Power Amp Arrangement

General Tech Discussion - Anything amp-ish goes!

Moderators: zaphod_phil, Daviedawg, Graydon, CurtissRobin, colossal

Post Reply
User avatar
HenryMars
Unrated
Posts: 7
Joined: Sun 12/17/06 2:00 am
Location: Unknown

Power Amp Arrangement

Post by HenryMars »

I have been toying around with a few ideas on power amp design using 2 EL84's. I want a Class A output. I was looking at a push-pull class A arrangement but I think I would rather parrallel connect the power tubes. The only disadvantage I can think of off the top of my head is that the increased plate current and lower total Zp will increase the cost and weight of the output Xfmr. Any thoughts?
0 x

Francis_Vaughan
Occasional poster
Occasional poster
Posts: 347
Joined: Wed 03/29/06 2:00 am
Location: Adelaide, Australia

Post by Francis_Vaughan »

There are a few things here.

The alternative to PP is Single Ended. This brings with it a number of complications. A PP topology has no static magnetic field in the transformer core - this is because when quiescent the bias current in the two halves is the same, but runs in opposite directions, cancelling out.

On the other hand, a SE topology has a significant static field. Indeed, if the system is biased properly it will have a static field half the maximum possible. Worse, the field never reverses (which it does in a PP) - it can only vary from zero to maximum. This means that for a given swing in magnetic field the SE transformer will see a peak magnitude of magnetisation twice that seen by the PP transformer. To avoid saturating the core of the transformer the core must have twice the cross sectional area. This can lead to a transformer that is up to three times the weight. And the price pretty much goes up with that. In order to avoid the dangers of saturation the transformer is usually air-gapped - which can allow a bit more leeway in the design.

Sonically PP and SE sounds different. As a broad generalisation a symmetric (PP) is half wave symmetric, and leads to cancellation of any even order harmonic distortion products generated in the output stage. The SE stage has no such symmetry and thus has a higher proportion of even harmonics in its output than the PP does. As a rough guide the even harmonics are smoother, warmer and more consonant in sound. The odds are more edgy and brasher. But bear in mind that the 3rd harmonic is nothing more than an octave plus a fifth, so it isn't a trivial thing to nail down.

There is also the "true" class A issue. Suffice it to say that the vast majority of class A PP amps are not true class A and are much closer to heavily biased class AB. SE can't be anything but true class A. (OK, you could under bias, but it would sound utterly broken.)

When I got interested in amps I was sure that SE and warm sound was for me. Until I started listening. That sound in my head - my sound - turned out to be a pair of EL84's in PP. :D
0 x

User avatar
HenryMars
Unrated
Posts: 7
Joined: Sun 12/17/06 2:00 am
Location: Unknown

Post by HenryMars »

I kind of want to parallel the tubes. I have been researching transformers. The one that I like for the job is a Hammond 1627SEA, but it is expensive. Are there any more resonably priced SE Xfmrs? The 1627SEA seems to be well shielded.
0 x

Alexo
Verbose poster
Verbose poster
Posts: 1676
Joined: Thu 09/21/06 2:00 am
Location: Upstate NY, USA

Post by Alexo »

You can always get a "Universal Single Ended Output Transformer," a 125ESE ought to handle a pair of EL84's for under 40 bucks, they are less hi-fi than the 16xx series, which some take as an advantage.
0 x

Post Reply