Amek Angela Mic Preamp Transistors

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

pinebox

Well-known member
Joined
Mar 11, 2015
Messages
117
Hello, my friend has an Amek I am refurbishing for him, and he has a lot of variance in microphone level across the console measured at the insert send right after the mic pre. I am thinking the transistors are suspect, since its design requires five ZTX214s on each side of the signal. He has one module that was reworked by Jim Williams, part of which was removing the ten ZT214s and replacing them with two 2sa1084s. He was told to just copy these mods but these 1084s are long out of production. I found another thread discussing the KSA992B as a possible replacement in a similar Amek preamp. This is more into preamp design than I am familiar with, would this be a drop in replacement or would I need to recalculate R8 and R9 to a new value. (I am assuming that these were changed on the JW mod looking at the board, I would use this updated value going forward) Any advice on what to read to learn about this would be great, thanks!
 

Attachments

  • Screen Shot 2022-07-21 at 11.05.25 AM.png
    Screen Shot 2022-07-21 at 11.05.25 AM.png
    477.7 KB · Views: 10
  • IMG_0136.JPG
    IMG_0136.JPG
    128.7 KB · Views: 7
  • IMG_0137.JPG
    IMG_0137.JPG
    124.2 KB · Views: 7
R8 and R9 appear to just be input biasing resistors (allow for base current flow into the input transistors). They just set the input impedance of the pre-amp (along with the phantom power resistors).
I found an old ZTX214 datasheet which said it was intended to be compatible with BC214. Have you tried using those in place?

Is the audio distorting? If the differences are only in level then bad transistors would not be my first suspicion.
 
R8 and R9 appear to just be input biasing resistors (allow for base current flow into the input transistors). They just set the input impedance of the pre-amp (along with the phantom power resistors).
I found an old ZTX214 datasheet which said it was intended to be compatible with BC214. Have you tried using those in place?

Is the audio distorting? If the differences are only in level then bad transistors would not be my first suspicion.

Thank you, I just looked at 5 transistors cascading into each other and figured any early drift would probably be amplified by the end of the line. I dont hear any distortion at regular volumes, but I have not measured THD at all.


BIG VARIATIONS IN THOSE MOST OFTEN IS ABOUT ESR IN THE 2200Uf/6.3v IN SERIES WITH GAIN POT

(sorry about yelling, caps-lock on but I'm not retyping)
Interesting. I will swap two and see if levels follow. Thank you. Shout it from the rooftops.
 
For more information about this micpre, dig out Paul Buff's TransAmp application note. The Amek is very, er, similar.
ZTX trannys are used because they are 1A large geometry transistors., and are thus relatively quiet for the price. This was the point of the TransAmp. Buff used regular large G transistors; (complimentary pairs in the TA series for VCAs). You could use any ZTX 550/551/750/751. BC560? (PNP things).
Back then, using multiple (lowish) noise trannys was cheaper than 2 matched expensive, esoteric, one manufacturer only, hard-to-get ones. (And other posts here will attest to the absence of those transistors now).
At the time, the LM394 had only just come out, and was expensive back then! Multiple regular trannys seemed very attractive.

Is the JW modded one noticeably different, ie better, than the others, and how is this being determined?
If JW changed R8/9, change to those values on a standard Amek one, and see what happens and see if it makes a toss of difference.
10 trannys per mic pre is a lot to change for possibly minimal improvement. It's not as if this is a lousy micpre, it's very well thought of as the Amek2500.
For noise and distortion and mic pre discussions, dig out the Rec Eng/Prod articles by Paul Buff.
 
Now that I could be bothered to look up the ZTX214 datasheet, it's an version of the BC214 standard low-noise tranny. This does not pooh-pooh the large geometry discussions.
R8/9 are 56K on the 2500 schematics; what did JW change them to? Does not seem wildly critical to me!
 
Using multiple transistors IN PARALLEL (not cascaded) was one approach to getting low noise when fed from low impedance mics. the same circuit will WORK with only one pair, or (if you have 50 pairs in parallel (inside a LM394), the results are simply a variation on the theme and hand soldering 100 transistors would take a lot of time, effort and board space. You can only build gear with what is available and at the time the circuits were originally built the ZTX214 were 'it'. Other preamp manufacturers either used aternative transistors or different designs altogether. Some transistors are better (or worse) in terms of distortion over the voltage and current range they were used at, which could be observed by looking at their characteristic curves. The terms 'low noise' as used by transistor manufacturers is often misleading because it all depends on the current range you are operating them at, and the input and output impedances the transistor 'sees' which can be very significant. If you go to extremes and quote mic preamp noise figures with a very low input termination, the 'numbers' can look great but then most real world mics (and the wiring from the mic through to the transistors in the preamp) are probably greater than is 'optimal'.
Noise quoted when the input is shorted can look great but no mics have zero output impedance, or at least would have no signal either.
Looking at the pictures in the original post I can see a number of original capacitors from when the desk was built so the FIRST job that needs doing is a full recap of the whole desk rather than messing around with some transistors. If you measured the ESR (at a range of frequencies) you will find that none of the capacitors are anything like what a brand new capacitor of ANY make will show.
 
