tubelab.com said:
I thought the 6BK4 might have some usefulness as an output tube used to drive some sort of electrostatic speaker system where high anode voltage up around 10 kV might be employed. The tube can handle up to about 50 kV IIRC.
What plate voltage did you try it at for audio, as an output or linestage/preamp?
I just connected it up to my power supplies and turned the knobs trying to get it to conduct. At 550 V on the plate and +25 on the grid (limit of my power supplies) I was drawing a few microamps. The behavior was very nonlinear. The thought was for preamp applications, since max plate current is 2 mA by some books and 1.6 mA in others. It is designed to operate with 26 KV on the plate. The maximum ( non conducting ) voltage is 46 KV in my old RCA book. No max ratings are given in my newer RCA book.
I did not consider HV applications for electrostatic speakers, since I have no room for them.
I did not consider HV applications for electrostatic speakers, since I have no room for them.
The gain and transconductance figures for the 6HV5A are for high plate voltage and high peak current, well outside the range where one could operate the tube continuously without melting it into a pile of glittering slag. The mu should be about the same at lower current, but I would expect a much lower transconductance than the figure of 30k mumhos cited in the data sheet. At 1.2kV plate voltage, 35W dissipation is only 29 mA.
Compactron idea #2
How about a push-pull set of 33GY7A? Reasonable dissipation and your power supply diodes built in. Build 'em in stereo and hang the four heaters in series from a isolation transformer and the dampers as a doubler for + 300 or so B voltage.
Not enough sleep and too much coffee makes me get strange thoughts
my2c
Jorge
How about a push-pull set of 33GY7A? Reasonable dissipation and your power supply diodes built in. Build 'em in stereo and hang the four heaters in series from a isolation transformer and the dampers as a doubler for + 300 or so B voltage.
Not enough sleep and too much coffee makes me get strange thoughts
my2c
Jorge
I found soma tubes in the warehouse that look similar to the 6HV5. The 6JH5 and the 6JK5 look like the 6HV5 and have the same specs. The 6HS5 has a wimpier plate and is only rated for 30 watts.
I tried connecting the 6JK5 up to my power supply and twisting a few knobs to see what kind of voltages this tube would need to work. With 400 volts on the plate and ZERO volts on the grid (beam plates connected to cathode) the tube was drawing about 15 mA. If this tube was going to work on a reasonable plate voltage, positive grid voltage would be needed.
OK, I turned the grid voltage up, with +5 volts on the grid, the tube was drawing 200 mA. I didn't have time to plot curves to determine linearity. From this 5 minute test, it seems that this tube could operate from a 400 to 500 volt supply with positive bias. Gain is very high.
Just to see what would happen, I connected the beam plates up to the plate. Zero bias current was 50 mA and oscillation was present. Any knob twisting caused extreme TV jamming oscillation and 200 mA of current.
Some time in the future, I will plot some curves, but I don't think that it will be any time soon.
I tried connecting the 6JK5 up to my power supply and twisting a few knobs to see what kind of voltages this tube would need to work. With 400 volts on the plate and ZERO volts on the grid (beam plates connected to cathode) the tube was drawing about 15 mA. If this tube was going to work on a reasonable plate voltage, positive grid voltage would be needed.
OK, I turned the grid voltage up, with +5 volts on the grid, the tube was drawing 200 mA. I didn't have time to plot curves to determine linearity. From this 5 minute test, it seems that this tube could operate from a 400 to 500 volt supply with positive bias. Gain is very high.
Just to see what would happen, I connected the beam plates up to the plate. Zero bias current was 50 mA and oscillation was present. Any knob twisting caused extreme TV jamming oscillation and 200 mA of current.
Some time in the future, I will plot some curves, but I don't think that it will be any time soon.
When I finally get around to building the 6HV5A amp, I'll be using plate and grid stoppers, and some cathode resistance on each tube - otherwise, I'd expect them to burst into song, especially operating in parallel. The beam plates will be nailed to the cathode right at the socket. As far as I've seen, the only 6HV5As worth messing with are from GE and Westinghouse - the plates on these are 20-30% larger than other brands. GE boasts of a superior bonded cathode coating that reduces the incidence of shorts. Given the high gain, the cathode to grid spacing must be microscopic...
I don't have a dead one to dissect, YET. I can see inside that the grid is made of a lot of closely spaced turns of really fine wire.
The 6HV5's that I have are both GE's. The other tubes are various brands. I was torturing a Zenith 6JK5 last night. Construction looks identical to the GE 6HV5. I had to go up to 55 watts to get the plate to glow, and even at that level you had to turn the room lights off to see it.
