4 power tubes vs 2

[Dave H]

Think I read somewhere with 4 tubes against 2 the amp damping factor is more improved. If so more output tubes means better bass control, that.s nice.

I haven't worked it out mathematically but my gut says that four tubes have half the output impedance of two tubes but the transformer used for four tubes has half the impedance ratio of the one used for two tubes so it comes out to be the same in the end.

[Gingertube]

What Dave said:

2 tubes on each side of the push pull - the 2 (parallel) tubes act like a single composite tube with 2 x the gm, 1/2 the rp and the same mu.
That is why you generally want an output transformer of half the primary Raa when you use a quad of output tubes - to give same output impedance and hence electrical damping of the speaker (BUT does nothing to change the mechanical damping of the speaker) .
Actual REAL Speaker damping is the RMS value of the Electrical and Mechanical damping factors.

Beyond that there are subtle differences which usually comes down to personal taste and some times real (but ususally imagined) golden ears.

The transition from pure Class A at low volume to Class AB happens at a slightly higher signal (volume) level with a quad of output tubes, which accounts for stories about some great players preferring a 50W Amp (Pair of output tubes) to a 100W Amp (quad of output tubes). At the transition point (at a particular signal level) the amp output impedance suddenly doubles and peaks of the sound loosens. Probably notice this most on note attack.

AND

Straight personal preferences for lower gm tubes (5mA/V or less) like 6V6, 6AQ5, 6L6 vs higher gm tubes (up around 10) like EL84, EL34, KT88 etc. - They do have different sound.

The arguement regarding gm gets really messy when you use 2 parallel tubes (double the gm) on each side particularly when taking into account that - for example, an EL34 has a book value gm of 10.5 mA/V BUT in most guitar amps they are biased at 70% dissipation at which point the gm is down to around 7 mA/V but rises back toward that 10.5 mA/V on signal peaks when extra current is drawn. That is, the gm and internal rp of the tube (or composite tube of double the gm in the case of a parallel pair) is non-linear with signal level anyway.

All this is beyond my ability to perceive real sonic differences.

Having said that I really do like the quad of 6V6G London Power "Standard" amp I built for a friend. To complicate the issue further he usually runs that amp with one output pair in fixed biased pentode mode and the other pair in cathode biased triode mode.

One of those issues with lots of opinion, a few theories, and next to zero proven facts.

Cheers,
Ian