
I always wanted to try using a power opamp as a class A2 driver. I finally got around to some experiments. I managed to have some sucess, especially using a 1636 (ala Darling amp) to get ~1.2W out. Still experimenting - I need to try push-pull, and some different tubes. Here's the info:
Class-A2 and a power opamp driver
Let me know if you can think of any candidate tubes. The power opamp can only put out about +/-36V @ 50mA, so stuff like 811's are not workable. So far I've tried the 1636, 2E26 (OK but not a lot of advantage over A1), and 6V6/6L6/6W6 (worse than A1). Seems like some of the newer semi-useless 9-pin-mini RF tubes might do well here?
Pete
Small power valves were usually either:
- not intended to take grid current, so the grid may overheat
or
- intended for Class C, so not especially linear
Exceptions are QQV series tetrodes (e.g. 6360 in USA), which have a rugged grid for Class C but are also intended for linear applications (mainly AB1). Another possibility is 5B/254M, but I would expect that to behave like a 6L6 or 807.
- not intended to take grid current, so the grid may overheat
or
- intended for Class C, so not especially linear
Exceptions are QQV series tetrodes (e.g. 6360 in USA), which have a rugged grid for Class C but are also intended for linear applications (mainly AB1). Another possibility is 5B/254M, but I would expect that to behave like a 6L6 or 807.
Looking at the grid drive 'scope picture on your website makes me wonder if the TI amplifier chips would have an issue with asymmetrical power supplies. If the supplies don't have to be symmetrical the supplies could be optimized for the tubes needs.
I am planning to build a se amp with a 1619 dht in triode mode. The datasheet shows the positive grid curves for the pentode mode. I manually checked out the triode connection in the a2 region and it looks promising. A +_25v swing with 2k load goes from 135 ma to 20ma. 180v,0 bias and 70ma quiescent current.
Alfredo
Alfredo
Looking at the grid drive 'scope picture on your website makes me wonder if the TI amplifier chips would have an issue with asymmetrical power supplies. If the supplies don't have to be symmetrical the supplies could be optimized for the tubes needs.
Good point - the IC doesn't care where ground is, so that would work well.
Pete
I just got a bunch of VT61A dual triodes that have ability to sink up to 17ma of grid current...
Is that the same as a DET19 or 2C34? I have a few of those, on my list to try. They seem like they would work great in A2.
yeah, close to the 2C34 apparently, although I'm having a difficult time tracking down a useful datasheet.
Would a 6aq5 work? I tested a preamp that used the 6aq5 triode strapped, on a B+ less than 60v the sound was quite clean on my digital designs ear buds.
The 3C24 looks like a great candidate. It could be zero-biased with symmetric grid swing, 550V B+ or so, and putting out 4-5 W Po, but needing an 18K Zpri. It may still want some feedback, but that should be easy to arrange with the opamp. The quiescent power at ~14W should light up the anode a little.
Cheers,
Michael
Cheers,
Michael
Attachments
Tried a few more tubes in SE this morning:
2C34 / RK34 / DET19 (both sections in parallel): Very promising... B+ 200V, VP ~180V, IP ~ 60mA, zero bias, 2.5k plate load - 1.4% THD @ 1W, 3.3W at 5% THD. I tried cathode bias and upping B+ but performance was worse.
3C24: Need higher impedance OPT and maybe higher voltage, but... at B+ = 400V, IP ~ 30mA, zero bias, 7k load - about 4% THD at 1W.
703A: Can't believe it can really live long at 20W dissipation on that tiny plate... B+ = 300V, IP ~ 70mA, zero bias, 7k load - 3.3% THD @ 1W, 5% @ 2.2W
6360 (triode connected, both sides in parallel): Needed to add 500 ohms cathode bias... B+ = 300V, IP ~ 50mA - 1W THD 2.8%, 5% @ 2.8W
2C34 / RK34 / DET19 (both sections in parallel): Very promising... B+ 200V, VP ~180V, IP ~ 60mA, zero bias, 2.5k plate load - 1.4% THD @ 1W, 3.3W at 5% THD. I tried cathode bias and upping B+ but performance was worse.
