Hi!
In order to cool a telcom device I am using 4x 12Vdc fans from Delta Electronics (the exact type is FFB0412VHN-F00). The fans have a 3-wire interface and I operate them by sending PWM pulses at their power supply wire. The pulses are generated from a PIC-microcontroller and are "converted" to 12V pulses through two FETs (see attached schematic).
The issue is that the fans seem to disturbe the 12V power supply and as a result they generate several different problems, like bit errors at the transmissions etc, which seem to disappear when I remove the fans. A logical thought was that the pulse current drawn periodically from the fans is causing small dips in the power supply and I was able indeed to verify that with an oscilloscope. Therefore I built an LC low-pass filter in order to make the 12V quieter. And it seems to work when the duty cycle of the PWM pulses is low.
What however I still not understand is that when I give full power to the fans (the duty cycle is 100%) then I still have the same problems as before. And that actually still puzzles me because at full power there are no pulse currents drawn from the 12V... I have a feeling that the filter has worked at low duty cycle values because it creates an additional voltage drop and thus the fans receive a lower rms power supply.
Does anyone have any idea on what to do?
In order to cool a telcom device I am using 4x 12Vdc fans from Delta Electronics (the exact type is FFB0412VHN-F00). The fans have a 3-wire interface and I operate them by sending PWM pulses at their power supply wire. The pulses are generated from a PIC-microcontroller and are "converted" to 12V pulses through two FETs (see attached schematic).
The issue is that the fans seem to disturbe the 12V power supply and as a result they generate several different problems, like bit errors at the transmissions etc, which seem to disappear when I remove the fans. A logical thought was that the pulse current drawn periodically from the fans is causing small dips in the power supply and I was able indeed to verify that with an oscilloscope. Therefore I built an LC low-pass filter in order to make the 12V quieter. And it seems to work when the duty cycle of the PWM pulses is low.
What however I still not understand is that when I give full power to the fans (the duty cycle is 100%) then I still have the same problems as before. And that actually still puzzles me because at full power there are no pulse currents drawn from the 12V... I have a feeling that the filter has worked at low duty cycle values because it creates an additional voltage drop and thus the fans receive a lower rms power supply.
Does anyone have any idea on what to do?