I found these power transformers about a month ago while trolling Ebay. I bought 6 of them. They are NIB and weigh in at 10 pounds each. The mailman was not happy. The seller calls them:
NOS POWER TRANSFORMER 120 + Watt, Tube Amp #7019
Item number: 220628640284
There are 3 6.3 volt windings, a 44 volt bias winding, and a 235 HV winding. It was originally intended to operate with a voltage doubler, but I have found another solution. Use 2 transformers with a bridge rectifier on each HV winding to make two 300 volt supplies. Wire them in series, so you have 0-300-600 volt DC supplies. Just right for high powered red boards!
NOS POWER TRANSFORMER 120 + Watt, Tube Amp #7019
Item number: 220628640284
There are 3 6.3 volt windings, a 44 volt bias winding, and a 235 HV winding. It was originally intended to operate with a voltage doubler, but I have found another solution. Use 2 transformers with a bridge rectifier on each HV winding to make two 300 volt supplies. Wire them in series, so you have 0-300-600 volt DC supplies. Just right for high powered red boards!
It looks like the feeding frenzy is over for now. They may return later. I bought 3 about 5 weeks ago. I played with them and contacted the seller for 3 more. He said he had a few more. I explained that a forum post would sell a few more, but he didn't seem too interested. He said he would relist them again when he had time.
They are cloth over some sort of plastic or rubber material. These seem to be of 1960's vintage, but still new in the box. They even smell like an old transformer. Not sure about asbestos but I would wash your hands after handling. That should be the case when playing with lead solder since trace amounts of lead can be absorbed through the skin. Asbestos is only a problem if inhaled.
I would use an Antek toroid. Probably a 2T230. It has 2 230 volt secondaries and 2 6.3 volt secondaries.
thanks Tubelab for the Antek model # - do you figure it'd run hot or just warm with 8-EL84 + 2-12ax7 class A load? - that's their spec and assume is met @ 50Hz - outputs will be Hammond 1K2 as already bought years ago. Does it make more sense sonically to pick a higher voltage then go with choke input and tube rectifier or will the solid state rectification sometimes sound "better"?? - only SE I built was parallel 2a3 with 5842q with solid state diodes on the output and tube rectifier on the 5842q
Last edited:
Regarding the 10 pounder: Do you think those extra 6.8V windings
could handle dancing to the tune of push pull cathode followers?
Roughly along these lines? I'm not yet sayin its practical to drive.
Just asking if heaters to cathodes is manageable problem or no?
No center tap on two out of three. Resistors???
could handle dancing to the tune of push pull cathode followers?
Roughly along these lines? I'm not yet sayin its practical to drive.
Just asking if heaters to cathodes is manageable problem or no?
No center tap on two out of three. Resistors???
Attachments
Last edited:
Only testing will tell. I built a P-P cathode follower. It works and has some serious bass. I used an Antek toroid for heater power and simply tied one end of each winding (one heater pin) to the cathode.
Yes it was drive voltage limited, but this is a case where you can run a chip amp into a backwards OPT (resistor load about 15K) and get zillions of volts since no drive current is needed.
You can use a technique to keep the voltage across the tube constant to really reduce the distortion. See simulation. You can also build a full up "augmented cathode follower" which can be configured to have gain. A voltage gain of about two was about the max I could get before instability comes up.
http://www.diyaudio.com/forums/tube...wer.html?highlight=augmented+cathode+follower
One thing about that design, bugs me...
If I'm gonna have to drive the full Miller of those MOSFETs,
why am I bothering with 6KG6 underneath? Seems to me
that you just dumbed two perfectly good pentodes down
to diodes for nothing but maybe crossover smoothage...
Could do same with a valve rectifier. Or square law diode
strapped MOSFET, or square law Schottky diode at much
lower voltage. Does this really ply 6KG6 to advantage?
Source follower seems to be doing all the actual work...
Maybe its just a philosophical viewpoint... I couldn't do it.
But if your heater to cathode problem was manageable,
that much at least proved something useful.
If I'm gonna have to drive the full Miller of those MOSFETs,
why am I bothering with 6KG6 underneath? Seems to me
that you just dumbed two perfectly good pentodes down
to diodes for nothing but maybe crossover smoothage...
Could do same with a valve rectifier. Or square law diode
strapped MOSFET, or square law Schottky diode at much
lower voltage. Does this really ply 6KG6 to advantage?
