trafoless microphone preamp wish list

[pstamler]

[quote author="BYacey"]The Valley circuit has an input impedance of 50K to 100K depending on which app notes you look at. pstamler, why would you say the SM57 requires a low impedance load? I have used SM57s with the Valley circuit, and they sound great. As well, the SM57 has an internal transformer between the voice coil and the XLR of the mic.[/quote]

Hearsay, I admit, but the hearsay comes from Scott Dorsey, whose ears I mostly trust. I personally can't stand SM57s into any preamp at all, so I'm not really a good unprejudiced witness, but Scott swears they perform a lot better into a tranny with an input Z of maybe 600 ohms.

Peace,
Paul

 

[chrissugar]

[quote author="pstamler"]
I'm told it has multiple low-gain amplifiers; you switch in and out however many you need for the gain you want.
[/quote]

NewYorkDave posted this:

There are only two stages, each with six steps from 10-35dB, for
thirteen steps from 10-70dB, with the second stage cutting in at 40dB.
Each step can be calibrated and is "gain staged" to optimize dynamic
range and distortion. Most of the circuitry in the photo is for
support functions, primarily: closed-loop voltage and current
regulators for each stage, ten discrete closed-loop supply regulators
per channel, output load compensation, and gain control.

Grant Carpenter
Gordon Instruments

there are only two gain stages.

chrissugar

 

[Rossi]

[quote author="BYacey"]The Valley circuit has an input impedance of 50K to 100K depending on which app notes you look at.[/quote]

Maybe this is a silly question. But doesn't the way phantom power is hooked up limit the maximum input impedance to just below 13.6 K ?

 

[BYacey]

That is true, but I switch the phantom resistors and caps out of the circuit completely when not required. Just a double pole switch.

 

[bcarso]

Thanks all for the continuing dialogue. Thanks Paul S. for clarifying the noise/distortion criterion in particular.

Re use of FETs: I get some encouraging results from sims using the Toshiba 2SK170, depending on the details, for low above-third distortion. Unfortunately I have doubts about the validity of the model----when I played with 170's once upon a time for the potential use as a square-law device for a quarter-square multiplier application, I couldn't fit well at all to the canonical equation that suposedly governs all FETs.

Then I saw a remark by Barrie Gilbert that modern FETs hadn't worked according to this oft-quoted law for some time---he didn't explain why, and when I pressed he referred me to Tsividis's book---i.e., I might as well have been signing up for a graduate course in FETs ;-).

Bipolars OTOH are well-known for harmonics out to as far as you wish to look. However, if the predistortion designs are used, as can be seen within the structures used for analog multipliers for example, a well-matched set of bipolars can give impressively low distortion over quite a range of currents. But the simpler structures are notably not low noise, although local feedback with resistors can help a lot (see the thread linked above from mic pre discussions dating back a bit, with PRR pointing out the noise degradation of current sources vs. resistors given the same available voltages).

Overall, I think we are sort of converging to a target spec even if we don't know how it is to be achieved---and that's what I really started the thread for. But there are some clues to the design here and there, and the Gordon preamp data and subjective accounts are percolating in my skull.

How much does that Grodon bad boy run?

 

[chrissugar]

[quote author="bcarso"]How much does that Grodon bad boy run?[/quote]
2600$ for the two channel version.

chrissugar

 

[bcarso]

Hmmm---they are not giving it away, but with that much stuff and the apparent performance it sounds fair.

 

[analag]

My soldering iron's been plugged in since Sunday.

 

[CJ]

The Valley Pre sounds just fine with the 57. And that's comparing it plugged into a V76, which will make a piece of tin foil off a prebaked TV dinner with the mashed pototoes stuck to the top sound like a Buls.

Remember the input impedance formed by two BJT' emitter-base junctions in parallel per side. Also, feedback has something to do with input impedance dosen't it?

The Gordon seems great on paper. I have never heard it. I do not know if output loading is an issue to the point to where you would need to write an algorithm (which apparently what is controlling the output impedance in the Gordon) to control it.

