SE, Parallel SE, PP and how it affects "watts"... Help me understand

[jonrpick]

I'm nearing the end of the initial design/layout phase of my next build. It's to be a parallel single-ended amp with a preamp loosely based on the "bass" channel of a 6G6-B, with Vibro Champ-style tremolo added in.

Should be pretty neat.

Anyway, I keep trying to get a better understanding of how much power to actually expect. This is more of an academic/curiosity thing, as I know that output in terms of "watts" doesn't necessarily correlate to "volume/loudness".

My question is really centered around parallel output tubes. We all know that push-pull is more efficient in terms of putting watts into the speaker. But I have yet to find anything definitive that will tell me what to expect in a parallel SE configuration. I *have* read things that indicate that it won't simply be double the output. For example, a 6W SE Champ won't suddenly be 12W if you slap another 6V6 in there wired parallel to the existing 6V6.

Although, one amp company built a harp amp using three 6V6's in parallel, and claims 15W.

(link removed)

So how much output will a pair of 6V6's in parallel produce?

The big "X" factor here, and what adds confusion (to me at least), is that the big difference between most 50W amps and 100W amps (at least on the schematic) is that the 100W amp appears to be a 50-watter with a 2nd pair of power tubes connected to the first pair in... parallel.

So... does running power tubes in parallel really have a significant effect on output power, or are the amp manufacturers just blowing smoke, or...??

What am I missing or not understanding here. I don't gig as a guitarist, so I'm not necessarily shooting for any particular end result. I'm just trying to understand. Every build or change I make to one of my amps is a learning exercise for me.

Thanks...

 

[clintj]

Tube datasheets list expected operating point for P-P and single ended class A. Class A plays by a different set of rules, and develops significantly less power per tube. 5W is typical of a 6V6 in that mode.

Two tubes in parallel should develop twice the power, in theory. There's other parts of the equation, though. The power supply needs to supply more current, the OT may need to be uprated to handle more power to avoid saturating, and so on.

 

[Wally]

Just curious, Jon.....why the bass channel?

 

[danlad]

My understanding was basically double up the valves (either way) and double the power. But as mentioned, so much depends on other factors and maybe doesn't matter so much for guitar as we don't care so much about clean power at whatever % distortion yadda yadda.

But the watts manufacturers give them seem just convenient markers. Y'know all push pull big valve output sections seem rated at 50 if there are 2, 100 if there are 4 a la Marshall. All smaller valve output sections rated 15 or 30 likewise a la Vox. Wonder what they really put out when measured?

 

[jhundt]

what they really put out, when measured, would probably depend largely on what standard they were measured to. Power ratings frequently are based on output power with minimal distortion and clipping, as explained in Wikipedia:

Typically, an amplifier's power specifications are calculated by measuring its RMS output voltage, with a continuous sine wave signal, at the onset of clipping—defined arbitrarily as a stated percentage of total harmonic distortion (THD)—into specified load resistances. Typical loads used are 8 and 4 ohms per channel...

but since most guitar players are looking for a certain amount of clipping and distortion, those "output watt" numbers become less meaningful.

 

[danlad]

True. No guitar tune is a 1khz sine wave producer so watts quoted is probably arbitrary, if based on a fair rule of thumb.

 

[danlad]

And I know my syntax is all over the place there...

 

[jonrpick]

Good stuff, guys... Thanks.

Just curious, Jon.....why the bass channel?

Mainly, because I want to know what it sounds like. It's not a direct copy of that circuit. I'm planning to use two 12AX7's (as of right now). V1b will be a cathode follower to driver a standard Blackface TxB tone stack (fixed mids), followed by two gain stages. A true 6G6-B bass channel uses fixed treble through a voltage divider, in lieu of a pot. The actual treble control is tied in following the next gain stage. That's where my preamp differs.

After replacing the stock 1-knob tone in my MMB with a TMB, and seeing how much loss there is, I decided to go with this setup for the increased gain (true gain, not distortion is what I'm after).

I'll be playing with cathode resistors and caps to try to dial in the preamp.

The whole parallel SE thing is an effort to keep the bass tighter and cleaner when the amp is turned up. It's not an overall "loudness" issue. As it is, my Vibro Champ has a bigger (15W) OT running into an efficient Cannabis Rex, so it can keep up on stage when I do occasionally play guitar in one of my bands. It did fall apart on a louder song though.

As far as why this and not a push-pull setup, well, I like the harmonic content of an SE amp. Between the SE output section and the hotter preamp, I should be in harmonic heaven!

 

[printer2]

Say you use a 6V6 into an 8k ohm OT that is rated for the power, you will make 6W. Then if you are running two 6V6's you need a 4k ohm OT that is rated for 12W and, you guessed it, you will be making 12W. You are running the same signal (voltage wise) and you are pumping in twice the current. Therefor twice the power.

P-P is a little different as the second harmonic distortion that each output tube generates is canceled in the OT. And since they stated the power output at a distortion level (say 10%) then the amp can be turned up some more before the output has a distortion level of 10%. So a pair of 6V6's in P-P are rated as putting out more power than two tubes in parallel SE.

But ignoring the distortion level does the P-P set really put out more power if we ignore the distortion and crank the suckers? Not really. The P-P transformer might be a little more efficient that the SE one but not enough to really get excited about. So why can we get more power out of a pair of tubes in P-P than in SE?

Because the SE amp is running Class A (or at least sort of, with the bias skewed a little one half the waveform can still sort of resemble the waveform while the other half is clipped) and the P-P amp is running Class AB. Because in the Class A SE amp the tubes are always passing current, (theoretically) adding up the watts dissipated across them give the limit on how much power they can deliver.

In Class AB each tube is turned off for a time allowing the plate to cool down. This cooling down allows us to pump more power through them. So the P-P set makes more watts than the SE double barreled Class A amp, or the P-P Class A amp. Hope this makes sense.

 

[mozzman]

Barely perceptible change in sound takes twice the power. 3db. I would build a push pull 6v6, or 6l6 amplifier.

 

[jonrpick]

I would build a push pull 6v6, or 6l6 amplifier.

...which would kill the very harmonics present in a SE amp that I'm trying to preserve...

I'm only trying to gain a little clean headroom, not overall volume.

 

[mozzman]

KT88.

 

[jonrpick]

KT88.

Maybe a pair of 'em...

 

[JD0x0]

KT88.

KT150
http://www.thetubestore.com/Tubes/KT88-Tube-Types/Tung-Sol-KT150

 

[Wally]

And big capacitance in the first stage....no matter what power tubes one chooses.

 

[mRtINY]

Barely perceptible change in sound takes twice the power. 3db. I would build a push pull 6v6, or 6l6 amplifier.

Um, no.

Double the power is 6dB. Double either the voltage or the current and you get a 3dB increase....


-tINY

 

[jonrpick]

And big capacitance in the first stage....no matter what power tubes one chooses.

Are you referring to the filter caps? Think 40uf would do it?

 

[dankilling]

Over-filtering can also cause hum- separate the filtering stages info you want to add more than about 47uf at a time.

 

[robrob]

Are you referring to the filter caps? Think 40uf would do it?

Yes he is. I recommend you use a solid state rectifier because Class A amps are always pulling almost full power so there's very little voltage sag. With solid state you can go much higher in first filter cap so I'd do two 40's in parallel for 80uF at B+1. You need the extra filtering because push-pull removes ripple hum (common mode rejection) and single ended doesn't.

Using a choke between B+1 and B+2 would be a good idea for ripple hum removal for a single ended amp.

 

[robrob]

Over-filtering can also cause hum- separate the filtering stages info you want to add more than about 47uf at a time.

? More info please.