Removing Power tubes to cut power?

[JLDFla]

I have heard that two power tubes [120 Watt, 4 6L6] can be removed to cut amp volume. Is this true and if so how do you know which two tubes to remove?

[collinsamps]

You would have to see how the pairs are wired to the OT primary to make sure, but most of the time pulling two inner or two outer works(Fender, marshall etc)

Also remember that it will double your output impedance when you remove two.

A switch can easily be installed to lift the cathode ground of two tubes if you plan on doing it on a regular basis.

You'll only get about a 3db drop in the end and IMHO it's not really worth it. I would go with a less efficient speaker myself, or both.

[JLDFla]

Thank you for the information. I see an Air Brake in my future.

[Luthierwnc]

+1 on collinsamp suggestion. If you pull two of four tubes, you double the primary impedance (half the paths to ground, twice as hard to get there). You have to double the impedance on the secondary to compensate. Hard to do in a combo. If you search this site you'll find some ways to put in a switch that doesn't change the impedance but takes the tubes out of the mix.

Back in the day, friends with Twins usually removed the two inner tubes but if they did it often, they would try to rotate the four tubes just to keep the wear more-or-less even. That flies in the face of the first paragraph but nobody much cared back then.

I haven't found it was worth the bother. The only thing I've found about cutting an amp down to half power was that you lose the low end.

sh

[martin manning]

If you search this site you'll find some ways to put in a switch that doesn't change the impedance but takes the tubes out of the mix.

I hope not. The output load is almost all reactive, so if two of the tubes are not conducting signal current through the OT, there's no way around the other pair seeing only half of the load that they did with all four working.

[pdf64]

Could some of the tubes be taken out of operation, without affecting the amp output impedance, by removing their control grid signal?

[Cliff Schecht]

Yeah but then you are literally cooking tubes for no reason. If you use a lot of hard to find older tubes like I do then this is not the smartest option. Also I'm not sure how two tubes in parallel would react if both are biased on but only one is swinging. It seems like the plate voltage flying all over the place would cause the other one to shift around its bias point which could cause unknown/unforeseen problems. Most output tubes don't like to be used past maybe 3 in parallel or the tubes start to fight each other. I've never tinkered with this so this is purely speculative though, maybe it would work fine..

Martin is right also about the output impedance changing when you pull a tube. There is no way around this aside from biasing the now single tube to a point in which its plate impedance matches what the pair was sitting at, but this would be something like 150% plate dissipation at idle for a single tube and would cook most anything you do this to.

Maybe look at how Mesa Boogie does their Simul-class stuff, I haven't studied it deeply but I'm pretty sure there isn't anything special in the output transformer (like gapping to support standing DC current for true class A operation) so you might be able to figure something out there.