audio-talk

Thinking about it Stephie maybe started PP for linearity up to output stage where the SE magic works too.

But then there is the asf-2 circuit that Nick used when we drove the 833a quite hard. Reading the article again in last few days I must be crazy not having built it yet, distortion tiny, capacitances cancelled bandwidth phenomenal, output impedance 3 ohm.

Thats the issue for me Paul, all the good things about se operation are blamed on distortion. And I'm pretty sure that not true unless they are used near their limit. As nick has just described

That's not entirely what I said, I was just saying that SE and PP distortion was different, I certainly didn't say that SE didn't distort, far from it. But I don't say that all the good things are due to distortion (not that you said I did), I say the good things can sound good despite distortion, but the distortion is a problem and that's partly why I don;t use one at the moment.

I'd put it to you that the reason you want to turn your push pull amp up is that most push pull amps cancel a small part of the signal.

And I'd put it to you that if they did something as simple as that it would be easy to measure. I think that a lot of the failings of push pull is down to the phase splitter and not push pull itself. The output stage of the DAC I made (and Mark has one off) is push pull remember. Also I remember Marks build of the Olson PP amp didn't sound like what I agree some PP amps sound like. I think there is also merit in SE amps not taking the transformer flux through zero.

There are many things solid state I have found that sound bad, once you have got past gross distortion (which is relatively easy, but it needs to be a lot lower that you might expect) all sorts of other temporal things start to be heard, memory effects and that sort of thing, but they can be solved. Simple valve amps generally avoid some of those sorts of problems so if they are operated in the generally linear area they will be easy to make good (after you sort out power supplies and slew rate and all the other things that matter). Simple SS amps like the MoFo also have many of the same characteristics.

Also to throw into the "turn up PP amps" mix, remember that the ear tends to translate distortion into loudness, so its easy to fall into the trap of turning up the volume until something distorts enough to sound loud.

Which is why I said that the KT120 sounds loud even when the volume is low, and is the loudest sounding of the beam tetrodes I’ve tried when strapped as a triode. A lot of 2nd and 243 times less third. The ear hears this as loudness: impressive in the short term but not so good for long term enjoyment.

If the ratio of 2nd to third is closer then the more ‘air’ and ‘sparkle’ and musicality there seems to be in the sound of an SE triode amp. Hence that’s why the 45 sounds so damned great and is so very listenable long term. Closest 2nd to third ratio. It’s not a ‘loud’ valve. It just ‘is.’

I expected to be shot down in flames and it didn’t take long until that happened, but I stand by it. Pure conjecture, totally unscientific, simplistic irritating and based on dodgy data it may be, but I hear what I hear and if you believe those sims then it all falls into place. Single-ended KT77s sound far better to me than either SE EL34s, KT88s or KT120s and of course the sims answer why that would be too.
KT150s would be very interesting to try. I predict from those sims that they will p*ss all over the rest of the big tetrodes, for naturalness and musicality. Less sparkly than a DHT but probably extremely liveable with long term. Akin to a KT77 on steroids. All that without even hearing one. But I’m fairly confident.
I specified a 1 Watt level because it would hardly be appropriate to try to simulate a 45 at ten watts output power.

I specified a 1 Watt level because it would hardly be appropriate to try to simulate a 45 at ten watts output power.

Yes, agreed, but its equally off to use a kt150 at 1W, I would have though a fixed percentage of max anode dissipation would have been more informative.

i've put the last link I posted here into the stacks section because I'm thinking it was missed here because I edited my post and it got quickly passed by. I do beleive it's relevant and personally thought it was a good overview of the whole distortion thing.

ed wrote: ↑Thu Dec 30, 2021 9:40 am i've put the last link I posted here into the stacks section because I'm thinking it was missed here because I edited my post and it got quickly passed by. I do beleive it's relevant and personally thought it was a good overview of the whole distortion thing.

Just read that. Seems to me that the key to avoiding complex and IMD type stuff is headroom, headroom, headroom and just in case there isn’t enough: yet more headroom. This either comes from high undistorted power, or extremely efficient speakers, plus minimising the number of amplification stages

Just thinking out loud with reference to that article. With a single ended 300B amp for example, where people regularly claim 8Wpc, you might have a 3W amp at most, before things start getting silly on complex music, 2A3? 1W, and a 45….er…..never mind.