All of the tubes that I have look identical, except for the 6HS5. It has much smaller plates, and it is a GE.
I am trying to dream up a minimum parts amplifier, but it hasn't come to me yet. I don't have time to build it yet anyway. I was digging through my warehouse looking for some small dual triodes when I found a big box marked sweep tubes. I couldn't remember what I put in the box, so I opened it and pulled out about 5 of these pulse regulator tubes. At the time I boxed up the tubes I did not know about the 6HV5, so I thought it was just annother sweep tube.
About a year ago I found that Jack at Electra-Print had used this the 6HV5 as a driver tube. 700 volts, cathode bias (negative grid voltage).
http://www.electra-print.com/hyperdrive.html
The 6HV5's that I have are both GE's. The other tubes are various brands. I was torturing a Zenith 6JK5 last night. Construction looks identical to the GE 6HV5. I had to go up to 55 watts to get the plate to glow, and even at that level you had to turn the room lights off to see it.
All of the tubes that I have look identical, except for the 6HS5. It has much smaller plates, and it is a GE.
I am trying to dream up a minimum parts amplifier, but it hasn't come to me yet. I don't have time to build it yet anyway. I was digging through my warehouse looking for some small dual triodes when I found a big box marked sweep tubes. I couldn't remember what I put in the box, so I opened it and pulled out about 5 of these pulse regulator tubes. At the time I boxed up the tubes I did not know about the 6HV5, so I thought it was just annother sweep tube.
About a year ago I found that Jack at Electra-Print had used this the 6HV5 as a driver tube. 700 volts, cathode bias (negative grid voltage).
http://www.electra-print.com/hyperdrive.html
Tubelab -
I looked up the 6JH5, and it takes a bit more filament current than the 6HV5A - 2.5A vs. 1.6A. I would suspect this means a somewhat larger cathode structure and more peak current capability. At 700V plate voltage, though, the difference may not mean too much. The 6JH5 may be a better choice if one is trying to squeeze out power using a lower plate voltage. I'm gratified but not entirely surprised by the results of your power torture test - those tubes have big, big plates.
I'm planning to take an old Hallordson 540VCT transformer and run it with a full wave bridge to get around 700V to run the 6HV5As. What's holding me up is figuring out a sneaky, sleazy way of getting a lower supply voltage for the driver tubes without having to throw in another transformer or blow big power in dropping resistors. If I have to use another transformer, one sneaky way of proceeding might be to dedicate a spare filament winding to driving a filament transformer backwards. If I use a transformer with dual/230V primary winding, I could get 300VDC or so to run the drivers.
I looked up the 6JH5, and it takes a bit more filament current than the 6HV5A - 2.5A vs. 1.6A. I would suspect this means a somewhat larger cathode structure and more peak current capability. At 700V plate voltage, though, the difference may not mean too much. The 6JH5 may be a better choice if one is trying to squeeze out power using a lower plate voltage. I'm gratified but not entirely surprised by the results of your power torture test - those tubes have big, big plates.
I'm planning to take an old Hallordson 540VCT transformer and run it with a full wave bridge to get around 700V to run the 6HV5As. What's holding me up is figuring out a sneaky, sleazy way of getting a lower supply voltage for the driver tubes without having to throw in another transformer or blow big power in dropping resistors. If I have to use another transformer, one sneaky way of proceeding might be to dedicate a spare filament winding to driving a filament transformer backwards. If I use a transformer with dual/230V primary winding, I could get 300VDC or so to run the drivers.
I was about to say the same thing. Put a bridge across the 540 vct winding and a cap across the bridge. Ground the negative terminal of the bridge. You will have about 700 volts at the positive terminal of the bridge. Connect a cap from the CT of the transformer to ground. Electrolytic is OK. You will magically have about 350 volts at this connection.
If you think about this for a while it is really two seperate full wave power supplies stacked on top of each other. Or it is like the + and - supplies in most SS amps, with the ground in a different place.
Doing this (or any bridge rectifier on a transformer designed for full wave CT operation) will subject the transformer to twice the voltage that it was designed for. Use caution with vintage transformers. I am thinking Antek toroid from Ebay.
Same for the OPT. With 700 volts on the B+, the peak voltage at the plate could go to 1400 volts. I don't think that Hammond had this in mind when they made the 125CSE. Make sure that the cases of these transformers are well grounded, even during testing.