OK, we are trading stuff already here. Bulletproof was one priority. Are SK170's tough enough for a front end? Can we somehow protect them without an xfmr? Are zeners quick enough?

Which distorts nicer, Fets or transistors?

I like the external pwr thing already. Now we are starting with minus 1000 db noise.

Question 1: Is there something out there that does everything we are asking for already?

Question 2: What does the Valley People lack as far as something we might end up with after all is said and done? The only criteria it lacks would be no feedback/gain control.

I agree 100 percent that the gain should be fixed. One setting for dynamics, one setting for condensers, and thats it.
The Langevin AM-16 is the best sounding sand state amp I have heard so far. It has no gain control. But it does have some hellishly good transformers.

This guy seemed real confident. I would like to hear some reviews of this thing.

 

[tk@halmi]

ten discrete closed-loop supply regulators per channel

What heck is this? Why do you need ten of anything in there?

 

[fum]

[quote author="analag"]My soldering iron's been plugged in since Sunday.[/quote]

Cool, do you think you could make me some XLR cables? :wink:

 

[chrissugar]

[quote author="CJ"]
This guy seemed real confident. I would like to hear some reviews of this thing.
[/quote]

I searched for some reviews for the Gordon preamp and I extracted only some relevant posts to spare you the reading of long threads.
Couldn't find a post that say it is not the most colourless, transparent preamp. It looks like it is some absolutelly special engineering.

Here are some:

"The Gordon Instruments Model 5 is the most neutral and the quietest preamp I have ever heard. I recommend it for everything when you want the sound of your mike to shine through. It's design is unlike any other preamp of which I am aware."

"He mentioned to me that after hearing it, Martina McBride (or was it Trisha Yearwood?) loved it so much she went and got herself her very own immediately with her own money!"

"if you are in need of the highest resolution and maximal extraction from your microphones,the Gordon Instruments microphone preamps,is your answer.I am a happy owner of the model 2 mic preamp.The preamp has a noiseless background,as i refer to my friends " black background",and it's strength lies also in handling *s* iformation/details.The model 5 and 4 is out now, which came as a result when Grant Carpenter was designing his not-yet released power amp.In a e-mail exchange with him,the new preamps actually offer higher resolution and are less colored(yikes!!) than previous models.model 5 costs $2900 with gain control, if you need a version with higher perfomance and quality parts( or sjall i sau deluxe parts) add another 1400.
I am not affiliated with the gordon instruments,just a fan here in europe."

"The absolute best preamp **ever** in terms of clarity, lack of distortion and overall great sound is Grant Carpenter's Model 3 (and now Model 4/5) preamps. The brand name is Gordon Instruments. Find him on the web, read his philosophy and get a demo. I can all but guarantee you'll agree this is the best preamp you've ever heard.
...
I own Millennia HV-3B, API, Amek/Neve. I have frequent contact with Grace, U-Audio and Focusrite Red. None of these come close in quality of sound. The Gordon is not a personality preamp so skip it if that is what you're looking for.
Forgive me if I sound a little hyper-enthusiastic about this, but this is one of the few audio products I really get excited about!"

"The Gordon mic pre is wider,bigger and flatter than the grace.
the grace has slimmer bass response and somewhat hyped higher frequencies ,compared to the gordon.the grace is somewhat more "forward/agressive) sounding."

"Upon our first audition we were so startled by the difference between the Gordon and the other pres that we immediately assumed we'd done something quite wrong, and/or that the mic feeding the Gordon was much better than the others. So we rotated the connectors at the mics, moving each mic to the next preamp channel and tried again. We got the same result, and that result coninued regardless of which mic fed which preamp channel. Further, no additional adjustment of preamp gain was necessary to maintain the within-a-dB calibration.
Now, all of these are truly fine preamps. I work regularly with the Great River and the Millennia, and I always appreciate them, and I enjoyed the opportunity to work even a little with the Grace. But the Gordon is something else in comparison. Given the specs and what I was hearing I can't convince myself that the Gordon preamp was offering some kind of fabulously euphonic coloration. On the other hand it almost strains my own beliefs to think there could be so much additional transparency available from any preamp.
I look forward to another round with a Gordon preamp, and I would love to run it against the GML. So far I have never heard any preamp approach the clarity and realism I heard from the Gordon "

