Mcintosh MC240 question.

[Indyryder02]

Hopefully this is an easy question for those of you with much more experience than I. (Or maybe even a little more haha)

Just recently rebuilt the power supply and bias board on this amp.

My question is…
For some reason, (maybe the specially designed output transformers?). The plate dissipation, negative bias and B+ all jump all over erratically with no load connected. I’ve never experienced this with any tube amp, let alone Macs. Normally any amp I’m testing etc.. the voltages etc are all stable without any load connected. I often test amps for initial health this way before connecting loads , scopes etc etc.
without any issues.
The amp is stable once speakers are connected, and while running signal through the amp. So seems to be fine. But I’ve never seen any tube amp that had to have outputs connected to remain stable? I’m just afraid I’m missing something else that’s not right. If not. Great! Learned a lesson on Mcintosh 240’s Haha but if this isn’t normal.. I don’t even know what to look at! Thanks in advance for any guidance or advice.

[Helmholtz]

Any tube amp must have a load connected.
Otherwise the amp can (self-oscillate and) produce kV OTprimary voltages putting the OT at risk of insulation breakdown.

[Indyryder02]

Any tube amp must have a load connected.
Otherwise the amp can (self-oscillate and) produce kV OTprimary voltages putting the OT at risk of insulation breakdown.

I’ve always understood that is only necessary when applying signal to the amp? I’ve never had any issues like this on any other amp without signal and no load. Surely there is something different or wrong with this particular amp.?

[Helmholtz]

Self-oscillation doesn't require applying a signal.
Instability depends on the filter poles within the NFB loop, NFB design and the open loop gain.
Open loop gain hugely increases without a proper load.

It is not safe to operate a tube amp without proper load, period.

[g1]

Lots of tube amps have shorting jacks at the output so there is a short rather than open if you forget to connect the speakers. Many of those you tested probably had that.
Other than that, consider yourself lucky. In most cases they won't go crazy with no signal and no load, but any fault or oscillation and you could have serious damage.

[Indyryder02]

Ok. Well. Yes then. Definitely sounds I’ve been lucky… on hundreds of amps now! And yes, generally I work on guitar amps with shunting output jacks.. but I’ve never ran into this on any other HiFi amp before either. Likely different NFB designs etc as you explained.
I actually learned the no load needed if no signal applied from Mr Carlson’s YouTube. But seems that must be only if it’s a healthy amp and likely shunted outputs etc.

My first time ever running into it. So it really was throwing me. But seems like it’s ok then. Just has to be loaded.

I appreciate the wise info. Just cause I’ve been lucky, doesn’t mean I shouldn’t play it safe always!

Thank you!!

[Helmholtz]

Hifi amps often use considerably more NFB than tube amps.
Also bad filter caps can facilitate oscillation.
It is often overlooked that filter caps are actually in the signal path.

[Indyryder02]

Hifi amps often use considerably more NFB than tube amps.
Also bad filter caps can facilitate oscillation.
It is often overlooked that filter caps are actually in the signal path.

I did replace all filter caps and the final coupling caps.
And I thought originally I had checked the bias, without a load. And it was stable.. which is what was really making me think I made a mistake of some kind after replacing the final coupling caps. But maybe I had it hooked up to speakers when I first checked it and now don’t remember.

Hmm, it’s possible I got the lead dress out of whack when doing the last caps. And maybe that’s creating an oscillation. I better check it with a scope to be sure. As it seemed to jump around a bit still at one point with the load. But, it’s been running now a couple hours with music through it. And it’s quite stable currently.

[g1]

There are a lot of inductors in that output circuit, not what I would consider a common tube power amp circuit at all. So to me, for it to act differently from what you are used to seems quite plausible.

[Helmholtz]

Don't forget that even flipping a switch or connecting meter leads injects a transient signal which could start oscillation with an unstable amp.