Thanks very much, Roger. Due to suggestions on the overall circuit topology, the design current (which I apparently never specified) may have just gone down to around 50 mA. I was using 100 mA as a number before (though I apparently neglected to specify it), so maybe John was right.
--Mac
(snip)
(snip) (The following advice is my opinion. Others will probably have other opinions)
The voltage rating on the caps includes either 85C or 105C operating temperature at the full ripple current rating, depending on the particular device, for a specified life. And the caps have a surge rating for brief exposures, higher than the rated voltage. Being conservative on temperature, voltage, or ripple current will boost the expected life.
That said, if you are not going to expect this board to tolerate the full commercial ambient operating range of 0 to 70C, and don't push the ripple current rating, then you can select devices that are the next higher rating than the absolute highest voltage you can ever expect them to experience (high line voltage, no load on the supply, as I went through, above, except that I didn't include the transformer regulation at no load), and with normal operating conditions they will provide a very good life expectancy. If in doubt about the actual worst case peak voltage, go up one more voltage level. After that, I think you are just wasting money and board space without gaining much additional reliability. If you have plenty of both (money and board space), go with a bigger capacity or higher quality, instead of twice the needed voltage rating.
Electrolytic capacitors are generally much better than they were when I got started, 40 years ago. And they keep getting better. Just don't buy the cheapest, underspeced brand. Some of my favorites, at the moment are the FM series by Panasonic. They beat my previous preference (the FC series) with higher ripple current ratings and lower internal series resistance with equal or longer expected operating life times, and are generally smaller, to boot.
Thanks, John. Once again, I truly appreciate you taking the time to share "your opinion," which seems to be a well-informed one.
;-)
--Mac
That kind of opamp has an input and output range that is a couple volts inside the supply rails,so +- 9 volts would allow inputs and outputs to work up to about +- 7 volts, etc.
Quite possibly, not very important. Opamps have a power supply rejection ratio that is usually above 100 db (less than 1/100,000th of the supply ripple gets into the output). And the ear is not very sensitive to 120 Hz to begin with, so a tiny trace of that in the signal is not very audible. The biggest problem is in high gain stages, like phono preamps and low level microphone preamps. But those stages often include extra supply filtering.
I think it might be worth an experiment to find out.
I need to provide a similar supply to an old rack-mount mixing module for my son. It uses exclusively 1458 and 4558 op-amps and has a three-wire supply, so in the absence of data, I'm assuming it wants from +-9 to 15 volts.
My question is, how important is regulation likely to be? The module doesn't appear to have any on-board regulation, or any discrete transistors. Is it possible that it has sufficient PS rejection without needing regulators?
Clifford Heath.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.