Preserving even order harmonics in push-pull

[bluefoxicy]

It seems AB1 push-pulls remove even-order harmonics (octaves) and leave odd-order harmonics (shit I don't want to hear). Why, and how do I intentionally cause/enhance even-order harmonics and/or reduce odd-order harmonics? I want to try to make a Class AB1 act like a Class A.

[Bruce / Mission Amps]

It seems AB1 push-pulls remove even-order harmonics (octaves) and leave odd-order harmonics (shit I don't want to hear). Why, and how do I intentionally cause/enhance even-order harmonics and/or reduce odd-order harmonics? I want to try to make a Class AB1 act like a Class A.

I'm pretty sure the harmonics that are eliminated in PP are the ones generated as distortion in the PP devices themselves... not the harmonics the amp is reproducing from the instrument or previous stages... if it did they would all suck ass as far as guitar amps go.

[Steve Conner]

The best way to enhance even harmonics is to use a single-ended output stage, IMO. Or if you must use a push-pull, use mismatched tubes.

In practice, I think the preamp tubes generate plenty of even harmonics that complement the odd harmonics from the cranked power tubes. The odd harmonics probably help add a bit of aggression to the tone.

BTW, Bruce is right, the power tubes only cancel their own even-order distortion harmonics, and then only if they're matched and driven by a balanced PI. They don't actually suck the even harmonics out of your tone.

[mooreamps]

I want to try to make a Class AB1 act like a Class A.

The way to do that is ; remove or reduce the signal drive to one of the power tubes.

-g

[Rob Mercure]

Bf,

To elaborate a little on the above - this is probably the greatest difference between a HiFi amp and a musical instrument amp (well, other than the vast use of precious metals and exotic materials <grin>). The HiFi amp's output section is "balanced" to infinity - or as close to infinity as the HiFi enthusiast's budget can get them (higher than their budget should get them IMHO) while, as Steve alluded, a guitar amp's output section's "flavor" is greatly influenced by the degree of AC - "signal" inbalance (DC balance is desireable in the output transformer). Or the "short hand" that I like to use: Music "production" vs. "reproduction." And obviously we're on the production side.

Rob

[bluefoxicy]

I'm pretty sure the harmonics that are eliminated in PP are the ones generated as distortion in the PP devices themselves... not the harmonics the amp is reproducing from the instrument or previous stages...

Right. I tend to rely on power amp distortion more than preamp but have been working on a multi-stage preamp overdrive to sweeten up the tone.

In practice, I think the preamp tubes generate plenty of even harmonics that complement the odd harmonics from the cranked power tubes. The odd harmonics probably help add a bit of aggression to the tone.

I probably should make a GAIN tube stage. :|

BTW, Bruce is right, the power tubes only cancel their own even-order distortion harmonics, and then only if they're matched and driven by a balanced PI. They don't actually suck the even harmonics out of your tone.

THD in Class A1 SE for a 6V6 is 14%, in Class AB1 PP it's 4%. Some of us rely on power tube distortion, esp. for blues and rock (EL84 crunch).

The HiFi amp's output section is "balanced" to infinity - or as close to infinity as the HiFi enthusiast's budget can get them (higher than their budget should get them IMHO) while, as Steve alluded, a guitar amp's output section's "flavor" is greatly influenced by the degree of AC - "signal" inbalance (DC balance is desireable in the output transformer).

Explain? Should I dim out the signal a bit to one tube, use different manufacture tubes (Tung-Sol vs Electro-Harmonix, look at the innards, shaped differently and all), etc?

[daz]

I've heard this notion before about mismatched outputs creating more complex tones. What i don't understand is if this is true, why has the GT notion that matched tubes is the way to go lasted this long and still not been knocked down in the consumers mind? I mean, we still have all these tube retailers offering matched tubes and people still lean that way. If it's BS i would have thought by now it would have long been proven wrong. Would we all be better off asking for mismatched tubes? I'm not disputing any of this, just asking out of total ignorance. Which is correct? Is it just because players are stubborn when it comes to change and have believed in matched tubes too long now to consider the opposite as an option?

[bluefoxicy]

I've heard this notion before about mismatched outputs creating more complex tones. What i don't understand is if this is true, why has the GT notion that matched tubes is the way to go lasted this long and still not been knocked down in the consumers mind?

See Global Warming.

Tubes that get matched within 5% fall within tolerance of a system using caps and resistors with likely 20% tolerance for vintage; these days I have lots of gold and red band (5%, 2%) tolerance stuff though. The matching is mostly futile because tubes drift based on temperature and on performance of the components they're hooked up to; and over time, wear unevenly from use and start to fall out of matching.

On the other hand, severely mismatched tubes can misbehave horribly. They can produce scary, grated, horrible tones and just sound like trash.

I'm not sure if having imbalanced tubes is a good idea, though I don't see why not to ask. I can see why having an absolutely matched pair is futile, an why slightly mismatched tubes would be fine or even interesting.

[Enzo]

daz, you are confusing two issues. Matched tubes will balance an output stage, so hum cancels, and potentially the power stage will have better fidelity.

On the other hand, some folks like the sound when things are NOT balanced. They will have more hum and less fidelity, but they like it that way.

it is not a matter of one is right and the other is wrong.

[Steve Conner]

A long time ago, I built a push-pull amp that had regulated B+, so it wouldn't hum whether the tubes were matched or not. It also had independent bias pots, so I could put any tubes I wanted in the two sockets, and trim them both to reasonable idle currents. Some of the combos I tried were:

One EL34, one 6L6
One 6L6, one 6V6
One 6L6, one KT66
One EL34, one EL37

And so on. At the end of the day, I decided that it didn't really make a massive difference, but I liked the tone best with two power tubes the same, and reasonably matched. Some of the combinations made interesting dirty tones, but they didn't clean up well at all.

I ended up using a matched pair of Tung-Sol reissue 6550s.

http://scopeboy.com/toastpix.html

[dai h.]

leave odd-order harmonics (shit I don't want to hear

IMO it seems to me the premise is wrong. Only evens would probably sound bland and uninteresting. I'd say you need odd harmonics and getting something to sound good is more complicated than a matter of getting rid of the "shitty" harmonics. Maybe taste would be a good analogy. All bitter might taste bad, but a bit of bitter mixed with sweet might make something taste better and more interesting than sweet only. I think you have to be careful not to oversimplify things, like "Class A good, Class AB worse, or the lower the cable capacitance the better (always), or new stock tubes bad, old stock tubes (always good), expensive parts better cheaper parts worse", and so on.

[black_labb]

i'd say that all even order harmonics sound a bit dull, where all odd order sounds too dirty and harsh. you need some odd and some even to get a good tone that isnt too boring or too harsh. how much of each depends on your personal oppinion. so, do you know that you like SE or PP, or are you just guessing?