of amplifiers.  

A 990 based version of the J99 would be my first choice of the kits.


A12

The A12 seems to be a good copy of an API-312.  The BOM is not available so there are a few unknowns.  

One unknown is the amplifier.  Is it a 2520?  API had problems themselves over the years making 2520's that maintained the original 2520 "sound".  That sound is result of the rather slow and asymmetrical output of that amplifier - technically not the greatest opamp - but when it's linked with that crappy $8 output transformer magic still happens.  I really like that preamp.  The harder you drive it (up to the point where it starts clipping anyway) the cooler it sounds.  Technical specifications are not everything.  So...  here the issue is what the amplifier really is...  api is not selling 2520's to people making clones if someone is making 2520's, which of well over 25 versions are they trying to clone and where are they getting the transistors?  This is not to say that they didn't figure it out but it certainly begs the question.  

Then there are transformers.  The very first 312 cards had a tall round input transformer that had it's own sound.  Some api-o-philes seek those out and pay premium prices for those.  Currently api uses, I think, a Jensen JE-115...  or JT-115 or whatever they changed the prefix to.  That's a good transformer and works well on the input side of a 2520 based preamp.  

The OUTPUT transformer is more of the trick.  Originally that transformer was a cheap "low spec" all-iron transformer.  They did not really get into anything esoteric in the spec... it was (probably)  this size, this many dBm, the primary and secondary ratios...  and that's about it.  It turned out that there were various iterations of the output transformer as there was in the opamp.  The steel laminations changed over time.  You can have grain oriented and non-grain oriented (isomorphic?) laminations and some of the problems over the years with changes in the sound of api product can be traced to those changes in the laminations.  API seems to have both the transformer and opamp issues worked out today...  but once again it is a question that comes up.  

MY advice with any api preamp clone is that you have to put one up against an early api 512 or 312 and compare.  The clones will probably sound good but will they sound the SAME as the api ones?  For a time api couldn't quite get it right so...  who knows.  Try before you buy if you want the true vintage api sound or simply expect to get something reasonably close and be happy.



OKAY - N72

I have expressed (ranted) my opinion of 1272's as mic preamps over and over but this is a new forum so...  here's the book.  

The N72 is yet another iteration of a modified Neve 1272 module.  This one looks better than most but instead of making a microphone preamp the way that Neve did they are doing what almost everyone else making "Neve" preamps have been doing.  At least Seventh Circle used a stepped switch combination of attenuation and amplifier gain control instead of a simple gain pot that works the preamp gain by removing feedback only.  I don't want to really knock Seventh Circle because the implementation looks quite good, but, it's a 1272 with a lot of gain added and NOT a clone of any mic preamp Neve ever used in a channel amplifier.  Seventh Circle could have added another three transistors and another deck on their sensitivity switch and gotten it right but they chose not to make the effort. I find that disappointing.

Up front - 1272's are NOT mic preamps.  They were specifically designed to be gain makeup amplifiers for passive-loss summing busses in Neve 80xx series consoles.  They can claim to be Neve style mic preamps...  not as used in 1073, 1084 and other "Class A" modules using BA183 (or the later BA283) and BA184 (or the later BA284) amplifier cards.

digression/history...

The BA183 and 184 amplifiers were based on amplifier circuits that were originally defined in the B100 through B105 amplifiers that were first used in the 1053 channel amplifiers (earliest I have seen anyway).  These were germanium, positive ground and largely PNP circuits.

The 1063 channel amplifier was the first module that was negative ground.  The two amplifier boards used in the 1063 were essentially NPN negative ground reworks of the original PNP circuits.  Same circuits - reversed polarity.  These were laid out on two boards - the BA 183 and the BA184.  The EQ filter networks were were on B180, B181 and B182.  Later on Neve came up with a real high gain quasi opamp in the BA-238.  That amplifier had enough open loop gain to use in a single stage preamp.  Neve needed to make this move because the older architecture cost more to build - it simplified the sensitivity switch for one thing and the plug in modules went onto a main board like the BA312 used in the 1081 and related "Class AB" modules.  The 238, 338 and 438 are all the same preamp for the most part and they are NOT class AB, they are class A.

back on track...

On the BA 183 board there is one preamplifier and one output amplifier.  Collectively (the preamp, output amplifier and transformer) are really optimized to run at gains up to about 55 or 60 dB.  Neve documentation suggests that the individual amplifiers could be run up to fairly high gains but Neve did not use them in those high gain regions in their mic preamps.

