Hi Chris.
You asked:
<< What types of noise sources within a typical power supply or product would
produce emissions in this frequency range? >>
Sorry if this is teaching you to suck eggs...
Any bridge rectifier/capacitor filter will generate Conducted EMI is that
frequency range, due to the 'commutation' effects of the forward conducting
period of the rectifier diodes.
The higher the value of smoothing capacitor, the longer the voltage on the
other side of the diode remains high, and the diode wont conduct until there
is sufficient voltage (0.7V'ish) across the diode to turn it on. So, all
the current to be delivered by the power supply, must be delivered to the
smoothing capacitors while conducting for a very short period, sometimes as
low as 100uS, or even 10uS.
The average current in must equal the average current out, so if you are
conducting only fo 100uS, rather than the full half cycle (10mS in the UK)
then in order to maintain average output current the diode current must be
10mS/100uS x Iout = 100 Iout. So if you have a 5A power supply, the diode
current will be 500A, for 100uS. A higher value of smoothing capacitor may
allow only 10uS conducting time, so the current would be 5000A for 10uS. In
reality these currents are rarely achieved, you simply see very hot diodes
and output voltage drop.
So, you have very high speed switching of very high currents just buzzing and
fizzing with high frequencies. Diodes can switch very fast, and I have often
measured huge amounts of conducted EMI up to 30MHz and above, though it then
starts to radiate... It's a situation where the non combatant will see
conducted emissions and increase the smoothing to try to remove it, when in
fact that just exacerbates the situation. A simple way to reduce conducted
emissions substantially is to reduce the smoothing capacitor to the point
where it becomes noticeable on the target system, then double it! You may
be very surpprised how 10,000uF capacitors can be reduced to a few hundred
with no significant increase in ripple, but a HUGE 20-30dB reduction in
conducted EMI.
Difficult systems benefit from 3 phase supplies to the bridge, you can get
away with no smoothing at all on non critical systems... Or using multiphase
transformers etc. etc... This 'peaky' current demand of course, being as it
is at the very peak of the supply voltage waveform, is the source of the
problem that is the purpose of IEC1000.3.2, etc. the generation of harmonics
back into the mains supply.
Sorry if I missed the point of your question, but I'm a simple soul and saw
what I thought was a simple and obvious cause. Or if it helps, I'm glad.
Chris Dupres
Surrey, UK.
-------------------------------------------
This message is from the IEEE EMC Society Product Safety
Technical Committee emc-pstc discussion list.
To cancel your subscription, send mail to:
[email protected]
with the single line:
unsubscribe emc-pstc
For help, send mail to the list administrators:
Jim Bacher: [email protected]
Michael Garretson: [email protected]
For policy questions, send mail to:
Richard Nute: [email protected]
