Hi all,
I just started learning about opamps and basic audio circuitry. I wanted to try and build the attached gain circuit with a NE5532!
Only thing I changed is R2 for 3.3k because with my +/-12v it'll clip otherwise. When I plug in my headphones (45ohms) at the other end It adds this distortion to the signal and I can't wrap my head around it!! Calculating the output current (lets say at 4v, with 100+45ohms =27mA) I stay in the limits of the opamp yet it distorts.
When I look at the datasheet for the opamp, I assume it shows a minimum load of 600ohms but can someone explain to me why minimum load is a factor?
it shouldn't be when I keep the current between nominal values right?
LEFT: SIGNAL MEASURED AT U1B OUT WITHOUT HEADPHONES PLUGGED IN
RIGHT: SIGNAL MEASURED AT U1B OUT WITH HEADPHONES PLUGGED IN
I just started learning about opamps and basic audio circuitry. I wanted to try and build the attached gain circuit with a NE5532!
Only thing I changed is R2 for 3.3k because with my +/-12v it'll clip otherwise. When I plug in my headphones (45ohms) at the other end It adds this distortion to the signal and I can't wrap my head around it!! Calculating the output current (lets say at 4v, with 100+45ohms =27mA) I stay in the limits of the opamp yet it distorts.
When I look at the datasheet for the opamp, I assume it shows a minimum load of 600ohms but can someone explain to me why minimum load is a factor?
it shouldn't be when I keep the current between nominal values right?
LEFT: SIGNAL MEASURED AT U1B OUT WITHOUT HEADPHONES PLUGGED IN
RIGHT: SIGNAL MEASURED AT U1B OUT WITH HEADPHONES PLUGGED IN
Your gain happens in the second stage with the first being a follower. I have a strong feeling that you'd get a more ideal performance if you were to do it the other way. A gain of 10V/V would reduce the loop-gain by 20dB whereas a gain of 1V/V (follower) would not. I would suggest that you try this before you try anything else.
First, when using someone else's (in this case, Rod Elliott's) schematic, make sure to provide a reference to the original.
Second, this is not distortion, this is instability. It may be related to a number of different factors, one of them is un-bypassed power supplies. Do you have capacitors in the power supply rails of the opamp? If not, read e.g. "Opamp Bypassing" by Rod Elliott and add something - 0.1uF film may be enough, 10uF or 22uF electrolytic wold be more usual.
Finally, after you have solved the instability problem, the NE5532 is not designed to drive such low impedance loads. Even though its output current is limited at somewhere between 10mA and 60mA, you should not expect a stellar performance with load impedance (much) below 600 ohms. You can use a different opamp with higher output current capability, such as NJM4556, or add a discrete current booster (e.g. Rod Elliott's Project 24).
Second, this is not distortion, this is instability. It may be related to a number of different factors, one of them is un-bypassed power supplies. Do you have capacitors in the power supply rails of the opamp? If not, read e.g. "Opamp Bypassing" by Rod Elliott and add something - 0.1uF film may be enough, 10uF or 22uF electrolytic wold be more usual.
Finally, after you have solved the instability problem, the NE5532 is not designed to drive such low impedance loads. Even though its output current is limited at somewhere between 10mA and 60mA, you should not expect a stellar performance with load impedance (much) below 600 ohms. You can use a different opamp with higher output current capability, such as NJM4556, or add a discrete current booster (e.g. Rod Elliott's Project 24).
Thanks alot! Bypassing the opamp worked like a charm!
I am sorry, I should have put a reference to the original circuit
I do have an extra question if u don't mind:
From the potmeter fully closed to about 10% open I get a signal which (when pot closed) starts at about 1MHz and decreases in frequency until pot is 10% open, then the source signal starts apearing. Any idea what this could be?
I'm sorry if these are obvious questions, after my working hours I'm constantly trying to learn and increase my knowledge about electronics. I would not ask questions If I hadn't already been pounding my head and google over them!
Have a nice day
No, it is a passive volume control before a two stage amplifier. If the pot was where R2/3 are it would be a Baxandall.
This is a Baxandall volume control. Elliott's schematic is an implementation of Fig. 23(b) from the attached article, so the volume control CCW end correctly goes to the output.
As for the instability, I have no ready answer. It, indeed, may be due to construction. Have a look at this TI's verified design for an example of this volume control that works.
As for the instability, I have no ready answer. It, indeed, may be due to construction. Have a look at this TI's verified design for an example of this volume control that works.
There is a slight problem with the circuit as drawn in the schematic in post #1 above. It may need a small ceramic cap, something around 100pF for the NE5532, in parallel with R3 (10k) for stability at low volume levels. Some experimentation with the value of that cap may be required, esp. with other opamps.
No, you read the article
In the article (attached to my post above), Baxandall in Fig. 22(a) introduces what you call the Baxanadall control, explains its limitation (very high gain at the top end of the control range), then suggests a modification to address that limitation in Fig. 22(b) and two ways to implement that modification in Fig. 23 (a) and (b). Fig. 23(b) is what OP it trying to make work.