BIG VARIATIONS IN THOSE MOST OFTEN IS ABOUT ESR IN THE 2200Uf/6.3v IN SERIES WITH GAIN POT

(sorry about yelling, caps-lock on but I'm not retyping)
In addition to the above I would also look at the gain pot (R18?) if the level differences are near maximum gain.
The hop-off and wiper resistance was probably around 2-3Ω when the pots were new.
That stray resistance is in series with the 10Ω gain limit resistor.

On a console that old I would perform level measurement comparisons between channels at 100 Hz or less - there are likely many bad capacitors.

I don't think input transistors are the OP's particular problem but:

A single pair of ZTX951 is an excellent PNP replacement for the existing paralleled input transistors.
I've measured them and they are quieter than the 2SB737.
The ZTX951 rbb is <2Ω maybe 1.5Ω.
I haven't used many of the ZTX951 but I've used hundreds of the NPN ZTX851 and found them to be very consistent.
There are other higher-gain ZTX-series that are also outstanding low-noise performers.
 
This schemo would be much clearer if the ZTX214 were represented with a proper PNP symbol and the correct orientation of collector and emitter.
Q9 and Q10, the emitter current sources, appear to have the CE terminals reversed.
(The emitter arrows on all of them aren't very wide either.)
Is that correct?
The way it's drawn looked odd to me as well.

Q1-Q8 appear correct.
I do like the way the "draftsperson" arranged Q1-Q8, the collector loads and the FB and Rgain elements on the drawing.
It's a little different than how we're used to seeing the front end drawn.
 
re: the schematic.
Valley People Dynamite has the circuit in a more viewable format, given that it is pretty much the same as the TransAmp version. It shows much more the long-tailed pair with the parallel trannys feeding the 2 opamps.
Also dig out the Graham Cohen stuff on this topology. It's what Google's for.
Also Douglas Self's "Small Signal (can't remember)" book, goes into this type of circuit as well.

Conclusion.
General concensus is that you are wasting your time pixxing about with the trannys when there is an awful ot more you should be pixxing about with.
If you have not done a local circuit recap recently, you are chasing your tail.
 
re: the schematic.
Valley People Dynamite has the circuit in a more viewable format, given that it is pretty much the same as the TransAmp version. It shows much more the long-tailed pair with the parallel trannys feeding the 2 opamps.
Also dig out the Graham Cohen stuff on this topology. It's what Google's for.
Also Douglas Self's "Small Signal (can't remember)" book, goes into this type of circuit as well.

Conclusion.
General concensus is that you are wasting your time pixxing about with the trannys when there is an awful ot more you should be pixxing about with.
If you have not done a local circuit recap recently, you are chasing your tail.
Yup, a search here about "Cohen" topology, and "Transamp" should reveal much discussion. I even wrote about the Transamp in my 1980 console article.

I will second the suspicion about gain pot and series capacitor.

JR
 
This is one of my favorite active mic preamp drawings that was given to me by the founder of Benchmark, Al Burdick, when he was operating out of his garage in Garland, Texas. Al had just discovered the 2SB737 and was really pumped about using it. I wrote some of the original documents for his products.

BenchMark_Mic_Pre.JPG
 
Confirming that Q9 and Q10 have reversed collector emitter terminals.
I agree they could use a better symbol.

Looks like Al was in agreement with you on how he drew his PNPs.
 
I've recently discovered the CMPT5087 and 5088 for low noise and I've been happy with their performance.
 
This is one of my favorite active mic preamp drawings that was given to me by the founder of Benchmark, Al Burdick, when he was operating out of his garage in Garland, Texas. Al had just discovered the 2SB737 and was really pumped about using it. I wrote some of the original documents for his products.

BenchMark_Mic_Pre.JPG
I built exactly that one, only with 2sd786 and "upside down"...
Thanks for posting it before!
 
Back
Top