Im sorry, bit of a rough day i shouldn't have responded as i did. Nitpicking because people get the thermeology wrong.. Youre not yourself when your hungry.. Just had an entire loaf of bread.
Ive been down that road before: my advice. instead of building monoblocks. Build two chassis one for the amplifier one for the power supply. And just push DC through the umbillical cord.
Only join all the supplies for your amplifier in the amplifier chassis at a central star ground and you're all set.
You can cut down on the losses somewhat if you use the LT4320. If you use reasonably good mosfets that have relatively fast switching times. A standard 25A bridge has horrible switching characteristics.
I use RURG6030 diodes for these kind of things nowadays, but i have a dozen 4320's i still have to use. im quite adept at rolling my own PCB's
I actually sold some regulators to someone that wanted to upgrade an existing amplifier, and using 4320 we actually got the headroom needed to run from a normal 10VAC winding for a 813 Push-pull. He was lucky enough that his transformer winder built in a lot of margin and it ended up fine.
For the HT, im quite the oddball, i have some very high performance tube regulator circuits to my name.
Ive been down that road before: my advice. instead of building monoblocks. Build two chassis one for the amplifier one for the power supply. And just push DC through the umbillical cord.
Only join all the supplies for your amplifier in the amplifier chassis at a central star ground and you're all set.
You can cut down on the losses somewhat if you use the LT4320. If you use reasonably good mosfets that have relatively fast switching times. A standard 25A bridge has horrible switching characteristics.
I use RURG6030 diodes for these kind of things nowadays, but i have a dozen 4320's i still have to use. im quite adept at rolling my own PCB's
I actually sold some regulators to someone that wanted to upgrade an existing amplifier, and using 4320 we actually got the headroom needed to run from a normal 10VAC winding for a 813 Push-pull. He was lucky enough that his transformer winder built in a lot of margin and it ended up fine.
For the HT, im quite the oddball, i have some very high performance tube regulator circuits to my name.
Agree with separate PSU chassis. Apart from heritage amps that I have, everything done from scratch has separate power supply.
And I love my tube rectifiers! I use 6C_3 series TV damper diodes. They have 350-400 mA continuous ratings, kilovolt voltages, and only about 10 V forward drop. Dont have to worry about inverse spikes killing SS diodes. If I need a 200 mA supply, I make it 400 mA continuous capable. Fun stuff.
And I love my tube rectifiers! I use 6C_3 series TV damper diodes. They have 350-400 mA continuous ratings, kilovolt voltages, and only about 10 V forward drop. Dont have to worry about inverse spikes killing SS diodes. If I need a 200 mA supply, I make it 400 mA continuous capable. Fun stuff.
Damper diodes, are really nice, however you really cant build a full bridge with them without the right transformers.
I did a board for a half bridge its in my gerbers thread here: V4lve lover's free gerbers thread.
I used three series connected BY2000 for each diode in the leg, with 2n7 2KV MKT over each diode. and 2.2 MEG for good measure.
One simple trick i found to keep dampers in there correct operating parameters is to use a 555 delay that shorts a resistor on the primary after some 30s, something like 50 ohms 25W wirewound. if you think about it: at a 1:10 step up that is reflected by the square of the turns ratio to the primary. This however only works if you have a separate HT transformer. You can also find a LM555 turn on delay in my thread about that.
Im gonna help a friend with his GM70 build, we found two identical tek transformers out of 545 scopes. Those puppies give 140VA + 2X 115VAC each, if you work it out you end up with six rectifiers and six capacitor for just the high voltage supply.
I did a board for a half bridge its in my gerbers thread here: V4lve lover's free gerbers thread.
I used three series connected BY2000 for each diode in the leg, with 2n7 2KV MKT over each diode. and 2.2 MEG for good measure.
One simple trick i found to keep dampers in there correct operating parameters is to use a 555 delay that shorts a resistor on the primary after some 30s, something like 50 ohms 25W wirewound. if you think about it: at a 1:10 step up that is reflected by the square of the turns ratio to the primary. This however only works if you have a separate HT transformer. You can also find a LM555 turn on delay in my thread about that.