In a Neve channel amplifier using 183 or 283 boards there is a second amplifier card that has three amplifiers, the BA184 or BA284.   Two amplifiers are used in the EQ and the third amplifier serves as a SECOND preamplifier in the mic pre that is switched in at around 60dB gain on the sensitivity switch.  A second preamp is required at higher gains because the individual preamplifier's open loop gain limits the maximum gain where they can operate while maintaining enough feedback to hold down distortion and maintain stability and bandwidth.

If you look at the front of a Neve 1073 channel amplifier you'll see an "off" position on the sensitivity switch between -50 and -55 dB.  At sensitivities -55 and higher the second preamp is switched in so that the gains of the individual amplifiers are held within their best operating range.  That is the way every channel amplifier using a 183 style output was done.  If people want to clone a 1073 style preamp they may want to start by looking at a Neve 1290 module instead of a 1272.  

Here is a schematic of a 1290 for informational purposes.

http://www.technicalaudio.com/pdf/neve/1290-EN10041.pdf

The only reason 1272's have become the standard for cloners is that when you break up an old Neve desk, you end up with a pile of 1272's that had been used for makeup gain on the busses.  BA184 and 284 boards don't turn up.  BA183 and 283's do in the channel switcher/router modules but once someone has one of those and the transformer the temptation to make TWO preamps is too great.  They don't use the second 183 to get another preamp and there you go.

The output transformer.  Seventh Circle is probably using a Carnhill 69095 - that is the correct choice - they are calling it out by a different SCA part number but they are getting it from Carnhill.  This is a gapped core transformer and would be the replacement for the Neve LO1166.  Gapped core transformers will tolerate DC current through their windings without saturating.  This is mandatory if you are building anything with a 183 style output.  They are on the ball with this choice but I have seen a few amplifiers of this type that don't use a gapped transformer and they have significant distortion and are marketed as having that "warm Neve Class A sound".  2% distortion at 50Hz at +4 output is NOT the Neve sound.  

Now...  I need to admit that I have a pair of 1272 based mic preamps.  They sound fine at lower gains but they definitely loose headroom and generally crap out at higher gains.  It AIN'T a Neve mic preamp.  It's a preamp made from Neve components.

The Seventh Circle N72 is probably just fine and very Neve-like up to moderate gain levels but if you have a low output mic that needs a bunch of gain you may want to go with something else.  A lot of mics today are pretty hot so this whole high gain issue may not crop up.

The rag I am chewing on is that blah blah '72 preamps are not what Neve made as mic preamps and I am generally annoyed that most people making these things imply that they are deliver something that really isn't.  Once one is made aware of all this one can make a purchase based on real knowledge of what is going on.


Anyway - that's pretty much my whole "thing" on Neve 1272's as mic preamps.
Logged
John Klett / Tech Mecca
http://www.technicalaudio.com

SeventhCircle

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
Re: DIY Mic Pre Kits?
« Reply #3 on: May 07, 2004, 04:27:55 pm »

Hi guys,

Thanks for the interest in our preamp kits and also for the comments.  It's nice to get some informed feedback, even if it isn't in the form of a ringing endorsement.  I'm sure Mr. Klett is quite busy and has better things to do than peruse our web site, and I really don't want to spam the thread, but I thought I'd just clarify a few points.  It's important to keep in mind that all the kits we offer are designed to be built with modern, commonly available components by persons who may have very little electronics experience, and so from the outset are not intended to be precise reproductions of the products they're based on.  In addition to sounding great and being affordable, the kits need to be easy to build.

Quote:

It looks like Seventh Circle is offering a faithful reproduction of the Twin Servo architecture in Kit form and as an assembled product.


Actually, at this time the J99 is only available as what we refer to as a "minimal" kit, which contains a set of PC boards, Grayhill rotary switch, Neutrik connectors, and a few other odds and ends.  We're planning to start offering it as a "full" kit eventually.

Quote:

In a quick look at the schematics and Bill Of Materials (BOM) I am confused.  Hardy 990C's are specified in the main part of the list but down at the bottom there is a separate BOM for an opamp. So...  it may be that you have a choice of 990C's which are expensive or, maybe, you can stuff an opamp in there that is not any derivation of a 990.


Quite right.  Since there's no opamp included, it's up to the kitbuilder to decide if s/he wants a pair of Hardy 990s, a pair of hybrid modules, or something else entirely.

Quote:

The difference between the opamp indicated in the separate BOM and a 990 is that the one in the separate BOM is simply an Analog Devices low noise AD797 IC opamp with a BUF634 high current buffer to boost output current.  That amplifier is going to have a lower output impedance than a 990 and will drive the transformer differently.  It will not sound the same as a 990 based Twin Servo.


