DOA AMEK 2500 preamp

[grrrunge]

Two questions though:
[list type=decimal]
[*]Someone mentioned that input caps are usually of much higher capacitance. How come? As far as i can see, the two 10µ in series with a 56K to GND forms a high pass filter with a corner frequency of 0.6hZ. Lowering the value of the input cap to 1µ raises the corner frequency to 2.8hZ - that still seems reasonable. A 470nF cap woud raise the corner frequency to 6hZ. Is there any reason why i should use a 10µF or higher value electro instead of smaller film caps?
[*]What's the function of the (ridiculously huge) 2200µF capacitor in the gain control? It takes up quite a bit of real estate Can i replace it with a more sensible value - say 220µF - without f**king anything up soundwise?
[/list]

I should probably note that i mostly record rock/metal stuff, and rarely use low frequency content below 20hZ in the mix anyway, so recording content around 0.1hZ doesn't make much sense to me. 8)

 

[bruce0]

Two questions though:
[list type=decimal]
[*]Someone mentioned that input caps are usually of much higher capacitance. How come? As far as i can see, the two 10µ in series with a 56K to GND forms a high pass filter with a corner frequency of 0.6hZ. Lowering the value of the input cap to 1µ raises the corner frequency to 2.8hZ - that still seems reasonable. A 470nF cap woud raise the corner frequency to 6hZ. Is there any reason why i should use a 10µF or higher value electro instead of smaller film caps?
[*]What's the function of the (ridiculously huge) 2200µF capacitor in the gain control? It takes up quite a bit of real estate Can i replace it with a more sensible value - say 220µF - without f**king anything up soundwise?
[/list]

I should probably note that i mostly record rock/metal stuff, and rarely use low frequency content below 20hZ in the mix anyway, so recording content around 0.1hZ doesn't make much sense to me. 8)

1) That was me... I see that the 56K resistors (sorry) in parallel with 4 PNP bases (and I don't see how it is biased and thus don't really understand how to understand the impedance of that combo, I can't even tell if it is common collector or common emitter ( I think common emitter which can be very low impedance) ) and yes a corner frequency of .6hz is fine.  A lower higher corner would be fine too.  Remember that the "corner" frequency is the 3dB down point, and the rolloff starts earlier, so sure you could use smaller caps but if you push up near the audio range you will effect the very low frequency.  I would not lower the cap value below the 5uf unless I understood the total input impedance.

2) I think the 2200uF does need to be that large.  The impedance of that part of the circuit is very low (maybe it is the parallel emitters of the 5 PNP's which apparently (see ricardo's note) can get as low as 30 ohms in series with 10 ohm resistor.  I don't know how to calculate the exact impedance of that part of the circuit.) so that size cap makes sense.  Note that it can be very low voltage and so will be small.  I think lowering it as you mention will roll off the low frequencies.

 

[ricardo]

• • Someone mentioned that input caps are usually of much higher capacitance. How come? As far as i can see, the two 10µ in series with a 56K to GND forms a high pass filter with a corner frequency of 0.6hZ. Lowering the value of the input cap to 1µ raises the corner frequency to 2.8hZ - that still seems reasonable. A 470nF cap woud raise the corner frequency to 6hZ. Is there any reason why i should use a 10µF or higher value electro instead of smaller film caps?
  • What's the function of the (ridiculously huge) 2200µF capacitor in the gain control? It takes up quite a bit of real estate Can i replace it with a more sensible value - say 220µF - without f**king anything up soundwise?

  The choice of input caps is rather subtle.  You need to know the Current Noise Spectrum of the input stage.  eg  for THAT1510/12, the correct caps are 22u.  In most cases, the LF roll-off isn't a consideration.
  
  The THAT datasheets & application notes have good discussions of other important points in the choice .. to do with CMR.
  _____________________
  
  The 2200u cap forms a LF roll-off with the GAIN pot in the feedback.  It is huge cos you don't want too much LF roll-off at maximum gain when R3=10 is the relevant resistor.  It is truly EVIL.  Cohen doesn't have it which is one reason why the Greens & other lesser copies are inferior.
  
  Huge size (for both i/p & f/b caps) is EVIL
  _________________
  
  IMHO, if you want a clean preamp, using THAT 1510/12 with their datasheet & AN circuits will be much simpler and give you much better performance .. especially for noise.
  
  Even simpler is to buy Expat Audio's Eden modules which are based on TI 1na163 and fully tested.  THAT & Eden will approach 1nV/rtHz which none of the Greens will do.
  
  But don't let me stop you from playing.  It's all good learning.

 

[grrrunge]

So here's where I'm at currently. I'll start layouting tomorrow, and get a few boards fabbed from OSHpark. Fun's around the corner

 

[grrrunge]

First set of prototype boards on its way

 

[grrrunge]

Boards shipped out today. Exciting exciting

 

[grrrunge]

...and recieved today. Can't wait for exams to be over!

 

[grrrunge]

Nah. Couldn't wait! I had to populate one of these before i hit the sack
Can't wait to bring it to the lab, and see how it behaves....

 

[peterc]

The other thing to watch out for, apart from Freq Response, is Phase Shift.

A guideline is to have high and low freq limits tenfold that of the freq response to limit phase shifts in the 20-20kHz range

So, at 2 hz there might be a substantial phase shift in your circuit, but this will be minimised at 20Hz, same with the upper limit.

    Someone mentioned that input caps are usually of much higher capacitance. How come? As far as i can see, the two 10µ in series with a 56K to GND forms a high pass filter with a corner frequency of 0.6hZ. Lowering the value of the input cap to 1µ raises the corner frequency to 2.8hZ - that still seems reasonable. A 470nF cap woud raise the corner frequency to 6hZ. Is there any reason why i should use a 10µF or higher value electro instead of smaller film caps?
    What's the function of the (ridiculously huge) 2200µF capacitor in the gain control? It takes up quite a bit of real estate Can i replace it with a more sensible value - say 220µF - without f**king anything up soundwise?

 

[grrrunge]

I never got around to finishing this project, so while I'm waiting for the 1176 boards to arrive, this gets a makeover

I hear good things about this one as being what  the "Langley preamp" is based on. Can anyone confirm that?
Did anyone else ever think abut trying this one out with DOA's in it?

Fresh schematic:

 

[grrrunge]

Boards laid out Looking forward to test this bugger...

 

[pedroplanet]

does anybody knows the power consumption for each channel?

 

[grrrunge]

does anybody knows the power consumption for each channel?

I tried simulating one channel of this in LTspice with NE5534 op-amps. the power consumption came in short of 0.6W.
I'm sorry i havent been around much lately, to follow up on my projects here. Real life came in the way of all the fun stuff :/

 

[pstcho]

Hello guys

Great project

Did you finish it?

Works fine?