"Well we had the pleasure of having one for about a month a couple years ago. Grant(the builder) wanted us to see what we thought. It is a great pre. I would say that it is the closest thing to a straight wire with gain that I ever have heard. The engineers who used it loved it except that the general feeling was that it lacked any color. Is just a preamp, nothing more or less. It is extremely fast. Probably the best sounding pre I have ever heard or drums and percussion.
It is a case of you need to really hear it to make your own judgement. It is very expensive. Way above the range of most people I know. Like Bill said, it is chocked full of components. I don't think I have ever seen so many parts inside a case that size. It looks more like a power amplifier than a mic pre.
It is a wonderful instrument, but you need to listen for yourself."'

"I saw Russ at the AES show and he wanted to make sure I knew about the Gordon. I have known Russ for a long time and have loads of respect for his work (Newboys, Sixpence None the Richer), He said that the Gordon, which he bought, was "the best preamp I've ever heard in my whole life." And this from a man who's owned Neve, Hardy, API and more. He heard it and loved it. He said he uses it on everything. That tells me a lot.
Lynn Fuston
3D Audio"

"My friend Russ Long has a Gordon model 3. I've used it a couple dozen times or so, and it is just incredible.
I made the mistake of a/b-ing it with my brand new $2000 preamp. I was blown away how much better the Gordon sounded; easily worth the extra $1500. I sold my preamp shortly thereafter."

"The most holographic recorded sound I?ve heard. This was the Gordon Model 2. And it did it almost every time."


chrissugar

 

[analag]

Now that we have heard about the worlds greatest pre....what are we gonna build now. And I was getting excited about building this preamp, I'm not one for the idle talk so I'm going ahead with my Transformerless Fetcode spinoff.

Analag

 

[chrissugar]

Two projects are even better. One tube/jfet based preamp (like the Fetcode) and another exclusively solidstate preamp.

chrissugar

 

[Svart]

wow this was a hot thread yesterday, but now it's died..

If you look at the Gordon, take a close look at the relays. those aren't sealed or expensive relays.. those mylar caps aren't expensive either.. Obviously there is a trafo under that central cylindrical area, and man look at the number of caps!

I think they took a hard look at making sure the power was super clean..

I think that will get us farther than we usually go.

I think we should definately use lm317/337s with large amounts of smoothing and decoupling caps. Maybe now is a good time to test the FRED diodes as rectifier elements( and maybe source some of those new ones too), eh Bcarso?

Also on the gordon, look near the output XLRs.. BURRBROWN opamps.. yep.

Look near the inputs.. do those transistors have more than 3 legs? looks like AD mat-02s? also under those are more BB parts..

those shiny resistors are ultra precision, something like .01% I just bought some for a project and look just like those..

I'd like to see what you are thinking of doing Bcarso, maybe we could offer more ideas, I'd like to start helping getting this moving before the momentum dies and everybody bails for the next GroupDIY project of the minute... :green:

 

[NewYorkDave]

Maybe someone could get Grant from Gordon to drop in here. I don't expect that he'd reveal trade secrets, but perhaps he'd be willing to discuss general principles, etc.

I don't count myself as an active participant in this thread since I'm perfectly happy to build my preamps with input transformers. :green:

 

[bcarso]

Svart, thanks for keeping this alive. I am doing some models as we speak for sections of this paying special attention to the Stamler Criterion (sounds like a Robert Ludlum spy novel title) of noise-floor-level 5th and above. It is tough, but then that's what's exciting. At this point a sort of brute force approach is needing even higher rails that I expected and quite a fair amount of power dissipation too.

Since I as I mentioned I don't trust the Spice FET models I may have to breadboard something soon anyway, and then figure out how in the world to measure at these levels. I have another design, for a powered sub, that must make a Friday deadline with at least a schematic so this may be a little while.

At the same time, where this thread started was to gather a wish list. If it leads to more than one design, let a thousand preamps bloom!