Some people may think it sounds better!  The AD797 is one heck of an IC opamp, and equals or betters the 990c in just about every figure of merit except output current and maximum supply voltage.  The BUF634 can source 250mA of current, can slew its output at 2000V/uS, and has built-in thermal protection.  Built with the suggested parts and supplied circuit boards, these modules provide very high performance at less than half the cost of a 990c.  They're also a lot easier to make than a discrete opamp.

Quote:

The A12 seems to be a good copy of an API-312.


Well, not exactly.  The web site doesn't have all the information yet, so a little explanation is in order.  The A12 circuit board can be populated to produce a reasonable facsimile of an API 312.  It can also be built with alternate components to produce something much more like a Hardy M-1 than a 312.  A number of input transformers can be fitted, including models from Jensen, CineMag, and Lundahl.  A variety of output transformers can also be used.  It's up to the kitbuilder to decide what they want.

Quote:

One unknown is the amplifier.  Is it a 2520?


No, it's not.  As you say, API released numerous versions of the 2520 over the years, and we have no interest in nominating one version as the "best" and attempting to clone it.  When the A12 full kits become available in a few weeks, kitbuilders will have the option to use a hybrid IC amplifier (built with an OPA627 and BUF634), obtain a 2520 clone, or scrounge up a used 2520 on ebay.

Quote:

MY advice with any api preamp clone is that you have to put one up against an early api 512 or 312 and compare.  The clones will probably sound good but will they sound the SAME as the api ones?


The A12 full kits will contain transformers that we feel are quite close to the originals.  Both have cores with considerable iron content, and the input maintains the original 1:8 ratio.  Those looking for better API "fidelity" have the option of using a 2520 clone (like the very nice 1122 from Avedis at Brent Averill).  Or they can buy an API!

Quote:

The N72...(is) a 1272 with a lot of gain added and NOT a clone of any mic preamp Neve ever used in a channel amplifier.


Quite true, but the fact that Neve didn't do it doesn't mean that for most applications the circuit doesn't perform quite well.  Unlike many 1272 conversions, the N72 runs the first stage with enough feedback to maintain low distortion and wide bandwidth.  The overall gain of 70dB (including transformer step-up) is achieved by taking advantage of gain available in the output stage.

Quote:

Seventh Circle could have added another three transistors and another deck on their sensitivity switch and gotten it right but they chose not to make the effort. I find that disappointing.


Sorry to disappoint you!  A few things here: First, we might have gotten three more transistors on the board, but not nine more resistors and three more caps.  The web site photos need updating, but the current N72s also have local regulators on board.  That, coupled with the wide PC traces used throughout, preclude the addition of a another gain stage.  Second, the rotary switch is a three deck switch, one deck for the attenuator, the second deck for first stage gain, and the third for output stage gain.  Third, and most importantly, why bother with unnecessary complexity?  The 70dB of available gain is plenty for most applications, and the two-stage circuit can provide it with excellent fidelity.

Quote:

Seventh Circle is probably using a Carnhill 69095 - that is the correct choice - they are calling it out by a different SCA part number but they are getting it from Carnhill.  This is a gapped core transformer and would be the replacement for the Neve LO1166.


We order them with a different number, but the N72 uses Carnhill transformers for input and output.  The output transformer is essentially an LO1166 with a universal mounting clamp and flying leads.  We feel they're the best choice currently available.

Quote:

The rag I am chewing on is that blah blah '72 preamps are not what Neve made as mic preamps and I am generally annoyed that most people making these things imply that they are deliver something that really isn't.  Once one is made aware of all this one can make a purchase based on real knowledge of what is going on.


We couldn't agree more.  We don't claim that the N72 is sonically indistinguishable from a 1073, 1290, or even a 1272.  It is what it is, a very high quality, Neve-like preamp based on the BA183/283 circuit.  But people in search of Neve-like tone have been anything but disappointed with the N72.

With all the aspiring recording engineers out there these days, we're hoping that at least a few of them will choose to improve their signal chain by building a few kits, and learn a little something in the process.

Please feel free to e-mail me with questions!

-Tim
Logged
-Tim Ryan

Marik

  • Full Member
  • ***
  • Offline Offline
  • Posts: 215
Re: DIY Mic Pre Kits?
« Reply #4 on: May 09, 2004, 12:19:43 am »

Kevin Carter of K&K audio has a RCA BC2b kit and some others. If anybody is interested I can let him know and he probably could stop by with more details.
Logged
Mark Fouxman
Samar Audio & Microphone Design
www.samaraudiodesign.com
Pages: [1]   Go Up