Fixup said:
R1 must be close to the op-amp, and the pot too.
I'm not a fan at all of long wires after volume pots, and a couple of tens of cm is long. IMO.
R1 is there just for protection if the pot fails. It's not so frequent to happen, but if the pot becomes 'open' (bad contact, infinite impedance), the input of the op-amp will not be referenced to ground and you will have DC on the output.
It is a rare occasion, but can happen. If you have a faulty source component that outputs several volts of DC, you can kill a volume pot.
Any balancing resistor will be in parallel with the pot and change it's taper.
I've recently found, with higher value resistors anyway, that some chips can't be used with this circuit with no DC blocking input cap as a large DC offset is produced at their output. And this swings as the pot is moved.
I have a plan to insert a BUF04 or whatever, between pot and chip if necessary and if the sound quality of that and a better opamp is better than no buffer and a lesser chip that tolerates this circuit.
I've recently found, with higher value resistors anyway, that some chips can't be used with this circuit with no DC blocking input cap as a large DC offset is produced at their output. And this swings as the pot is moved.
I have a plan to insert a BUF04 or whatever, between pot and chip if necessary and if the sound quality of that and a better opamp is better than no buffer and a lesser chip that tolerates this circuit.
Fixup said:
That 100k resistor is there because it has a 10k pot before it. It's just for protection, and it must be a higher value than the pot, read obove.
I would not use 100k as input impedance if there wasn't a volume pot.
Fixup said:
Mistakes?
Ha!
Of course, if you use an input cap (and kill the sound of the pre), you must have a resistor after it.
We agree, then.
But even without an input cap, I use a large value resistor after the pot. It doesn't make any harm (soundwise), and it's a protection.
Look, use it if you want.
Or do I have to 'xplain that you should use a fuse on your PSU?
IanAS said:
No, it won't.
The resistor is 10x greater in value than the pot.
IanAS said:
For bipolar input op-amps that is true.
And even more serious if you don't match impedances.
But the circuit I posted is for the OPA627.
That's the problem, you guys think that the op-amps are all the same and you can change them on a circuit at your will.
You have to know what you are doing
Sometimes you need to change everything around.
I agree with Carlos here. R1 will give the op amp a well defined input, ground potential, if the wiper of the pot has bad contact and goes open. Of course it must be large enough so it doesn't affect the taper too much. I would perhaps consider an even higher value, but it is a trade off what to settle for, since a large value can cause too high DC offset in the case of bipolar inputs.
Fixup said:
For bipolar input op amps, yes, that may be the case, but hardly for FET input ones.
Yes, but less with the resistor than without. There is quite a difference between 100 k and infinity.
The resistor will actually lower the voltage noise since it is in parallel with the impedance already present at the input, thus lowering the total source impedance. The effect will be negligible, though, since the resistor has such a large value. It could have an adverse effect on current noise in some cases, though.
Fixup, I have already pointed out the role of the 100k resistor.
I also said I would not use 100k as input impedance. There's a volume pot there.
Christer also correctly said that the resistor is in parallel with the pot.
If the pot fails, you won't have DC on the output, as the OPA627 is a fet input op-amp, which is 'immune' to mismatched impedances.
And that schematic is for the OPA627.
In the case of a bipolar input op-amp, the 100k resistor would not prevent DC on the output, but it would be much smaller than without it.
The resistor is there for protection, but it can't be too low, or it will affect the tapper of the pot.
Exactly what happens to those 'gaincloners' who use a 50k pot in front of a bipolar power op-amp (LM3886, etc.), with 22k from NI to ground.
I also said I would not use 100k as input impedance. There's a volume pot there.
Christer also correctly said that the resistor is in parallel with the pot.
If the pot fails, you won't have DC on the output, as the OPA627 is a fet input op-amp, which is 'immune' to mismatched impedances.
And that schematic is for the OPA627.
In the case of a bipolar input op-amp, the 100k resistor would not prevent DC on the output, but it would be much smaller than without it.
The resistor is there for protection, but it can't be too low, or it will affect the tapper of the pot.
Exactly what happens to those 'gaincloners' who use a 50k pot in front of a bipolar power op-amp (LM3886, etc.), with 22k from NI to ground.
With those car amps don't they use an oscilator to break up the +12DC into AC that can then be transformed to the required higher voltage?
If so some decent capacitors in the right places might help. Black Gates maybe. In the switch mode supply for DVD players there's usually one electrolytic in a critical place that needs upgrading and also put X7R dipped ceramic 0.1 or 0.47 mic caps across all the electrolytics and from the pins of the chips to earth.
It's probably only a UK thing this next comment but Maplin are selling the +12V (or so) 1 Farad 'power' capacitor for £39 at the moment. Was £100 or so last year I think.
If so some decent capacitors in the right places might help. Black Gates maybe. In the switch mode supply for DVD players there's usually one electrolytic in a critical place that needs upgrading and also put X7R dipped ceramic 0.1 or 0.47 mic caps across all the electrolytics and from the pins of the chips to earth.
It's probably only a UK thing this next comment but Maplin are selling the +12V (or so) 1 Farad 'power' capacitor for £39 at the moment. Was £100 or so last year I think.
You should have the power rails bypassed as close to the opamp as possible. I'd suggest 100nf + 100uf on each rail and also a 100nf between + and - rails (directly over the chip for this one).
Carlos and others have much more experience here than me though. There's a lot of info on bypassing here in this and other threads; you might need to dig a bit.
Carlos and others have much more experience here than me though. There's a lot of info on bypassing here in this and other threads; you might need to dig a bit.
Having said X7R multilayer ceramic, i think it's felt they can be microphonic, and polyprop 0.1 mic might be better, but a lot larger.
The ceramics are good for geting rid of the high frequency noise as their indcutance is very low. Three is parallel is a lot better than one, on my scope anyway. So should be good in the switched mode supply and just after it.
A few posts up i think Carlos mantioned a 390n across the chip.
The ceramics are good for geting rid of the high frequency noise as their indcutance is very low. Three is parallel is a lot better than one, on my scope anyway. So should be good in the switched mode supply and just after it.
A few posts up i think Carlos mantioned a 390n across the chip.
Sorry guys, all this sounds like Greek to me. I am a noob and wouldn't have an idea where to begin. You guys think that I shouldn't just swap the OPA 2604 with the OPA 2132 or 2134? Originally, t came with an OPA 7??. I have forgotten the number, and the manufacturer suggested to switch to the OPA 2604. I did, it's nice but found the mid section rather glaring. Wanted to see if there were other chips that would make it sound better.
Thanks.
I am a noob and wouldn't have an idea where to begin. You guys think that I shouldn't just swap the OPA 2604 with the OPA 2132 or 2134? Originally, t came with an OPA 7??. I have forgotten the number, and the manufacturer suggested to switch to the OPA 2604. I did, it's nice but found the mid section rather glaring. Wanted to see if there were other chips that would make it sound better.Thanks.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.