3C24: Need higher impedance OPT and maybe higher voltage, but... at B+ = 400V, IP ~ 30mA, zero bias, 7k load - about 4% THD at 1W.
703A: Can't believe it can really live long at 20W dissipation on that tiny plate... B+ = 300V, IP ~ 70mA, zero bias, 7k load - 3.3% THD @ 1W, 5% @ 2.2W
6360 (triode connected, both sides in parallel): Needed to add 500 ohms cathode bias... B+ = 300V, IP ~ 50mA - 1W THD 2.8%, 5% @ 2.8W
Attachments
Hello Pete.[IM[/IMG]
I always wanted to try using a power opamp as a class A2 driver. I finally got around to some experiments. I managed to have some sucess, especially using a 1636 (ala Darling amp) to get ~1.2W out. Still experimenting - I need to try push-pull, and some different tubes. Here's the info:
Class-A2 and a power opamp driver
Let me know if you can think of any candidate tubes. The power opamp can only put out about +/-36V @ 50mA, so stuff like 811's are not workable. So far I've tried the 1636, 2E26 (OK but not a lot of advantage over A1), and 6V6/6L6/6W6 (worse than A1). Seems like some of the newer semi-useless 9-pin-mini RF tubes might do well here?
Pete
If you want Flea power in a cool old bottle try a 6G6G.
DT
All just for Fun!
Last edited:
Tried push-pull-parallel 2C34/RK34's. One opamp set up inverting and the other non-inverting, with gains of 24. Zero bias, 175V on the plates for about 10W dissipation per tube... 8k plate-to-plate (it's what I have sitting around).
Without making any effort to balance or match tubes, 0.5% THD @ 1 watt and about 7 watts at clipping (output tubes, not opamps were clipping at this level).
I need another power amp like I need a hole in the head, but I think I will have to build a real amp using this and a 1641/RK-60 rectifier (another 2-cap tube). Maybe I should call it "Mickey"...
Using NOS plate caps and sockets, the caps and sockets will cost double what the entire tube compliment costs. RK-34 / 2C34's are going for $4.00 each...
BTW, data is here http://www.mif.pg.gda.pl/homepages/frank/sheets/081/r/RK34.pdf and here http://www.mif.pg.gda.pl/homepages/frank/sheets/139/d/DET19.pdf
Pete
Without making any effort to balance or match tubes, 0.5% THD @ 1 watt and about 7 watts at clipping (output tubes, not opamps were clipping at this level).
I need another power amp like I need a hole in the head, but I think I will have to build a real amp using this and a 1641/RK-60 rectifier (another 2-cap tube). Maybe I should call it "Mickey"...
Using NOS plate caps and sockets, the caps and sockets will cost double what the entire tube compliment costs. RK-34 / 2C34's are going for $4.00 each...
BTW, data is here http://www.mif.pg.gda.pl/homepages/frank/sheets/081/r/RK34.pdf and here http://www.mif.pg.gda.pl/homepages/frank/sheets/139/d/DET19.pdf
Pete
Attachments
I will have to build a real amp using this and a 1641/RK-60 rectifier (another 2-cap tube). Maybe I should call it "Mickey"...
Sounds great - I've got about 35 in my stash waiting for some action !
Cool fun project!
I've been picking up various types of tubes suitable for this little board over the years. Always hope for a way to play with them. Looks like I got the last 2 boards!
Michael- how would you apply feedback for the 3c24?
-Kent
I've been picking up various types of tubes suitable for this little board over the years. Always hope for a way to play with them. Looks like I got the last 2 boards!
Michael- how would you apply feedback for the 3c24?
-Kent
- Status
- Not open for further replies.
- Home
- Amplifiers
- Tubes / Valves
- Flea-power class-A2 using a power opamp driver