HF AC DHT Filament Supplies

[pre65]

I think the 26 is a special case.

Certainly the DIY hi-fi DHT modules are very versatile and a cheaper but as versatile low noise board would be welcome. I believe they are (like Tentlab modules) current and voltage regulated.

AND, they need to come with printed instructions (PLEASE) like Mikes e-chokes did.

[IslandPink]

Long time ago someone e-mailed me saying he'd had great success, and had I tried it. (I hadn't.) 100 kHz I believe it was.

Both Paul and Nick have tried going down this route, but abandoned it. You need to get them involved & see what conclusions were reached, and if there was anything else to be tried .

[Paul Barker]

I didn't actually abandon it, the 212 HF heater supplies are still on the output stage I use for the 833a nowadays.

It wasn't easy. but it sounded OK.

[Paul Barker]

I wouldn't promote it above a DC supply. The original consept in those days was that ac heating may be superior, so if we lift the ac above the audible range we retain the benefits of ac but without the hum.

Not sure any of it made any difference. Not sure DC is bad.

My list to try would contain a Steve Bench AC heating hum compensator, rather than hf AC heating.

the 212 I had running at home when a number of people came over had no hum at all but was 50hz AC heated. Cancellation was at play. I don't think I have bettered that sound yet.

I think that amplifiers whose 50hz ac heaters are blended to cancel one another would be the optimum. But DC heating sounds good enough.

[Andrew]

Yes, I am, I need a few and so there's no sense in not doing it as a group buy, what VA do you need?

Andrew

0.00000001W up to 1234567890MW....... thats fair enough innit!

Meanwhile back in reality, maybe a board that can be adapted to work with everything from a 26 up to say2A3/300B types?

DTB

300B is no problem, that's my sweet spot, as is 6b4g, voltage will be programmable, has a nice slow start and there should be no need to adjust when valves are changed, its a current source too so should be as good as AC.Hope to have a listen soon, based upon thatI'll decide if its a go-er - might sound pants for all I know.

How much VA does a 26 need?

[Dave the bass]

Yes, I am, I need a few and so there's no sense in not doing it as a group buy, what VA do you need?

Andrew

0.00000001W up to 1234567890MW....... thats fair enough innit!

Meanwhile back in reality, maybe a board that can be adapted to work with everything from a 26 up to say2A3/300B types?

DTB

300B is no problem, that's my sweet spot, as is 6b4g, voltage will be programmable, has a nice slow start and there should be no need to adjust when valves are changed, its a current source too so should be as good as AC.Hope to have a listen soon, based upon thatI'll decide if its a go-er - might sound pants for all I know.

How much VA does a 26 need?

1.5V @ 1.05A.

A 45 too, thats another popular DHT, it'd be great if it could cater for all the regular suspects in DHT-land. Maybe.

[Nick]

The faulty assumption I found about HF supplies (and why I stopped bothering) was the assumption there is no hum, there is, its just at a very high frequency. It may be better if you are at the 300kHz range switch modes often work at, but the problem with 100k or so becomes apparent when you sweep the amp with a 15kHz to 20kHz sine wave. In theory my ears should hear nothing. In practice...

[steve s]

Yes nick I don't think there are any real answers to using AC on DHT,s

the higher the voltage the more the hum, as it makes up part of the signal, a few weeks ago I succumbed and bought a pair of tent labs modules for my next px25 build

[simon]

A momentous event steve...

[Dave the bass]

It's just not right I tell ya!

[Mike H]

It's quite possible that my e-mail correspondent who reckoned 100 kHz AC was "better" is only 'cause it was adding something to the output that he thought makes it "better". But is in reality HF hash heterodyning with something in the treble range. So it sounds like the treble is "more".

[Paul Barker]

it could easily be that Mike.

However Phil used SMPS on his 833a and Nick and I had SMPS on our joint 833a and on that day nobody ran out of the room when either 833a amps were playing. So whatever frequency smps works at seems to be viable.

SMPS's are a cheap viable alternative to extreme engineering DC power supplies which are kept to 100hz religiously as if any other frequency is anathema to audio.

The limitation with smps is the voltages available from the cheap ones, the lack of adjustment and the cold start current when a cold valve filament shocks the smps with a dead short. But sonically if those 833a amps are wrong I can't hear it.

[Nick]

Yes, but the SMPS in those cases were producing DC, not HF AC. I have found though that a couple of stages of CMC and caps will get the noise of most SMPS's down to the mv range.

[pre65]

Yes, but the SMPS in those cases were producing DC, not HF AC. I have found though that a couple of stages of CMC and caps will get the noise of most SMPS's down to the mv range.

Whats CMC ?

The SMPS on my 833a filament is a 12v one, and I use RCRC to get it down to 10V.

[jack]

Yes, but the SMPS in those cases were producing DC, not HF AC. I have found though that a couple of stages of CMC and caps will get the noise of most SMPS's down to the mv range.

Its not the DC output of SMPSs that cause most issues, its the EMI they radiate into the surroundings and the cr*p they put back onto the mains (which appears elsewhere in the system). The culprit is in the essence of SMPS operation - switching large currents through inductors - unless the inductor is screened and careful attention is taken in all aspects of design and use, they can be "Interesting".

Earth paths are also an issue for the same reason - large switching currents also need a return path...

For OP stages, not so much of an issue; for preamps, different story.

Some of my smaller SMPS designs have been used in commercial valve stomp boxes etc. It was a challenge,,,