Up until reading a recent issue of EDN I was routing the V+ and V-power traces on the bottom of my PCB's -- for simplicity sake. This entails a VIA when the trace emerges to mate with the opamp pins -- I hadn't realized that the inductance of the VIA could significantly affect performance (high speed designs include audio filters -- if you use high bandwidth opamps or cfb opamps). The authors claim that the distortion from improper power trace layout (using vias) is measurable.
Is it germaine to DIYaudio? I leave it to you folks who take apart and put back together your clockers, or anyone who has tried to work with high speed current feedback amplifiers.
Is it germaine to DIYaudio? I leave it to you folks who take apart and put back together your clockers, or anyone who has tried to work with high speed current feedback amplifiers.
You have potentially the same problem if you lay them out on top. I saw a photo somewhere showing the power connecting by wire virtually next to the opamp power pins and leaving the PCB at right angles. This is only practical when you have a a single package to supply.
More practical is to be sure the power supply is as clean as possible since I don't think truly pure DC wouls cause a problem. I suspect this is one of the advantages to battery powered instrumentation.
More directly to your question: it's concievable but not inevitable. As for PCB layout, I would place a higher priority on keeping the output section traces as far ftom the input signal as possible. An output inductor is especially nasty in the regard. Problems in this area are much more common. After that, keeping ground paths and power paths as short as possible and at the edges of the board as much as possible is desirable.
More practical is to be sure the power supply is as clean as possible since I don't think truly pure DC wouls cause a problem. I suspect this is one of the advantages to battery powered instrumentation.
More directly to your question: it's concievable but not inevitable. As for PCB layout, I would place a higher priority on keeping the output section traces as far ftom the input signal as possible. An output inductor is especially nasty in the regard. Problems in this area are much more common. After that, keeping ground paths and power paths as short as possible and at the edges of the board as much as possible is desirable.
I read that and got the impression it was more intended for the digital crowd - high speed data busses and such. If you are talking digital audio or DSP it applies. Otherwise I don't think we have much nanosecond rise time to protect in our audio stuff. That is just an impression though.
There is measurable and then there is noticeable. What does the inductance measure - I have since pitched the magazine - and then calculate the effect it will have on audio frequencies.
There is measurable and then there is noticeable. What does the inductance measure - I have since pitched the magazine - and then calculate the effect it will have on audio frequencies.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.