OPA2134 Preamplifier

[maouna]

Hi. Im building a stereo preamplifier based on OPA2134 .

I would like to know whats the influence of feedback resistor value R19,R18 on sound.

On the schematic i chose them to be typical 10k and 33k. If i choose them to be 1k and 3.3k by keeping the same ratio,how the sound would be affected?

[SY]

Changing those values will have almost no effect on the sound. It will load the output of the opamp a bit more, but still well within its drive capabilities.

[Michael Bean]

Maona,

If the circuit is a competent design, the physical layout is done well, and components used are from known and reliable vendors, none of the parts should affect the "sound" of the circuit (I do not use expensive "boutique" parts and never will.)
Using lower values in the feedback network will lower the noise, but using high "R" values at the non-inverting input and low "R" values at the inverting input can cause excessive offset voltage at the output. Either the output should be AC coupled, or feedback network should be reconfigured for equal resistance at both of the opamp inputs and AC coupled. The 150k at the non-inverting input is too high. If you change that to something like 50k it will lower noise and maybe even get the offset voltage under control while still giving an input fc under 2Hz.

And I hope you aren't planning to actually use +/- 18 volts to power that chip, life expectancy would be short.

Mike

[maouna]

thank you both for replying.so these are the changes im going to apply

[counter culture]

The 150k to ground on the non-inverting input is shunted by the source impedance; since this is drawn as a voltage generator, it's not possible to say what the Johnson noise will be or what the impedance seen by this input is. Typically, however, the reduction in noise figure of the stage due to lowering the feedback loop resistance in an arrangement like this by a factor of 10 will be ~1dB and you could reduce the values to 150R/450R since the load is 100k and the OPA134 will drive 600R. It's necessary to perform a more accurate calculation or do measurements to reliably establish the exact numbers.

This will be insignificant (inaudible) in the majority of cases, but a couple of dB lower noise figure can be significant in some circumstances. This is an impact on noise (hiss), other aspects of the sound will be unaffected to all intents and purposes.

[Mooly]

Originally Posted by Michael Bean

.............And I hope you aren't planning to actually use +/- 18 volts to power that chip, life expectancy would be short.

Mike

That's within the recommended range

----------------------------------------------------------------------------------------------------

DC offset is minimal as this is a FET opamp so there are no worries on unbalanced bias "currents".

[Mooly]

Also if you are really interested in sound quality and tweaks then have a read at this thread and in particular some of the links I post too,

A Different Opamp Compensation Technique.

[Michael Bean]

Mooly,

So you recommend running components right up to thier maximum ratings? I feel much better leaving some safety margin.

Mike

[Mooly]

Originally Posted by Michael Bean

Mooly,

So you recommend running components right up to thier maximum ratings? I feel much better leaving some safety margin.

Mike

I would be happy at the 18 volt limit for these opamps providing the PSU was well behaved. They are designed to work at that voltage 24/7. Its not like running a transistor at max Vce where other factors come into play.

To quote from the data sheet,
"OPA134 op amps are easy to use and free from phase
inversion and overload problems often found in common
FET-input op amps. They can be operated from
±2.5V to ±18V power supplies."

[AndrewT]

and the 134 will run cooler than the 2134.