Hi
I'm assembling phono preamp based on op amps and stumbled upon weird situation (maybe just for me).
I'm using ne5534 for tests, non inverted input, ground resistor 200Ohm, feedback resistor 1k or 10k (switching gain, putting 1k in parallel to 10k).
And I was confused that with 1k (lower gain) there is a lot of audible noise (hiss, white noise etc) and with 10k it disappears!
Is that normal or is it faulty opamp (or for ne5534 is it common) or resistor or what?
I'm assembling phono preamp based on op amps and stumbled upon weird situation (maybe just for me).
I'm using ne5534 for tests, non inverted input, ground resistor 200Ohm, feedback resistor 1k or 10k (switching gain, putting 1k in parallel to 10k).
And I was confused that with 1k (lower gain) there is a lot of audible noise (hiss, white noise etc) and with 10k it disappears!
Is that normal or is it faulty opamp (or for ne5534 is it common) or resistor or what?
Maybe it oscillates with the 1 kohm connected, although an NE5534 should be stable when the gain is at least 3.
Do you have power supply decoupling right next to the NE5534?
Is the input wire very long?
Does the DC voltage at the output make sense?
If you have an oscilloscope with a 1:10 passive or an active probe, what does the output signal look like?
Do you have power supply decoupling right next to the NE5534?
Is the input wire very long?
Does the DC voltage at the output make sense?
If you have an oscilloscope with a 1:10 passive or an active probe, what does the output signal look like?
Decoupling 100n caps are close enough (few millimeters)
Tested without input wires, only 50k to ground.
I'll try to look with oscilloscope a bit later
Tested without input wires, only 50k to ground.
I'll try to look with oscilloscope a bit later
50k to ground on the input will create a lot of noise and a lot of output DC offset. Perhaps with the 10k it is saturating the output to one of the rails, thus suppressing any noise?
Worst case DC offset for 47k input resistor and 10k/200 ohm feedback is 3.5V for the 5534A. What supply voltages are you using?
Worst case DC offset for 47k input resistor and 10k/200 ohm feedback is 3.5V for the 5534A. What supply voltages are you using?
So it's oscillating for some reason, around 1.25 MHz.
Can you show a photo and/or schematic of your circuit?
Can you show a photo and/or schematic of your circuit?
On all my opamp circuits (especially the ones that I have upgraded to newer, faster opamps), I put a 22pF COG capacitor around the feedback resistor to eliminate any oscillation tendencies that may be brought on by older, poorer layout techniques that didn't have to cope with super-high bandwidth ICs. This limits the bandwidth of the circuits to about 220 KHz, but that is more than sufficient for any audio circuit. I also put a 100nF XLR capacitor right across power pins 4 and 8, on the solder side, for power supply bypass.
Do you have bulk decoupling on the PSU rails? Where are the ceramic 100nF caps? I only see film...
For a gain-of-at-least-three compensated op-amp, the attenuation of the feedback network has to be at least three times around the point where the loop gain becomes unity. With a capacitor shunting the feedback capacitor, that requirement isn't met, unless the capacitor is smaller than half the input capacitance of the op-amp.
Which is why layout gets important - stray capacitance between output and inputs exists already and can be a stability problem.
@MarrcelvdG: I do not see the input capacitance listed on the device's datasheet. How does one know this?
In general, when the capacitance is specified it is a general, one-size-fits-all value like 5pF.
The actual value will very much depend on the case style, PCB, etc.
The FB capacitor is useful to compensate for the input capacitance, as it eliminates FR peaking and overshoot, but it has to be correctly calibrated once the circuit is finalized.
The actual value will very much depend on the case style, PCB, etc.
The FB capacitor is useful to compensate for the input capacitance, as it eliminates FR peaking and overshoot, but it has to be correctly calibrated once the circuit is finalized.
Oscillation defeated!
I noticed that when I removed C7(18) oscillation disappeared and then I noticed 0.01uF in parallel to R13(26) 100k and it is not in original schematic, so I removed 0.01uF and put C7 back - no oscillation and no noise !
Maybe for the first time Chinese guys decided to put an extra component ))
I noticed that when I removed C7(18) oscillation disappeared and then I noticed 0.01uF in parallel to R13(26) 100k and it is not in original schematic, so I removed 0.01uF and put C7 back - no oscillation and no noise !
Maybe for the first time Chinese guys decided to put an extra component ))
You could measure it, guess it, add a capacitor from the negative input to ground or simply not use a capacitor across the feedback resistor unless the op-amp is unity gain stable.
