AMEK 2500 mic pre: how does it work?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

Hi Olaf
welcome

no answer to your question right now but I'll investigate the PDF's and see if anything comes to mind
 
Thank you for making something that is good better!

will try to see if i can use it in my own design.
 
You might want to build the more refined Green look at the green very close and ask yourself what are the 22uf caps doing around the input transistors??? Very clever circuit.

I believe the 2500 is an older design.
 
Guys building this would be a waste of time compared to the green why build something that is an infearer(sp) design. Then you need to match 8 transistors a pre etc......

Drop matched 2n4401s in a green I an going to play with the current in the input pair with on eof the current setting Rs.

The green has a bootstrapped input pair. The green /9098/M type is a real nice design IMO.
 
It should be a fine pre. I meant my posts in a helpfull way. You might be able to use just 2 input transistors or a matched set if you build any more.

R11 and R12 ajust the current for the input pairs
 
Regarding the difference amp at the o/p, here's my take.

The same gain could be realized by leaving the 18k's and the 10k feedback removing the 5k6 (r18) and making r19 10k. The gain either way is still 10k/18k or approx .556. So this begs the question why the extra 5k6 and change r19 to 3k6?

So I am thinking (still up for new ideas here) that this will improve differential input impedance. One problem with the regular 4 resistor diff amp is that you can either configure it for matching common mode input impedance or matching differential mode input impedance, but not both (the New Audio Cyclopedia goes over this in the chapter on Consoles and Systems). 10k in parallel with the 5k6 gives 3k6 . I did some quick calcs and sims and the differential in imp. does inprove but not drastically (from about 28k and 13k with the 4 res config to 21k and 15k with the 2500 config) so I cannot say wether this was there design objective or not. Just my initial conclusion.

Any others thoughts out there?

As far as the overall gain with the circuit as it is, it would be

Av = ((r8+r11)/(r3+r26)+1) * (r10/r21) where r10//r18 = r19

Thats my take (for now) anyhow.
Brian
 
I'm veruy interested on the sound of it, please let me know if you like it!

:sam:
Fabio
 
> the 2500 has a second feedback loop trough a 1M resitor (R7 and R23 in the original schematic). What are these good for?

The transistor collectors sit at about 7V DC. The designer choose to AC-couple that DC level away with C6/C8 and R7/R23, so U2's inputs could sit near DC ground. No significant AC feedback here.

> that the output op-amp seems to have different amplification factors for the - and + signals!?

Gain is the same. R10||R18=R19. I believe the reason is to reduce the gain inside the feedback loop around U1. The 5534 can be compensated for unity gain, or de-compensated for better speed if the feedback loop loss is greater than 1. With these values the feedback loss is over 3 (about 3.77) so the 5534's compensation can be much lighter.

I don't see how this is "infearer" (inferior). It isn't how I would sketch it, and seems more complicated than really needed, but the proof i in the listening. Some of those mystery design decisions may be based on listening tests. (Or just adapted from other designs, or to use up excess resistors....)
 
Okay fellas,

Feel free to tell me if I am nuts but isn't this a prime circuit to use a pair of Toshiba SK386's FETS ahead of those op amps??

I've got a box of those 'cute' buggers in my shop somewhere ...

Best Regards,
Carl huff
 
inferior in that the 2500 uses alot more matched parts.

I do like the balanced output drive used in the new 2500 build better than the green because you don't have to worry about grounding out the - output like you do with the green

The Motzart /green /9098 uses bootstapping on the input pair. C4 and C5 on the schematic of the motzart from the link at group DIY. IMO C4 and C5 are backwards on the schematic.

I am still working out in my mind what the bootstrapping gives in the green besides a Higher input Z because the in phase signal from the emitter of Q1 and Q2 are feed to the 10K, 47K nodes.

The bootstapping could have something to do with noise and CMRR I think.
 
F.F. has a very nice microphone pre design using fets at his site.
 
[quote author="Gus"]...C4 and C5 on the schematic of the motzart from the link at group DIY. IMO C4 and C5 are backwards on the schematic...[/quote]I'd try the Panasonic bipolar SU series here.
 
no need it will have a .6v of so voltage because of the base to emitter voltage. I would think a polar would be better.
 
This is my first post here, but I have been following discussions here and on the old forum for quite a while. This is a great place.


Thanks a lot for the B-Pre schemo. I have one, too, and I quite like it.

Theres one minor mistake I found: the pot next to the gain control is actually 1000 u. You might have mistaken it for 100u, because it is very low voltage and therefore relatively small.

The 2SB737 is obsolete, unfortunately. I talked to a designer for another German audio company, who said he was quite fond of that particular transistor. The usual replacement is 2SA1084.

On my unit I paralleled the electrolytics with .1 u WIMA polyester caps, which improved the sound a bit (more transparent).
Thanks for the info on the diodes. So changing them for a better type would improve noise? Would be a cheap and simple mod. My unit is pretty low noise already.

Another thing that I'm wondering about is the opamps. Back then B used the BE 027 and BE037 dual and quad opamps. In an old broshure they even advertised them as higher quality replacements for standard opamps such as TL07x and (I think) even NE5532. I still don't know what exactly those opamps were. I asked B several times, but they never replied. Given the high quality claim I always assumed they must be significantly better than a lame LM348. I know that there is a B-parts list somewhere on the web that suggests LM348 or something similar, but I don't think that's what they actually were. I have a couple of Composers that use different opamps depending on when they were built: BE027, BA4558, NJM4580. The by far worst sounding of them is the BA4558 one; its sound improved dramatically when I swapped it for a NE5532. The 4580 is lowest noise, but I don't really like its sound. The BE027 one sounds fine to me.
Bottom line: I'm hesitant to replace the BE opamps in my Ultragain 2000, because you can't get the original part anymore, and I'm not sure if anything else is really an improvement. Hey, maybe I'll build a pre according to your schematic for experiments.
 
[quote author="olafmatt"]I've installed some resistors to operate the opamps in 'forced Class-A' as it is done in Focusrite gear.

Just connect a 10k resistor between the output of the diff op-amp to the positive supply rail. Another one goes from the phase-reverse opamp to the pos. rail and probably a third one in high-pass filter stage (not on the schematic).
So far it sounds a bit more transparent at low levels and I haven't found anything negative about it (check whether the power supply gets too hot!).
[/quote]

Does anyone have a schematic (or could draw something up quickly) to illustrate this please? I read an interview with The Man Himself where talked about biasing the opamps this way, but I can't visualise it.

Actually this would be a good addition to Tim Ryan's electronic learning page.

I must apologise for my lack of electronic knowledge actually, it's a bit hard trying to wrap your head around this sorta gear. Even harder when your old man designs satellite antennas for a living...
 
Thanks a lot, Olaf, for the schemo and also the thing about forced class A. Is there a particular reason why you reduced the input impedance? The rated impedance according to the manual is 3k btw.

I still wish I could get more info on the BE-opamps, but I have no idea how. B-officials usually aren't too talkative. To be honest, it might be a good idea to change direct references to "Behringer" or "Ultragain". B never give out schematics, not even for obsolete products. So I suppose they wouldn't like to see schemos on the web. Although I don't think it's illegal as long as it's your own drawing, not the manufacturer's own.
 
Back
Top