Too bad... 😀
(Just kidding) 😉
Do you like how it sounds? Enjoy. Don't like? Search for real reasons why, but 5% of sag can't be a reason.
I haven't got the second mono block built yet, so it's too early to say if I like how it sounds, I was just thinking that 20V may be a lot for hifi. My baby huey only sags about 2 volts, that's what curious at this point.
I'll continue on with the second channel.....thanks everyone.
Think again, Boywonder
(sorry...)
I think you can do better. For EL34s in triode AB, biased as you have (see attachment), there should only be about 40mA current increase from quiescent to full-signal sinewave (You should also be able to approach 30W output). 20V sag for 40mA current increase is 500 ohms equivalent DC resistance! You should be able to get close to 100 ohms with good transient performance and decent DCR parts. The problem is your voltage tuning of a too-high voltage transformer down using the 2u2 1st capacitor. That introduces the extra sag.
Cheers,
Michael
(sorry...)
I think you can do better. For EL34s in triode AB, biased as you have (see attachment), there should only be about 40mA current increase from quiescent to full-signal sinewave (You should also be able to approach 30W output). 20V sag for 40mA current increase is 500 ohms equivalent DC resistance! You should be able to get close to 100 ohms with good transient performance and decent DCR parts. The problem is your voltage tuning of a too-high voltage transformer down using the 2u2 1st capacitor. That introduces the extra sag.
Cheers,
Michael
Attachments
Thank you Michael Koster......So what are my options? increase B+ a bit by a larger first cap and decrease bias current or get another transformer with lower V and higher current?
My PS caps for the 380V are a 2.2 uf Solen 630V, and a 220 uf Panasonic TS-UP 500V. The choke is a 8H 150R 125ma.
My PS caps for the 380V are a 2.2 uf Solen 630V, and a 220 uf Panasonic TS-UP 500V. The choke is a 8H 150R 125ma.
Last edited:
The secret is to put a mercury rectifier in front of the regulator, then it is pure sweetness😀
Seriously, aren't we talking about a p-p plate supply here? Maida is efficient, cheap, and much stiffer than a bunch of choke DCR. It really shines on bass-heavy electronic type music. And it performs pretty well, too. It has limitations which are easy to understand and design around, as well.
You'd have to be a masochist to shunt regulate a p-p plate supply of any significant size.
have you check the caps for leakage......also what output load are you using and which way round is it.
Well I conducted a few more experiments.....I paralleled a 20uf onto the first 2.2uf film cap, that raised the B+ to about 420V, and put a 10K R in place of the 6.8K to keep the volts on the driver about the same as it was. I still had about 15V of sag. Then I bypassed the CL-090 on the hot mains, that raised the B+ to about 426V with no change in sag.
So I buttoned it up the way it was, and listened to some music (Richard Thompson & Tanita Tikaram). It sounds quite good, although it's one channel.
I then connected a voltmeter to the B+ and another voltmeter to the speaker terminals while playing music at a pretty loud volume (speakers=90db) and the B+ voltage hardly budged (2-3V). The output voltage was averaging about 5V, so about 3+W. It's amazing how little power is really needed.
So I buttoned it up the way it was, and listened to some music (Richard Thompson & Tanita Tikaram). It sounds quite good, although it's one channel.
I then connected a voltmeter to the B+ and another voltmeter to the speaker terminals while playing music at a pretty loud volume (speakers=90db) and the B+ voltage hardly budged (2-3V). The output voltage was averaging about 5V, so about 3+W. It's amazing how little power is really needed.
I'm surprised changing the first cap didn't reduce the sag more (based on PSUD2). It sounds like you are down to mostly the DCR of the components at this point. The higher B+ might give you somewhat higher power and current draw at clipping. The good news is it's probably not such a bad thing to have a few hundred ohms DCR in the power supply of a "moderate" class AB amp at this power level.
Cheers,
Michael
Michael: I also thought that a much larger first cap would make a big difference (only because I had tried everything else that I could think of).
I'm still puzzled about the voltage sag because my baby huey has next to nothing for sag and both supplies are fairly vanilla CLC topologies with SS rect, Panasonic TS-HA or TS-UP caps and hammond 159 series chokes.
I've also convinced myself that the 1 Kz sine wave test at clipping is quite a bit more severe than listening to fairly loud music on my 90 db speakers. Also, when listening to music and watching the B+ voltmeter, it appears that the 220uf reservoir cap is able to keep the B+ up pretty well.
I'm still puzzled about the voltage sag because my baby huey has next to nothing for sag and both supplies are fairly vanilla CLC topologies with SS rect, Panasonic TS-HA or TS-UP caps and hammond 159 series chokes.
I've also convinced myself that the 1 Kz sine wave test at clipping is quite a bit more severe than listening to fairly loud music on my 90 db speakers. Also, when listening to music and watching the B+ voltmeter, it appears that the 220uf reservoir cap is able to keep the B+ up pretty well.
Last edited:
One remaining question is your actual B+ current draw, at zero signal and at clipping. What's the DCR of your transformer? It all should be predictable in PSUD2 with the correct inputs.
as i have seen one in your tread can i ask, which type of diode would work best as in g2-anode strap.
(if i remember right this helps to stop capacitance problems between the anode and the grids)
(if i remember right this helps to stop capacitance problems between the anode and the grids)
Pointy: I'm still a newbie....so I'm following Poindexter's Schematic which uses a 2A 100V schottky in series with a 332R for tying g2 to plate
50ma bias current at no signal, 60ma bias current at clipping per tube. Plus 9 ma total for the driver tubes (which I didn't measure-short a couple voltmeters)
so 109ma at no signal and 129ma at clipping total.
The transformer is 360-0-360, with 180 ohms R across the ends, and 90R each end to center tap. 378V unloaded.
PSUD model attached. My actual first cap is 2.2, not 2.5u as in the model. The transformers are vintage, so they are probably rated at 117VAC and my house is easily 120VAC+. The diodes are Fairchild stealth 1200V/8A.
Edit: The more I think about this, the more this doesn't make much sense to me. if the current is only going up by 20 ma total at clipping, I would assume that the transformer could live with that easily......
Attachments
Last edited:
- Status
- Not open for further replies.