As far as the Gordon, it sounds like the neutral-type preamp to beat or equal. It does probably entail comparable complexity. I think most of the ICs in the Gordon are in supporting roles like d.c. servos, current sources, etc.

As far as the voltage regulators go, I would use discrete-based designs for lower noise and higher performance. Maybe 317-ish parts for preregulators.

As far as the diodes, it would be fun to see if the SiC parts make a difference, other than just having higher voltage drop.

One thing that bedeviled another design I did was not so much the reverse-recovery snaps but just good old peak charging current magnetic emissions from the toroid. There were some unavoidable big loops that coupled to the trafo and made life difficult. It has driven me towards building in extra voltage and throwing it away in resistors to limit peak currents, and also specifying the trafo for higher current than needed to begin with, so as to steer things away from saturation.

Gordon is probably mag-shielding that big iron, as well as using it at a small fraction of its rated current, to allow that kind of proximity to what all say is an ultra-low-noise design.

 

[Svart]

Bcarso, I see a lot of other "cheapo" opamps in that picture too, but only see the *good* stuff near the i/o. I would bet that you are correct in them using servos, but would a servo *need* to be a such a high quality part? I mean I assume that they are spending where it matters most and saving where they can get away with lesser parts..

I also assume that they are trying to stay away from caps in the audio path, maybe even discrete throughout? It doesn't look that way to me though..

I offered the lm317/337 stuff as example because even those are a bit better than the 78xx parts we usually use but I now see that we are headed above and beyond these! I'll dig through some resources and find some circuits that I've come across and maybe we could find something useful.. I've been using a linear PSU in some equipment driven by a Vreg in a preregulator fashion driving some power BJTs. I'll see if that might stoke some ideas. You mentioned higher rails.. what's the target Voltage and current? any ideas on any other specs you'd like to see from the PSU yet?

I'm very interested in the SiC parts too.

I like your idea on the toroid and resistors. I've used similar setups for motor drives.. frequently using low value, high power resistors to kill spikes in current from startup and flyback and using overkill high VA trafos( i'll only use a toroid..).. it works like a charm though, and is still cheaper in the long run than trying to get around the noise and loss after the fact.





:thumb:

EDIT: http://www.national.com/ms/LB/LB-15.pdf an interesting but dated paper on high stability regulator circuits.. lots of expensive precision resistors needed... :?

 

[CJ]

There is a lot on the Gordon spec sheet at the site.
The unit is radio controlled so they can put the amp near the singer, but control the gain from the control room.

I talked with Grant for about an hour because nobody was stopping at his booth. It was just a banner, and a preamp.

There is a micro controller on board that does some control work on the fly. And a lot of other support chips, like Brad said. I think all those relays have to do with the power supply. Maybe he is changing supply rails to change gain or some weird thing. I know he is too smart to put relays in the signal path.
There is also ramped phantom.

That board is all one piece so it has to be hand soldered. There is too much flex for a wave machine. You can get the amp with an upgrade that includes a teflon board and caps and a few other things. He has all his caps custom made.

Asisde from the spec sheet, there isn't much more he could tell you without giving away secrets I am afraid.

 

[bcarso]

Wow radio controlled as well! At this rate maybe this WILL get up and get you a beer!

I hope this guy is successful---sounds like a great talent.

Perhaps he needs some marketing hype ;-). Must be some way to work the designer's name in to it. Probably most here are too young to remember the ads for Grant's scotch, that were variations of "As long as you're up, get me a Grant's." That was used for other gags about research grant money (As long as you're up, get me a grant) etc.

Let's see: "Grant Carpenter's hand-crafted preamp gave me a woody?" No, probably not...would appeal to some but repulse others :grin:

What was that line Dr. Morbius uses in Forbidden Planet when he shows the engine of the Krell's power source: "One does not look on the face of the Gorgon [Gordon] and live." Interesting---that movie is mentioned as having influenced John Carpenter too. (bring up Van Cleave's Twilight Zone music here)

Svart, I'm not sure of the rails yet but probably at least +/- 35V at this point.