Obviously 98dB speakers are not going to be enough with most of the popular or rare low power DHTs (71A, 45, 2A3, PX4) in a reasonably sized room. 107dB for those perhaps. Hence Steve’s big baffles for example.
I’m just thankful I have such a tiny room, with big efficient speakers that I deliberately designed so as not to overwhelm the space with bass, or else the performance of my flea watt 45 amp would be completely unable to be realised.
If I still had my big Mets, (94dB) no SET apart from KT150, 845, GM70 or 211 would be a realistic proposition for proper low distortion music reproduction. Thinking back, it was only the small size of the room that allowed me to just about get away with 2A3s with the Mets. Explains the push-pull obsession I developed when I had them.
Thanks for that Ed. It brought a few of my scatterbrained sonic thoughts together quite nicely.

It's not headroom it's staying on the linear range that's needed. I have said for years 1w is the most you should expect from a 300b

To add to that. The class b examples in that doc show why headroom on its own is not enough. Not that anyone actually does class b.

Nick wrote: ↑Thu Dec 30, 2021 11:01 am ….1w is the most you should expect from a 300b

I suppose As usual I wasn’t being clear. Staying linear is what I meant i suppose.I thought that with big headroom (a wide, linear acceptance envelope for music signals) one that copes easily with peaks, it should be easier to stay within that sweet spot range.
I know I’m not using the proper technical jargon, for which I apologise, but it’s a difficult world to inhabit and nigh on impossible to explain your thinking when you are to all intents and purposes a layman when it comes to electronics. I know enough to get by and be able to build things that make a decent sound when connected to a pair of speakers.
Apart from the aforesaid field of speakers, which I do know about, the rest of it baffles me.

I think it would probably be best if I don’t talk about things I don’t know enough about. I’ve been doing it for years on here and isn’t the definition of lunacy, repeating the same process over and over and expecting a different result? Shutting the FU when I don’t know what I’m on about, will make a good new year resolution. Keep quiet, observe and learn. I’ll make that my signature.

Let’s see how long I keep it up.

ed wrote: ↑Wed Dec 29, 2021 5:40 pm The discussion around H2 and H3 has been going on in fet circles for ages. I can't lay my hands on it but there is a paper discussing the various merits of each distortion, I think by Nelson Pass.

anyways, I think this is a relevant discussion:

https://www.diyaudio.com/community/thre ... on.330451/

There is also the H2 generator which Nelson released which goes some way to scratching the itch which some have:

https://www.firstwatt.com/pdf/art_h2_v1.pdf

edit...found it:
https://www.passdiy.com/project/article ... d-feedback

Thanks Ed, I must admit to not taking the time to read those links.
I'd read the last one before, but it was good to read it again,

Especially Nelsons conclusion

If you want the peak distortion of the circuit of figure 13 to remain below .1% with a complex signal, then you need to reduce it by a factor of about 3000. 70 dB of feedback would do it, but that does seems like a lot.

By contrast, it appears that if you can make a single stage operate at .01% 2nd harmonic with a single tone without feedback, you could also achieve the .1% peak in the complex IM test.

I like to think the latter would sound better.

I've always gone by what I think sounds better, and learnt not generally to trust what others may write, but all these links appear to be similar to my idea of why some things sound better to me.
And like Steve, I'm no expert in technical jargon, but that does not stop me identifying what, and eventually why, some things sound better to me..

Yep, but Nelson conclusion is:

I like to think the latter would sound better.

Thats not the same as "does sound better". Just saying.

Something those links do ignore is that loudspeaker are not restive loads. Feedback does more than just reduce distortion. Look at the MoFo, that works because of the internal feedback of a source follower creating a low output impedance.

Nick wrote: ↑Fri Dec 31, 2021 12:21 pm Yep, but Nelson conclusion is:

I like to think the latter would sound better.

Thats not the same as "does sound better". Just saying.

Something those links do ignore is that loudspeaker are not restive loads. Feedback does more than just reduce distortion. Look at the MoFo, that works because of the internal feedback of a source follower creating a low output impedance.

Of course he's been careful with his wording nick, one thing we have all learnt of the years is that there's more to any amplifier than the topology in how it may sound.

one thing we have all learnt of the years is that there's more to any amplifier than the topology in how it may sound.

Exactly, which is why I have been arguing that this whole thread fails for that very reason.

If it sounds OK and YOU are happy who cares. There are many things you can do with Tubes apart from water and liquids although some tubes correctly treated and modified could become a small acquarium! From illumination and heating to miniature amps and Tesla Megavolt generators the imagination is the limit. I don't have spice apart from what I add to fish etc, just look at the curves and take it from there.
For the 0.0001% distortion it has to be sand amps, and then the best speakers you can afford. Forget source material, a digital joke these days. Many years ago I used to record only 2 tracks at a time from an album and record them on a dolby deck, (near the centre one track at a time!). More than 2 tracks and I could notice the tracking error distortion. I used to listen to the distortion not the music!
Enjoy your "Valves"!