Hi Roger,
Your question isn't as easy to answer as you might think.
How an opamp actually sounds comes down to more than just specifications. They do have their own unique sound and even two similar specced devices can sound totally different. The 4558/9, 1458, 741 types are totally out of the running nowadays.
I have the data sheets for most of the common audio opamps (thanks) and I have used the TL072 extensively in the past. The TL072 tends to be pretty docile in use and although it has a fairly reasonable slew rate it also has a high output impedance (compared to many other faster devices) and that also helps keep things stable.
Noise isn't really an issue here and it's quite a complex subject. I tried to find something easily readable on it for you...
Noise Figure & Other Stuff
The TL072 is an ideal one to try because it will be a major step up in sound quality.
If you do fit a socket and want to experiment then I also really like the OPA2604 and the OPA2134 and, to play safe and avoid problems you could always fit a small cap of around 4.7 pF (4.7 picofarad) across R17 and R18 (the feedback resistor) to ensure absolute stabilty.
Your question isn't as easy to answer as you might think.
How an opamp actually sounds comes down to more than just specifications. They do have their own unique sound and even two similar specced devices can sound totally different. The 4558/9, 1458, 741 types are totally out of the running nowadays.
I have the data sheets for most of the common audio opamps (thanks) and I have used the TL072 extensively in the past. The TL072 tends to be pretty docile in use and although it has a fairly reasonable slew rate it also has a high output impedance (compared to many other faster devices) and that also helps keep things stable.
Noise isn't really an issue here and it's quite a complex subject. I tried to find something easily readable on it for you...
Noise Figure & Other Stuff
The TL072 is an ideal one to try because it will be a major step up in sound quality.
If you do fit a socket and want to experiment then I also really like the OPA2604 and the OPA2134 and, to play safe and avoid problems you could always fit a small cap of around 4.7 pF (4.7 picofarad) across R17 and R18 (the feedback resistor) to ensure absolute stabilty.
@ Mooly
LOL I know these are not easy questions. 😀 I appreciate you sharing your time and your knowledge. And thanks for the ESP link.
As to output impedance: I am not finding that on the TL072 spec sheet. Is there a way to calculate that from other specs so that I can compare among different opamps?
LOL I know these are not easy questions. 😀 I appreciate you sharing your time and your knowledge. And thanks for the ESP link.
As to output impedance: I am not finding that on the TL072 spec sheet. Is there a way to calculate that from other specs so that I can compare among different opamps?
The TL072 has a 128 ohm series output resistor (look at the internal circuit in the data sheet) and that gives isolation. The effective output impedance when the opamp is in use is still effectively extremely low though.
Output impedance as seen at any opamp output is related to the closed and open loop gains.
If this is a subject that you are really interested in then this book by Doug Self is excellent and very easy to read. Maybe a library could get one...
The Audio Power Amplifier Design Handbook
Output impedance as seen at any opamp output is related to the closed and open loop gains.
If this is a subject that you are really interested in then this book by Doug Self is excellent and very easy to read. Maybe a library could get one...
The Audio Power Amplifier Design Handbook
The NJM4558 datasheet I have got from datasheet catalog specifies a input noise spec of 1.2 microvolt at 30 khz and r2=1 kohm. these numbers cannot be directly compared to sqrt nv/sqrt hz numbers, you have to do an integral. And companies are always playing around with the test frequency and source impedance of the noise figure to look good. So it is hard to tell apples vs oranges. I put sockets in my disco mixer, and ST33078's hiss less than NJM4558's. ST33078 is 4.5 nv/sqrtHz @ 1 khz & 100 ohm source impedance. I think the difference between NJM and NJR prefix is a corporate reorganization or something. I tried some NJM4560's out of a Peavey amp in the socket, they sound about the same as the St33078's. The spec on the TI RC4560 is 1.2 microv noise @ 30 khz and 2 Kohm r source, which is about the same as the 4558, but the ones I have sound better. I don't have any tL072, I had some trouble with the op amps whanging into the power supply rail and am afraid of an IC that latches up in those conditions. Soldering of traces on 0.1" separated pins is not a 100% controlled process in my case.
- Status
- Not open for further replies.