audio-talk

izzy wizzy wrote: ↑Tue Jun 08, 2021 1:01 am Did that make any audible difference? I've heard how technically a full wave bridge utilises the transformer better; more efficiently as it's always doing something rather than one half being cutoff each half cycle. Of course changing the transformer and the rectifier config is two things at once but I had to ask.

It’s interesting.
The toroidal transformer I had in there was a 300VA job, which let’s be honest here, is bloody ridiculous. I mean there’s derating and there is stupid.

Back on subject: after a bit of comparison, the sound with big toroid and hybrid bridge has the better bass. The centre tapped transformer/rectifier gives better midrange and treble but the bass is not as good. The centre tapped rectifier/transformer makes more of the strengths of the 45, so the trade-off works for me.

As for the technical reasons behind the differences; apart from the lower source impedance presented by the big toroid I don’t know.

I used to use toroids but compared to EI types, found the toroids let though more hash so EIs sounded a bit more natural. This was in preamps so bass not really affected.

I dont buy the theory that bridge is better because full winding of transformer is conducting throuhout. Always used the benefit of a centre tapped secondary for what it is. Fit for purpose. Ensures I can full wave rectify with 100% valve rectification. All my power supplies are made this way and one thing I dont have to reinvent is that. Neither do I buy the hybrid rectification of transformers not evolved for valve rectification.

All my se amps had traditional historic power supplies with choke input as a MUST. The sound of different rectifier valves have always affected the sound of an SE amplifier. I doubt the addition of silicon is a benefit to these traditions. Breaking down an se amp built this way, the music is played by the interaction between the valve which rectifies, and the single output valve.

Go a step further and valve shunt rectify, eliminate power supply capacitors then the music is largely played between the shunt valve and the output valve, and the valve rectifier plays a part, with no capacitors to influence the sound, whose influence is much greater than tube rolling. My last capacitorless amplifier Mo asked me to add a black Gate wkz. It sounded considerably worse. It wasnt required and it altered the sound for the worst.

the winding arrangement of the transformer never gave me a problem.

What's discussed in this article makes even more sense(to me) when "sound" of transformers or psu" becomes a topic.

https://richardsears1.wordpress.com/201 ... teraction/

A discussion would be interesting.

Right!
After yesterday’s session at Steve’s and the sheer fabness that were the globe 45s, I got to thinking about whether I could get the STs somewhere near their performance. Could I use the tone of a different driver valve to make up the performance deficit? It turns out that yes, I could. OK, it may seem a quick 24 hour turnaround, but I’ve put a hell of a lot of thought into what I’m about to describe below.

It has to be remembered that I don’t have a lot of money, and simply can’t afford the exotic output transformers that might have made the job of approaching the sound of the globes a lot easier. That’s just the way it is so no point dwelling on it. With expensive OPTs right out, it fell to driver valves to provide the answer.

I whittled it down to two choices: ECC81 or ECC88. ECC88s are nice and linear, but in the past, I’ve found them a bit sterile sounding and besides, my HT is a bit too high for them. I could have used 6n1P, to ameliorate that, but I’m not that keen on them either; a bit dark for my liking. So almost by default it became the ECC81 as driver.

I can hear the groans now, but that’s all I have available. They looked reasonable with a 15K load. So I proposed an LED biased scheme running at 200Va, 2Vk and 7.5mA current. I’ve used one half per channel, which means that there is one valve less on the top deck. I also decided to shorten significantly, the signal paths, by putting the noval socket directly between the two 45s and turning one of the UX4 sockets round so that the filament pins go on the outsides and the plate and grid pins go to the insides. Everything is then able to connect directly using just the component leads, which go point-to-point and are as short as possible.

I’ve used my prized GEC A2900 as the driver valve. Although this valve has very similar electrical characteristics to an ECC81, it is actually an industrial computer type valve like the 5687, but with a gain of 60. It is more linear than the ECC81, is rare as hen’s teeth, goes for £200 NOS and guitarists kill for them as reverb drivers.

The result of the work is a much closer approach to the sound of the globe 45 than I had yesterday. The A2900 is great with the ST45, giving a big soundstage and a clear, crisp and clean sound with just the right amount of warmth and space to allow an ST to achieve 95% of the sound of a globe.
Next job is to find something to use on the front part of the top plate, to cover up all the holes. For now though, I’m well chuffed.

Im glad you've found a better sound with the ST’s. Im living on very old memories but I dont recall the difference from ST to globe was quite so transforming. It was the most dramatic tube rolling Ive ever heard, and Ive done a lot of it. So It seems you may have solved it.

Cressy Snr wrote: ↑Sat Jun 12, 2021 11:13 pm Next job is to find something to use on the front part of the top plate, to cover up all the holes. For now though, I’m well chuffed.

Maybe move the knobs to the front (cue DtB)? Perhaps replace them with some skirted retro types that fit snugly into the holes in the top plate?

...knobs....snugly.....holes.....

Trying to resist.... its not easy...aaarrrgggh!

Paul Barker wrote: ↑Sun Jun 13, 2021 6:55 am I’m glad you've found a better sound with the ST’s…
…So It seems you may have solved it.

Yep, it seems to have worked out nicely.

Ray P wrote: ↑Sun Jun 13, 2021 7:50 am
Maybe move the knobs to the front (cue DtB)? Perhaps replace them with some skirted retro types that fit snugly into the holes in the top plate?

I’ve decided to use something I used years ago: guitar pick-guard material. I didn’t use much on the above amp (the pearlescent red strip) but it is tough stuff and looks very good if implemented properly.
It comes in blank sheets 18” x 12” for use by custom guitar builders. The aim is to make all the bolts invisible apart from a half inch margin all round where the top plate attaches to the wood frame. It won’t be pearlescent red, it’ll be more subdued this time.

AFAIK it's called traffolite. At least thats what it was called from Pearson's engravers in Halifax.

Thats the stuff. We used to use it for all the labels on control panels.

Ali Tait wrote: ↑Sun Jun 13, 2021 3:28 pm Thats the stuff. We used to use it for all the labels on control panels.

Ditto.

Spent a while today doing a mod to the rectifier scheme, so that either 5 or 4V rectifiers, can be used depending on which socket you plug into. The 5V octal socket is at the rear, with the 4V socket in front of it. A Mazda UU4 is in use at the moment. The octal socket for 5V rects can just be seen at the back.

I like these Mazda UU* rectifiers. I’m a sucker for how they look. Don’t care how they sound

This UU4 works beautifully, allowing an even bigger soundstage. It’d be thrashed to destruction with 2A3s and whatever was driving them (max output current is 120mA) but it is cruising running 45s and an A2900 at 75mA total and seems to be in its element.
It’s a British/American valve alliance that certainly delivers the sonic goods. Very, very nice sound.

TBH, I’d forgotten just how valuable these little audio meet-ups can be.