I am building a small power supply for some audio equipment I have. What I want to do is introduce some sort of bandpass filter before the transformers to only allow 50/60hz frequencies into the supply. Is it better to just make one using caps and inductors or better to find an already make PCB package that I can just solder to the board? It has to allow 3 Amps of current at most. Can you guys give recommendations for both?
Regards,
Christan
Regards,
Christan
I'm not an expert, here, but...
I'm not sure why you need that. What other frequencies are you concerned about? If it's only higher frequencies that you're worried about, maybe you could use a (simpler) LC low-pass filter, instead of a bandpass. A low-frequency bandpass filter for large-ish currents will be more bulky and more expensive.
On the other hand, assuming that you have a pressing need for a narrow bandpass filter for 50-60Hz, remember that you can also use a notch filter to make a bandpass filter, by taking the difference between the filter output and the input. For only three amps, maybe you could use a power opamp (chipamp) for the difference amplifier. I mentioned that method because there are probably quite a few existing designs for 50-60Hz notch filters, available, that you might be able to scale for your current-level requirement. However, I don't know if that would be a practical approach, or not, in your case. There might also be problems with the inrush current.
Also, there is a very good free passive filter design software program downloadable from http://www.aade.com . But I've never used it for very low frequency filters, or high-current filters.
I also recommend simulating any filter that you plan to use, using the free LTspice (SwCadIII) software, from http://www.linear.com . Remember to specify the series resistance of all inductors and capacitors, by right-clicking on them (using the ESR, spec'd or calculated for 50-60Hz, for the caps). Besides plotting the frequency response, you should also plot the power dissipation of each component (Alt-Left-Click on component), after a transient-response simulation run. And then you can get the average dissipation by Ctrl-Left-Click-ing on the plot label. Make sure that the caps' ripple current specs are not exceeded. And remember that the ESR of electrolytic caps can vary tremendously, with temperature, often much more than it varies with frequency.
Good luck.
I'm not sure why you need that. What other frequencies are you concerned about? If it's only higher frequencies that you're worried about, maybe you could use a (simpler) LC low-pass filter, instead of a bandpass. A low-frequency bandpass filter for large-ish currents will be more bulky and more expensive.
On the other hand, assuming that you have a pressing need for a narrow bandpass filter for 50-60Hz, remember that you can also use a notch filter to make a bandpass filter, by taking the difference between the filter output and the input. For only three amps, maybe you could use a power opamp (chipamp) for the difference amplifier. I mentioned that method because there are probably quite a few existing designs for 50-60Hz notch filters, available, that you might be able to scale for your current-level requirement. However, I don't know if that would be a practical approach, or not, in your case. There might also be problems with the inrush current.
Also, there is a very good free passive filter design software program downloadable from http://www.aade.com . But I've never used it for very low frequency filters, or high-current filters.
I also recommend simulating any filter that you plan to use, using the free LTspice (SwCadIII) software, from http://www.linear.com . Remember to specify the series resistance of all inductors and capacitors, by right-clicking on them (using the ESR, spec'd or calculated for 50-60Hz, for the caps). Besides plotting the frequency response, you should also plot the power dissipation of each component (Alt-Left-Click on component), after a transient-response simulation run. And then you can get the average dissipation by Ctrl-Left-Click-ing on the plot label. Make sure that the caps' ripple current specs are not exceeded. And remember that the ESR of electrolytic caps can vary tremendously, with temperature, often much more than it varies with frequency.
Good luck.
Hi,
toroid or EI transformer?
Toroids are susceptible to DC from the mains. This can happen when the positive half of the Mains AC waveform does not exactly match the negative half. This can vary from second to second being dependant on other consumers on the power line.
A DC blocking capacitor in the primary feed to the toroid can solve a mechanical hum that is due to the DC coming in with the wanted AC signal.
The DC blocking capacitor is a high pass filter and would complement a low pass filter. Both filters attenuating the out of band frequencies that your transformer and PSU do not want to see.
http://sjostromaudio.com/joomla/index.php?option=com_content&task=view&id=62&Itemid=27
toroid or EI transformer?
Toroids are susceptible to DC from the mains. This can happen when the positive half of the Mains AC waveform does not exactly match the negative half. This can vary from second to second being dependant on other consumers on the power line.
A DC blocking capacitor in the primary feed to the toroid can solve a mechanical hum that is due to the DC coming in with the wanted AC signal.
The DC blocking capacitor is a high pass filter and would complement a low pass filter. Both filters attenuating the out of band frequencies that your transformer and PSU do not want to see.
http://sjostromaudio.com/joomla/index.php?option=com_content&task=view&id=62&Itemid=27
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.