I was thinking of using the 5K Transcendars I have on hand for this amp. I'll have to call their shop and find out if the things can take the voltage. I was also thinking of the Antek transformer, but that's another $50 for a speculative project, and I have all this NOS power iron that otherwise would be fancy paperweights... Too bad - I have the other two Antek tube power transformers, but not the 540V one.
BTW, the idea of using the CT is a nifty one - thanks.
BTW, the idea of using the CT is a nifty one - thanks.
I want to use the 125CSEfor the initial experiments to determine the optimum load impedance. Then I will figure out what OPT to use.
I also have the other two Antek transformers, but not the 550 volt one, so I will see if the 400 volt (dual 200 V) transformer will produce enough juice to power up a pair of these tubes. I also have about 1000 pounds of assorted power transformers to try, so something will work. It just takes time to try them all. The big Antek (720 volt) will generate about 1KV with a bridge. This would surely toast the Hammond 125CSE. Besides, I have already put it into one of my Simple SE amps, but it is total overkill for that job.
I was hopeing to come up with a power supply with just the right amount of voltage so that I could operate the tubes with zero bias. The voltage required (I'm guessing 500 to 600 volts) would be easier on the OPT. I have already used both the Hammond and the Edcor at 520 volts without fireworks.
I'm not too worried about the bias, being reconciled to a fair amount of cathode resistor to provide local feedback to linearize the tube and keep it from taking off like a rocket. One could conceivably run a single tube amp. The feedback wouldn't be as pretty as your classic two-tube cathode feedback number, but it could be forced to work...
I haven't been able to get in touch with Transcendar yet, but the description of their 5k SE transformers on Ebay indicates that they are hipot tested to 3kV - that's a good start. The only thing I need to find out now is whether they will support a +/- 700V swing on the primary. I have two electrophoresis supplies now - one goes to 500V, the other to 2kV. I should be able to use these to bench test a breadboard.
I bought a couple of these from Ebay, but neither one of them works. I have a Fluke 407D which goes to 555 volts. For higher voltage, I usually rig up a variac to a BIG 2000 volt transformer and a rectifier.
I have several 3K ohm Transcendars. I have operated them with a B+ of 520 volts without issue, but I have not tried 700 volts yet.
The 500V electrophoresis supply I have is from Bio-Rad, and it appears to power up and work, though I haven't tried to source any output current from it yet. It's a neat and reasonably compact package. The 2000V beast is a Dan-Kar, and I suspect that it is a rude and crude setup using a rectified microwave oven transformer and a variac, though I may be wrong. I haven't hauled that one down to the basement yet, but its time will come.
If you're set on using around 500V B+, your next step might be to set up a tube on a breadboard and plot a set of curves to see what happens at that voltage with some positive drive. I suspect you won't need too much, so your grid current may be miniscule. I'll end up doing the same thing at 700V. I still have the setup on the bench that I used to characterize the 6CD6GAs for enhanced triode operation, though the Dan-Kar supply will make things simpler and safer if it actually works.
Yes that is the next step. I am more concerned with linearity than whether there is grid current or how much. I am not afraid of using some sand to feed the grid current. Been there, done that, it works for me.
If the tube turns out to be non linear at 500 to 550 volts, I will go higher. I was looking to stay in the range of cheap transformers and capacitors. As the voltage goes up the price goes up.
I just received a huge 8H, 250 ma potted choke (Hallordson NOS). With the monster Antec 720V toroid, this should get me around 650 volts and gobs of current., enough to light off two pairs of 6HV5As. As Nelson Pass would say,
BUA, HA, HA, HA, HA, HA.....
I sure hope the 6HV5As deliver the goods after all this. If they don't cut the mustard, any suggestions for another tube that would benefit from this sort of B+ with similar plate dissipation (triode preferred)?
BUA, HA, HA, HA, HA, HA.....
I sure hope the 6HV5As deliver the goods after all this. If they don't cut the mustard, any suggestions for another tube that would benefit from this sort of B+ with similar plate dissipation (triode preferred)?
The first and most similar thing that comes to mind is the 811A. This is a DHT transmitting triode with a Mu of 160. In most circuits the grid wants to be in the same range as the 6HV5. Dissipation is 45 watts and it likes a lot of voltage. It fits the same socket as a 300B, which are common. It has a plate cap, which is considered a drawback by many. Many people have made good sounding amps with these. Grid current (A2) is required for most common (less than 1KV) operating points.
Svetlana once made the SV572-160, but it has been discontinued and is now hard to find. There is a High Mu variant of the 211, the 838, It is also hard to find. I would also consider a triode wired 813 at this voltage range. It would live forever considering the 100 watt dissipation ratings. They look cool and put out a lot of light.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.