SET vs. PP? Naw, it's not what you think!

[NOSValves]

Maybe I missed it, but no one seems to have mentioned that 2nd order distortion is canceled in the PP circuit. Even using the tubes in their linear portion, that point is likely to give a different sound from SE.

I have a PP Class A DHT (2A3) amp, and it can do some things very well. I plan to pull it out soon and try it on the 98db Tannoys.

One of the very best amps I ever heard was Komuro's PP 845 amp. No reason you can't use great triodes in a PP amp.

It's been my experience that the really *great* SE and PP amps do sound similar, particularly on efficient speakers where the distortion level is very very low. And this is from a guy who has NO PLANS to switch from his SE amp.

Tubino,

I have seen you metnion 98db about your Tannoy's a number of times but I keep seeing specs on the net at 101db am I missing something?

Craig

 

[dbsupercrew]

reading this makes my head hurt.........no flames.........just one heck of a headache........great info..............

 

[tubino]

I have seen you metnion 98db about your Tannoy's a number of times but I keep seeing specs on the net at 101db am I missing something?

The 98 number is the 1watt/1meter number that is comparable to what everybody else uses. The 101 is some anechoic/half-space complication that seems to come down to the regular number plus 3 db.

Still, 98 isn't a bad number for a 15" that acts more like a mini-monitor than a horn. I've been too busy to hook up flea amps to them, but I'll get around to it. I'm very curious to know what a single 2A3 or 1626 would do on them. I've got a 1626 amp (circuit is Bob Danielak's "Darling" -- one of the all-time great DIY amps for first-timers) waiting to have the dust blown off after a long hiatus. Next month when the wife and kids are gone...

 

[heekma]

Maybe I missed it, but no one seems to have mentioned that 2nd order distortion is canceled in the PP circuit. Even using the tubes in their linear portion, that point is likely to give a different sound from SE.

Ahh...Tubino, I knew that, but didn't think to factor that in-you're absolutely right-even *assuming* that the linear portions of the tube characteristics are *similar* that center-tapped output on the PP amp cancelling 2nd order harmonics and some hum would indeed make a difference in the sound! :yes:

Hooked up to an O-scope, they would indeed "test different and measure different" as Craig mentioned-you might just see a little extra "bump" creeping into the waveform on the SET from the addition of the 2nd order distortion to the fundamental that would be lessend or even cancelled entirely from the PP amp.

See, I think this has been a pretty good thread about SET-nobody's tried to propose one is better or worse than the other, just what exactly it is that contributes to their differences-there's a lot of platitudes and commonly referred to descriptions about their differences-that old saw of which is better; PP or SET?-but there doesn't seem to be more concrete discussions of why. Leave which is better to the individual, but discussing why they're DIFFERENT, not neccessarily BETTER is some interesting stuff!

Best,
mojo

 

[heekma]

Craig,

See, no need to don the fire-proof Suit! :thmbsp:

Glad you came back in to make some great points as always. Now, just to play the devil's advocate with Thatch's point...

With the question of parts count as a contributing factor, since capacitors can act like a "low-frequency filter" if the frequency dips low enough to *approximate* a kind of DC rather than AC (that's my understanding, if I'm off I would surely welcome being corrected!) then could the number of coupling caps and their value be a contributing factor to "shaping" the sound of a PP amp more than a SET-especially if there are more of them in the circuit?

This is just the kind of stuff I was curious about! I feel like this thread is getting to some of the actual reasons it sounds different-who cares about better or worse!? Some of the things discussed would be good reasons for why different tubes which share a *similar* characteristic curve when run in their linear portion would still manage to sound different!

Best,
mojo

 

[heekma]

It's been my experience that the really *great* SE and PP amps do sound similar, particularly on efficient speakers where the distortion level is very very low.

tubino,

Your experiences with and opinions of great amps of both flavors addresses my question nicely:

Mainly, that a pentode and a triode, having *similar* characteristic curves in their most linear portion should, and in your experience DO sound similar-when kept in that most linear portion of the curve! That has been the driving force behind my curiosity:

If they share a similar reproduction curve in their linear portion, then they should reproduce sound in a similar way. When a person reads all the hyperbole on the net about pentodes vs. triodes it seems they are worlds apart. However, when looking closely at them, in at least some ways there should be more similarities than differences.

