What causes motorboating with 4 ohms?

[tmblack]

I have repaired an amplifier and it works correctly with 8 ohm loads.
When I connect a 4 ohm speaker the amplifier starts to softly buzz and then grows into a full motorboat sound.

I addes more caps across PS and this does not help. Adding 10 ohm resistor at output does allow it to drive 4 ohms but what a waste of power.

Is the PS too weak?
Any suggestions?

Thanks

Tom

[alaskanaudio]

I would suspect that the amplifier is unstable even with an 8 ohm load. Loading it with 4 ohms puts it over the edge and the oscillation that the instability produces becomes audible as motor boating and an also as a rise in noise level, buzzing as you describe it. This buzzing noise is probably due to the onset of nigh frequency oscillation.

The instability may be due to an inherent design problem which may have been aggravated by having replaced active parts with ones with higher gains or changes in other characteristics. It is also possible that the amplifier may still have a part installed that was slightly damaged by whatever caused the amplifier to fail originally. Further trouble shooting may be needed to find such a part. Don’t rule out a potentially damaged trace on a circuit board or other problem that may affect circuit continuity, grounding or a change in the value of a part such as a resistor.

I doubt that your problem is caused by the power supply. An oscilloscope would be nice to have so that you could see what is happing at the output of amplifier at various signal levels and loading. Care should be taken since oscillation in a power amplifier can lead to failure of the output devices and also damage tweeters. A variac with an AC current meter would also be nice to us so that the mains AC input current to the amplifier could be monitored during trouble shooting. Many amplifiers will show an increase in power consumption when oscillation starts.

[tmblack]

Thank you for your suggestions Alaskanaudio.
I used an AC voltmeter than is 0 when there is no load. I'll get an oscilloscope to confirm your suspections.

Tom

[Eva]

If the output devices of your amplifier are TO-3 metal can MOSFETs whose cases are used as source terminal [like 2SJ50 and 2SK135] and the circuit is a source follower, then : Try grounding the heatsink to a point near the local supply decoupling capacitors of each channel