Volume pot between two NE5534 pre, what is happening?

Hello there,

I recently experimented with my NE5534 pre amp, and found very interesting result. I would like to hear your opinion.

Here's what I did. NE5534 is not regarded as high end op amp here, but praised by many as decent. I set each op amp as 4x gain non-inverting amp.

Exp1. Added volume pot in front of two serial connected op amp.
Exp2. Added volume pot in between two op amps.
Exp3. Added volume pot behind two serial connected op amp.

Not knowing exactly how much it will affect, I hoped Exp2 gives least effect of volume pot in the sound chain. And, to my surprise, Exp2 actually sounds most transparent to me. Oscilloscope measurement of square waves from 20 Hz - 20 kHz seems very similar to all 3, though.

Does anyone have idea why, if any, Exp2 can sound better than the other two configurations?

Thanks,
Doug
 
it is a design choice, not even by the company, it depends on the designer unless told otherwise.
some like pioneer 1250 used a x1 buffer before vol/bal controls. next model 1280 they did away with the buffer, who knows why, to save a $ on some cheap parts?
I like to buffer the signal before the volume control, small amount of gain x2. that way the sources see the same load Z and it works better with higher Z sources.
pre-amp gain. in the past the std was set at 150mV in (typical tape phono out) and 1V out (typical PA input to get full output)
any design should have a drawing showing signal flow and gain/att stages.
now that we have cd and other sources that put out all kinds of different full scale levels, the pre-amp design can be different depending on the sources that you want to accept. a cd puts out 2V FS, too much for a PA connection, so you only need atten not gain.
Doug Self is the master at using the ole 5534/5532, review his designs.
 
why would it be surprising that exp2 sounded the best ?

Not knowing the reason, I though all 3 configurations must sound the same.
As other person said, maybe Exp2 setup is somewhat better because it gives constant input and output impedance to the source and power amplifier.

But, does volume pot become variable load to the source? Or does it become variable source to the power amplifier?
 

rayma

Member
2011-04-29 8:37 pm
Not knowing the reason, I though all 3 configurations must sound the same.

You didn't say what value of pot you used, or the source component output impedance, or the amplifier input impedance.

Placing the pot after both op amps will generate more distortion because of wiper loading by the amp, and also possibly
from overloading in the second op amp due to less headroom. High frequencies will be more rolled off due to higher Rout.

With the volume pot first, loading of the source is maximized.
 
Last edited:

rayma

Member
2011-04-29 8:37 pm
The pot I used is Alps blue velvet 50k. The source is Sony SACD 333es, and the power amplifier is Odyssey Stratos stereo.

With the pot last, and a 2V CD input x 16 gain, you will have overload in the second op amp.

With the pot last, a 50k pot will have as much as 12.5k output impedance. This could roll off the highs.
 
Last edited:
With the pot last, and a 2V CD input x 16 gain, you will have overload in the second op amp.

With the pot last, a 50k pot will have as much as 12.5k output impedance. This could roll off the highs.

That makes sense. Now I reduced the opamp gain to 2x, but still there is 2V x 2 x 2 = 8V at the output when pot last.

Thanks for the opinion.
 
A gain of 4 is only a little bit more than the minimum gain of 3times.
The 3times gain lower limit will probably have a phase margin of 45degrees, or a bit more.

A gain reset to 2times will perform badly.
At 2times gain the amp will be close to oscillating and will oscilate with a bit of capacitance on the output.
At 3times gain the amp will show severe overshoot on faster signals.
At 4times gain the amplifer will show overshoot that is clearly audible as not being part of the original signal.