The idea that this is in fact true in some cases, and in others not-and the reasons for each is what I find most interesting!

Best,
mojo

 

[tubino]

mojo,

I'm a bit uncomfortable with this equation of SE vs PP with triode vs pentode. I was only commenting on the first comparison, and didn't say anthing about the second. In fact I thought it was time to point out that there are PP amps that use DHTs, just to keep the two discussions separate. And there are SE pentode amps too.

There is a pattern of characteristics that make most SE amps sound different from most PP amps. On the SE side you have the inductance/capacitance tradeoff in the OPT, the magnetization of the OPT with unbalanced DC, th llack of cancellation, etc. On the PP side you have the phase inverter, and typical lack of perfect balance over the dynamic range, even in the halves of the OPT. And I'm just way way oversimplifying.

The triode vs pentode thing is not simple either though. Child's Law, miller effect, knees/kinks in the transfer curve, sensitivity to variations in the reflected primary impedance... capability and exploitation of A2 possibilities, the adequacy of the driver to deliver, the difference with direct coupling vs. cap coupling as it relates to the items above...

When you look at the characteristic curves, none of that messy implementation stuff is addressed -- but you can't build an amp without it. It's easy to build a tube amp that works and even sounds good without understanding that stuff --heck, even I have done it!

But because of all that stuff above (which really matter!), I don't want to be on record as saying that triodes and pentodes sound similar when operated in their most linear portion -- without a chance to footnote it like crazy!!!

 

[heekma]

When you look at the characteristic curves, none of that messy implementation stuff is addressed -- but you can't build an amp without it. It's easy to build a tube amp that works and even sounds good without understanding that stuff --heck, even I have done it!

Tubino,

Beleive me, if through my ignorance I seem to be loading questions to put people on the hotseat, it's not my intention-it really is simple ignorance!

That quote above gets right to the heart of my curiosity-as I skim through the tube manuals I see a lot of characteristic curves that look very similar-at my stage of the learning curve that seems on the surface to imply they should sound similar-it's that "messy stuff" you mention contributing to their differences that interests me.

EDIT: You're right, I should've been more consistant in my posts; I was mainly discussing the *potential* similarities of a pentode PP amp and a SET amp-in essance the idea that a pentode and a triode having a similar linear curve would suggest in some instances a similar sound-and only the impact of that similar characteristic curve, not taking into account circuit and topolgy differences.

Most of the ham and tube basic electronics books I have don't delve too far into the things you mentioned-it's more of a basic overview of how they work, a little bit of the why and then on to the next topic. More than just basic, but also kind of "in depth lite" so to speak! I guess I'm reaching a limit on what these books have to say and reaching out for more answers.

Any suggestions for books I can look for that deal with some of the more "in depth" questions I'm having?

Thanks tubino, I'm really just curious and looking for answers!
Best,
mojo

 

[NOSValves]

Craig,

See, no need to don the fire-proof Suit! :thmbsp:



Best,
mojo

That is only because I'm being forum correct and biting my toungue

Craig

 

[NOSValves]

tubino,

Your experiences with and opinions of great amps of both flavors addresses my question nicely:

Mainly, that a pentode and a triode, having *similar* characteristic curves in their most linear portion should, and in your experience DO sound similar-when kept in that most linear portion of the curve! That has been the driving force behind my curiosity:

If they share a similar reproduction curve in their linear portion, then they should reproduce sound in a similar way. When a person reads all the hyperbole on the net about pentodes vs. triodes it seems they are worlds apart. However, when looking closely at them, in at least some ways there should be more similarities than differences.

The idea that this is in fact true in some cases, and in others not-and the reasons for each is what I find most interesting!

Best,
mojo

In some ways this take is absolutely true. Now factor in the effeceincy difference of the tubes and you will see why the industry left triode laying in the parts bins in the 50'. The interesting point that Tubino made is the when running the tubes head to head in the same topology the differences really become very small. The biggest problem auditioning examples of PP or SE of both tube types is you would have a heck of a time finding amps with comparable build quality with similar design Criteria. Not the mention power.

Craig