Source follower seems to be doing all the actual work...
Maybe its just a philosophical viewpoint... I couldn't do it.
But if your heater to cathode problem was manageable,
that much at least proved something useful.
Last edited:
Same thing, minus a few "critical" parts. Whats the diff?
Just beating the dead horse for sake of absurdity...
Both sims exceed rating of that 600VDS MOSFET by an
additional 750V on the downswings, either way...
Not to mention gates goin' splodie...
-----------------
Back on the more serious note:
You tie one end of heater direct to cathode, or midpoint
two resistor span to cath? If so, bout how many Ohms?
Or with indirectly heated type cathode, it won't matter?
Just beating the dead horse for sake of absurdity...
Both sims exceed rating of that 600VDS MOSFET by an
additional 750V on the downswings, either way...
Not to mention gates goin' splodie...
-----------------
Back on the more serious note:
You tie one end of heater direct to cathode, or midpoint
two resistor span to cath? If so, bout how many Ohms?
Or with indirectly heated type cathode, it won't matter?
Last edited:
Almost. Its certainly dead enough, but still has momentum...
I've moved the augie gate drives to the cathodes, bringing
the tubes back into the control seat. Bout 66 volts above
cathodes, bias currents begin to look similar to the original.
We aren't driving MOSFET miller anymore, but G2 currents
are big enough, this was drop in the bucket anyways...
But I notice something else odd about this sim? Applies to
the original as well. Even with 1000ohms stopper, G1 rises
+30 volt above the cathode, yet the sim only reports 9mA.
Is that figure realistic? It seems rather low...
Horse might be dead, but it's still shaking around a bit, so let me smack it at least once more.
I have several sims on my computer that all show that this is a viable idea. Yes I got carried away with the supply and drive voltages on some of them. It is too easy to build a killowatt plus amplifier in the simulator. The smoke doesn't smell bad, and the parts are free. The sound sucks though.
I built real working hardware about three years ago. I don't have the schematic here at work but the output tubes were 6336's (triodes) and the top device does indeed take its input from the cathode of the bottom device. I have used mosfets, triodes, pentodes, and agile SMPS's for the top device.
These were all experiments leading up to the Minitron amplifier design where the augie mosfet was replaced by a Microchip dsPIC based agile SMPS. The SMPS is modulated by the audio signal such that the plate to cathode voltage of the output tube remains constant, but the total heat dissipation is greatly reduced.
Well you could look at the design as a mosfet source follower with a tube in series with its output, or you could see it as a cathode follower with a noisy plate supply. The cathode follower has a high PSRR, so I believe it is the controlling device here. Keeping the plate voltage constant across the output tube removes its voltage variant parameter induced distortions from the audio signal.
MacDonald envisioned the augmented cathode follower circuit as a nearly perfect voltage amplifier and his applications were at low power levels. I built a few complete power amps with his circuits and they do sound pretty nice. All were SE designs, and I could not get any of them to simulate right. I still have one working amp. It's circuit doesn't match any of my sims though.
I have several sims on my computer that all show that this is a viable idea. Yes I got carried away with the supply and drive voltages on some of them. It is too easy to build a killowatt plus amplifier in the simulator. The smoke doesn't smell bad, and the parts are free. The sound sucks though.
I built real working hardware about three years ago. I don't have the schematic here at work but the output tubes were 6336's (triodes) and the top device does indeed take its input from the cathode of the bottom device. I have used mosfets, triodes, pentodes, and agile SMPS's for the top device.
These were all experiments leading up to the Minitron amplifier design where the augie mosfet was replaced by a Microchip dsPIC based agile SMPS. The SMPS is modulated by the audio signal such that the plate to cathode voltage of the output tube remains constant, but the total heat dissipation is greatly reduced.
Well you could look at the design as a mosfet source follower with a tube in series with its output, or you could see it as a cathode follower with a noisy plate supply. The cathode follower has a high PSRR, so I believe it is the controlling device here. Keeping the plate voltage constant across the output tube removes its voltage variant parameter induced distortions from the audio signal.
MacDonald envisioned the augmented cathode follower circuit as a nearly perfect voltage amplifier and his applications were at low power levels. I built a few complete power amps with his circuits and they do sound pretty nice. All were SE designs, and I could not get any of them to simulate right. I still have one working amp. It's circuit doesn't match any of my sims though.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.