SWR SM-900 Distorting

[rf7]

This SM-900 amp sounds distorted on the low notes mainly. On the scope, the output looks OK. Is there any way that I can make distortion easier to see on my scope?

I checked the rails and I found a 1.5Mhz oscillation that seems to be present on all of the power supply output voltages. I pulled the preamp and power amp connections and the rails look normal with the usual 120Hz ripple. I just connected the preamp only and I am seeing the oscillation on both rails, but it occurs very strangely. If I probe one of the 15V rails, it looks fine and stays that way. It is only when I switch back and forth from the positive rail to the negative rail that I see the oscillation. And at that point, it is on all the rails. And when I don't see the oscillation, I will see it come up when I power off the amp, for about 20 seconds. And to top if off, sometimes, turning the Limiter knob will make it go away. This seems like really weird behavior for a normal style power supply. Has anyone seen anything like this before? What is causing that? It almost seems like my scope probe is instigating it.

This amp has seen some heat, the plastic 5 sided input jacks looked very melted--I replaced them, the shorting switches were no longer working.

[MarkusBass]

You may see distortion better if you provide a low amplitude, high frequency signal of e.g. 15 kHz (and look at the signal with an oscilloscope). But, as you said, in this case it's a high frequency oscillation so I don't see a point of checking distortion.
I would look at electrolitic capacitors and ground connections in the preamp. What if you solder small capacitors to the rails in the preamp (of course with the amp being switched off)? Does it depends on gain and master pots settings?
I had this amp last week but it had some other problems. It's a great sounding amp.

Mark

[rf7]

You may see distortion better if you provide a low amplitude, high frequency signal of e.g. 15 kHz (and look at the signal with an oscilloscope).
Mark

Thanks Mark, that helped me to see it. Distortion is definitely more obvious at lower levels.

It turned out that the distortion was only on the left output, and was not coming from the preamp. Some other shop out here repaired this amp 4 years ago and they replaced an output module, but they never bothered to adjust the bias on it. The guy has been playing it like that for 4 years.

What if you solder small capacitors to the rails in the preamp (of course with the amp being switched off)? Does it depends on gain and master pots settings?

Capacitors did not help. And no, only the limiter pot seemed to affect the oscillation. But, the oscillation only seemed to occur after I probed the rails a few times. It's almost as if my scope probe was a factor in the oscillation happening. Weird.

I'm not hearing any apparent effects of that crazy 1.5Mhz oscillation on the output signal at all.

[MarkusBass]

Some other shop out here repaired this amp 4 years ago and they replaced an output module, but they never bothered to adjust the bias on it. The guy has been playing it like that for 4 years.

That's good for you. I hope you will get another satisfied customer . How could they forget to set the bias?

.... But, the oscillation only seemed to occur after I probed the rails a few times. It's almost as if my scope probe was a factor in the oscillation happening. Weird.

This is weird. An oscilloscope probe is nothing more that a large value resistor and a small capacitor. The rails shouldn't be sensitive to this. It is worth investigating. Isn't it a problem with an oscilloscope? And you are touching the rail with the probe and ground link is connected to the ground, isn't it? Oscillations shouldn't happen.

Mark

[Enzo]

A large value resistor a small value cap AND three of four feet of wire. That is an antenna in many places.

[rf7]

When I worked at Hughes Aircraft on digital electronics, the guys who worked on analog circuits were only allowed to use probes with a 1 foot long lead.

[MarkusBass]

A large value resistor a small value cap AND three of four feet of wire. That is an antenna in many places.

Yes, I agree. But we are here talking about a power supply rail and not about a very sensitive high-gain preamp input. It's still worth investigating .

Mark

[rf7]

That's good for you. I hope you will get another satisfied customer . How could they forget to set the bias?


Mark

Thanks Mark. I don't know how they did not set the bias. I heard the distortion right away when I played it.

Isn't it a problem with an oscilloscope? And you are touching the rail with the probe and ground link is connected to the ground, isn't it? Oscillations shouldn't happen.

Possibly. Yes the probe is on the rail and the ground is to the chassis.

I've never seen that much high frequency oscillation hash on a standard power supply. On a Switching power supply , it's a different story.

[Enzo]

And this hi freq oscillation is NOT on the output bus? generally if I see some signal on power rails it is being impressed theree by substantial current draw. What does the output look like?