🚀 go-pugleaf

CJT <cheljuba@prodigy.net> wrote in message news:<3EC87DD3.5030302@prodigy.net>...
> DaveC wrote:
> 
> > I agreed to help my friend find out why his 5-channel surround system 
> > (Harmon-Kardon; don't remember model number) has 60hz hum in all speakers. 
> > Hum is uneffected by the volume control. It is not in the headphone 
> > circuit. 
> > 
> > There is a small power supply circuit that feeds +15 and -15 vdc to some of 
> > the ICs on the power amp PCB. This circuit consists of a small transformer 
> > with c.t. secondary feeding a single bridge rectifier. Tied to the plus and 
> > minus side of the bridge are two 2200uF 35v electrolytic caps. Beyond each > > of these caps is a (pi?) filter: 47uF elect. cap (to ground), a series 
> > resistor, and another 47uF elect. cap (to ground). 
> > 
> > These two supply "rails" each feed 15v 3-terminal voltage regulators, one 
> > +15v, one -15v. 
> > 
> > The plus supply has ripple voltage on it everywhere in the circuit. I 
> > checked the voltage at the 2200uF cap with a scope and saw the ripple, and > > said "Bad filter cap" and replaced it. Ripple remains. I pulled each of the > > 47uF caps and tested them with a Fluke cap tester. They are within 1uF of 
> > spec. 
> > 
> > I had to leave my friend's place before getting to diagnose it further, but > > I can't think of what else could cause this ripple. I am aware of the 
> > epidemic of failing electrolytics. Is a capacitance test sufficient to 
> > declare a cap good? Or must I test resistance, also? Can a shorted bridge 
> > diode introduce ripple? I haven't yet checked the series resistor in the 
> > filter. If this changed value, could this cause ripple?
> > 
> > I've checked for bad solder joints. I can't imagine that the v.reg. can 
> > introduce ripple into the filter circuit. Can it?
> > 
> > Any other suggestions would be greatly appreciated.
> > 
> > Thanks,
> 
> Maybe one of the diodes in the bridge is shorted (although that ought to
> blow a fuse).  Even the biggest capacitor won't remove the ripple from
> AC!

The first thing to do with this sort of circuit is to check the DC
voltages before and after the linear regulators.

My guess is that you will find that the DC level at the output of the
positive-side pi-filter is less than 17V and the output from the
linear regulator is less than 15V. Without any head-room to regulate
within, the linear regulator won't do much of a job, least of all at
ripple rejection.

If you can work out the resistance of the resistor in the pi-filter
and check the DC voltage drop across it you will have a useful extra
insight into what is going on.

It could be that one of the IC's powered by the +/-15V rail is damaged
and pulling an excessive amount of current from the positive rail - if
this is the case disconnecting one channel at a time should start
localising it.

-----
Bill Sloman, Nijmegen