Nelson Pass said:There are three typical causes for intermittent mechanical hum
from the transformer.
1) Your circuit might intermittently draw very large amounts of
current. This is not very likely.
2) Your AC line might have some DC type noise on it, which
saturates the transformer. Lamp dimmers and other Variac
type controlled equipment are the usual culprit, and a filter
network is the usual answer.
3) Your secondary system might be drawing, on average, more
current from one rail than the other, in other words, different
+ to ground than - to ground. If the secondary windings are
mismatched and you are using only 1 rectifier bridge for the
two polarities, this can cause the same kind of saturation as
in (2). This is common when there is high DC offset at the
output through the speaker. Besides balancing out the draw,
you can also cure it by providing one rectifier bridge for each
rail polarity, assuming that you have dual secondaries and
not just a center-tapped secondary.
Is (3) more likely when two transformers ares used? (one for each rail)
1) Is not the case.
2) Is not the case unless 1mV DC average is too much. When the hum occurs the DC doesn’t rise. Max DC is 2.235mV measured over a period of 30’.
3) I can’t imagine both amps draw unequal currents exactly at the same time when not playing any music. Dc at the output is quite stable and within specs range.
The hum occurs definitely when my neighbour is welding.
I am however not able to measure big differences in the AC power condition then.
So how would a filter network look like?
I found an example in Elektuur, showing 2 100nF capacitors and a 30µH choke in between. I made it a time ago but never used it.
/Hugo
2) Is not the case unless 1mV DC average is too much. When the hum occurs the DC doesn’t rise. Max DC is 2.235mV measured over a period of 30’.
3) I can’t imagine both amps draw unequal currents exactly at the same time when not playing any music. Dc at the output is quite stable and within specs range.
The hum occurs definitely when my neighbour is welding.
I am however not able to measure big differences in the AC power condition then.
So how would a filter network look like?
I found an example in Elektuur, showing 2 100nF capacitors and a 30µH choke in between. I made it a time ago but never used it.
/Hugo
Attachments
A filter like this one will block mains DC (C1/4 and D1/2/3/4, the rest is EMI filter).
Another thing you could try is using a Zobel snubber on the secondaries, as described here (while I've used this to get cleaner power, in one case is shut up a singing transformer; but good luck calculating the correct values if you don't know the parasitic capacitance and inductance of the transformer, maybe trial and error...)
Another thing you could try is using a Zobel snubber on the secondaries, as described here (while I've used this to get cleaner power, in one case is shut up a singing transformer; but good luck calculating the correct values if you don't know the parasitic capacitance and inductance of the transformer, maybe trial and error...)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.