High value / standard datasheet resistor
Example LM4766
Rin = 47k
Rf = 20k
Ri = 1k
Low value :
Rin = 4.7k or 10k or 15k
Rf = 2.2k
Ri = 100R
If I use low value resistor I will have to increase capacitor values
Cin = 10uf
Ci = 220uf
What is the advantage and disadvantage for both ?
My experience : Low value sounds good but Amp running hot. ( or may be another reason )
Example LM4766
Rin = 47k
Rf = 20k
Ri = 1k
Low value :
Rin = 4.7k or 10k or 15k
Rf = 2.2k
Ri = 100R
If I use low value resistor I will have to increase capacitor values
Cin = 10uf
Ci = 220uf
What is the advantage and disadvantage for both ?
My experience : Low value sounds good but Amp running hot. ( or may be another reason )
IMO
High values of resistance will be affected by stray capacitances more than low values. This may affect higher frequency phase and/or response. It may affect amplifier stability.
This may or may not be significant.
The cap in the low value set (high Ci) will have a lower self resonance frequency. (at which it turns inductive in nature)
Again, this may or may not be significant and can be compensated with smaller values placed in parallel with the 220uF.
In either case this amplifier is able to drive the feedback network. I do not believe that the low value set is causing the heat issue.
Oscilloscope if you can.
High values of resistance will be affected by stray capacitances more than low values. This may affect higher frequency phase and/or response. It may affect amplifier stability.
This may or may not be significant.
The cap in the low value set (high Ci) will have a lower self resonance frequency. (at which it turns inductive in nature)
Again, this may or may not be significant and can be compensated with smaller values placed in parallel with the 220uF.
In either case this amplifier is able to drive the feedback network. I do not believe that the low value set is causing the heat issue.
Oscilloscope if you can.
High values => stray capacitance problems (as DUG said).
Low values => driving problems (although not with a power amp), possible heating effects in resistor, larger cap values means either lower quality dielectric and/or physically large components.
All electronic design is a compromise, so I smile every time I see a thread headed "ultimate/best ever/ideal" this or that.
Low values => driving problems (although not with a power amp), possible heating effects in resistor, larger cap values means either lower quality dielectric and/or physically large components.
All electronic design is a compromise, so I smile every time I see a thread headed "ultimate/best ever/ideal" this or that.
Another factor...Lower values of feedback resistors will contribute less noise.
A 1 kOhm resistor has noise of about 4 nV/rt-Hz.
A 100 Ohm resistor has noise of 1.26 nV/rt-Hz
Whether this matters depends upon how quiet your opamp is, and the size of the signals being amplified.
Update My Dynaco
Akitika GT-101
A 1 kOhm resistor has noise of about 4 nV/rt-Hz.
A 100 Ohm resistor has noise of 1.26 nV/rt-Hz
Whether this matters depends upon how quiet your opamp is, and the size of the signals being amplified.
Update My Dynaco
Akitika GT-101
using Dj's data, if the opamp has an input referred noise of 4nV/root Hz and the noisy resistor also has 4nV/root Hz, then does the combined input referred noise amount to 4*sqrt(2) or 4+4 nV/root Hz? , i.e. 5.7nV/root Hz or 8nV/root Hz?
I have tried both configurations
Standard or High Value creates low heat to chip
and
Low value runs hot but I like sound.
@AndrewT : Any solutions using low value resistor and low heat.
low value should not make the amplifier run hot.
A 1k0 feedback resistor on a 8ohms capable amplifier has a miniscule effect on the total current passed through the amplifier.
But an opamp designed to drive a 2k0 load and then asking it to also drive a 1k0 feedback resistor is a very different design problem.
One must consider the "whole" load on the output of the amplifier.
That whole load is the Rload you are driving, plus the feedback resistor, plus a cable capacitance, and maybe plus a feedback capacitor.
All these appear as parallel loads and the amplifier has to drive all of them.
Your "hot" chip is either due to a mistake in the build, or due to oscillation (another mistake in the build), or due to inadequate heatsink (yet another builder mistake).
Don't blame the chipamp.
A 1k0 feedback resistor on a 8ohms capable amplifier has a miniscule effect on the total current passed through the amplifier.
But an opamp designed to drive a 2k0 load and then asking it to also drive a 1k0 feedback resistor is a very different design problem.
One must consider the "whole" load on the output of the amplifier.
That whole load is the Rload you are driving, plus the feedback resistor, plus a cable capacitance, and maybe plus a feedback capacitor.
All these appear as parallel loads and the amplifier has to drive all of them.
Your "hot" chip is either due to a mistake in the build, or due to oscillation (another mistake in the build), or due to inadequate heatsink (yet another builder mistake).
Don't blame the chipamp.
Last edited:
- Status
- Not open for further replies.