• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Ultra Low Cost All Tube Headphone Amp Idea

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've been playing with a few things, trying to come up with a super cheap all-tube headphone amplifier and I think I may be very close. I eliminated the standard high voltage power supply in lieu of a "High voltage boost converter" from Ebay for $6. Using 12vdc input and it produces 250vdc while drawing 2A per channel from the source (that also includes filaments). It DOES require a ~100uF cap on the output (maybe less?) to clean up power supply noise. 100uF is all I had on hand. I highly recommend a dual monoblock configuration for stereo though. One boost converter WILL power two channels but it gets a bit warm when doing so. It's a pretty standard 12A?7/6bq5 single ended type of situation. I built my test setup with a 12at7 (again, just what I had on hand.) My main concern is with the output transformers. I found some Xicon 42tm114-rc signal transformers on Mouser for $2.70 each. 4.6k primary and 20ohm secondary. They're only rated for 200mW but they do get painfully loud. And I MEAN painful. We're talking plain-jane low impedance headphones though. There's another OPT that I'm also planning on trying for higher power needs. The Visaton 243-tr6.8. $10 each for that one. More expensive, but it may be needed in some circumstances like loudspeakers or maybe for power hungry headphones. The 6bq5 filaments can be run in series off of the 12v supply along with the 12a?7's filaments in series internally, so no separate power supply is strictly necessary for that. An inrush limiting resistor and slow blow fuse probably wouldn't be a bad idea on the power supply either.

So anyway... I'm not saying that the sound is "special" or better than this or that other amp. What I AM saying is that for the price this little amp sounds fantastic, and could very well open the door to people that don't have a ton of money to throw around on a fancy factory/custom built amp. It can easily be built for under $50, especially if you already have the tubes. Anybody have thoughts? Improvements? I just cobbled the design together from other schematics that I found on Google. I don't claim to be an expert or anything. Maybe somebody has already done this. Dunno. I looked around and couldn't find much info on those little power supplies. Hope someone finds it useful.

Included is an image of the boost converter I used (Search "High voltage boost converter" on Ebay) and a hastily scrawled schematic that I hope you can read.
 

Attachments

  • s-l1600.jpg
    s-l1600.jpg
    98.8 KB · Views: 306
  • IMG_20190722_020130119.jpg
    IMG_20190722_020130119.jpg
    565.4 KB · Views: 295
Last edited:
You'll need some good additional filtering on that PSU.

I've used 2, and found them noisy, a kettle on the boil sound, very odd. They seem stable enough at low power, but the fly back diode will blow with sustained loading above about 10W. Replacing that diode removes that mode of failure, but the next thing to go is the transformer (or FET). I'm unsure because either way the primary circuit winds up shorted, and I haven't removed the FET to check it off the board.
 
The datasheet for the OPT states that the unit is hipot tested at 2100v dc for one minute. I was wary about that myself though. I'm curious about the possibility of coupling the OPT to the 6bq5 in the same way that the 12a?7 is coupled to the 6bq5; through a coupling capacitor. That way, the OPT does not have to be connected directly to b+ and the 6bq5 draws power either directly from b+ or through a resistor. As a side note, I originally ran it at 335vdc. Didn't realize that I had it turned up so high. No ill effects on the transformers, though the sound was extremely bright and harsh with muddy lows. Sounds much better at the correct voltage.

This does not require any substantial or expensive filtering either. A single 100uF filter cap. It's quite possible that a substantially smaller capacitor would suffice, but I just didn't have one to try it with other than a 7.5uF, and that wasn't quite enough. It's dead quiet as tested, though I can also hear a "tea kettle" whistling sound without the filter.

In any event, I'm not finished with this design. I just had it running for about an hour and was so impressed with how well it worked that I had to share. Most tube designs out there are intended for very high-end "best of everything" use, and that prices a lot of people out. Chokes the size of bricks, 10uF solen film caps, fancy pants ALPS pots, oxygen free transformers that are hand-wound by elven mages and coated with unicorn snot lacquer from venus... No doubt whatsoever that the high end stuff is truly astonishing, but at some point you've got to stand back and realize that somebody took that design all the way to plaid when light speed would have been fine for a lot of people.
 
Last edited:
The datasheet for the OPT states that the unit is hipot tested at 2100v dc for one minute. I was wary about that myself though. I'm curious about the possibility of coupling the OPT to the 6bq5 in the same way that the 12a?7 is coupled to the 6bq5; through a coupling capacitor. That way, the OPT does not have to be connected directly to b+ and the 6bq5 draws power either directly from b+ or through a resistor. As a side note, I originally ran it at 335vdc. Didn't realize that I had it turned up so high. No ill effects on the transformers, though the sound was extremely bright and harsh with muddy lows. Sounds much better at the correct voltage.

This does not require any substantial or expensive filtering either. A single 100uF filter cap. It's quite possible that a substantially smaller capacitor would suffice, but I just didn't have one to try it with other than a 7.5uF, and that wasn't quite enough. It's dead quiet as tested, though I can also hear a "tea kettle" whistling sound without the filter.

1) the tea kettle whistling can be filtered with 100uF as you say. However the variable frequency burps created by the supply pulse skipping, or dropping frequency, results in a variable ripple waveform which is not sinusoidal, very audible, and can be anywhere between 10Hz and ultrasonic (or so it seems). I found that even with 220uF this was still present, and NB 220uF is too much, the in rush current makes the PSU chirp loudly (it doesnt like large caps one bit) I ended up ditching the cheap Ebay PSU as I had motor boating type sounds which were only isolated once I removed the supply and used a conventional PSU.

2) For my employment, I work testing electrical machines and transformers.

If a transformer is flashed/Hi-pot tested at 2100V this means:

Flash voltage to BS EN standards is twice working voltage plus 1000V - thus, the transformer is rated for a peak voltage of 1kV (withstand not operation).

I would have no qualms using it at 250Vdc B+. (Just ground the "earthy" end of the secondary, and fuse the primary appropriately)
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.