Im gonna help a friend with his GM70 build, we found two identical tek transformers out of 545 scopes. Those puppies give 140VA + 2X 115VAC each, if you work it out you end up with six rectifiers and six capacitor for just the high voltage supply.
I've built two PP 300B amps using AC filaments. With the output tubes properly balanced, they were dead quiet on my ProAcs.
True that high voltage power transformers with center-tapped secondary are things of the past. It is either NOS, or expensive custom job. I have several such transformers that I acquired over the course of years. One is 1,800 vct from a Heathkit transmitter, quite a beast. Another, which I am going to use in my GM70 project, is a new production German thing, 600 VA 2,200 vct, with 6.3 V windings, perfect for a 6CM3 full wave rectifier.
I am sure there will be interest in hybrid rectifier boards with half-bridge of damper diodes. 6C_3 tubes may not be the best for this because they use novar or duodecar sockets that are available only as NOS. But Russian 6D19P use standard 9-pin socket (noval).
I am sure there will be interest in hybrid rectifier boards with half-bridge of damper diodes. 6C_3 tubes may not be the best for this because they use novar or duodecar sockets that are available only as NOS. But Russian 6D19P use standard 9-pin socket (noval).
Attachments
So yep you can cancel the 50Hz hum with a balance pot and get no hum on each plate. The PP also cancels the hum too so you may find multiple positions of both pots where there is no hum from the speaker. However you will still see 50Hz sidebands on the 1KHz tone until you get both in the right place. The 100Hz hum does not cancel on each plate but does on the PP. So you will still be left with 100Hz sidebands on the 1KHz signal.
The 300B is quite linear and they may be quite small in size, I don't know as you would need a distributed cathode model of the 300B to simulate. I did see some actual measurements somewhere.
Last edited:
I notice there's a link here
https://www.diyaudio.com/forums/att...ing-power-applied-directly-heated-triodes-doc
Which rectifies the 50Hz and produces 100Hz and adds it into the grid with a pot so you can null. I think you would need to remove the higher harmonics. Maybe provide a smaller voltage to the diodes with a higher load to get more of a square law characteristic.
https://www.diyaudio.com/forums/att...ing-power-applied-directly-heated-triodes-doc
Which rectifies the 50Hz and produces 100Hz and adds it into the grid with a pot so you can null. I think you would need to remove the higher harmonics. Maybe provide a smaller voltage to the diodes with a higher load to get more of a square law characteristic.
Re. sockets, you can plug a novar tube into magnoval socket, but contact will be loose. In Russian ceramic magnoval sockets (which I would highly recommend), contacts can be bent with a thin screwdriver a little bit to improve the grip, but I would rather use the correct novar socket. Btw, damper diode data sheets recommend removing the unused tabs from the socket to prevent arcover if high voltages are involved.
That last part is only applicable when the damper is one without top cap is my guess. EY500 has the cathode and heaters on the base and the anode at the top. And conventional wisdom is to tie the heaters to the cathode.
I know which sockets you mean, i think they are called SK509 by Svetlana St petersburg but could be mistaken.
I had a crazy idea to use a thyristor behind a high voltage diode bridge to chop up the high voltage and regulate it at the same time. This produces some harmonics, but im quite certain i can get these down to the single digit mV range at which point a 8K primary transformer is going to step down the noise down to unnoticable levels.
If you make an enable pin on the control electronics for that you can disable the high voltage alltogether before the tubes are warmed up. You can also pull this pin low in case of overcurrent in the outputs.
There are a couple really fun IC's if you need to do a floating current measurement for some reason. You can use a HCPL7800 insulation amplifier for current sense. You can feed this HCPL7800 with a 1W DC/DC
Another possibility is the HCNR200 this 2 dollar part and 50 cents of transistors allows you to send analog signals with great accuracy across a 1.5KV potential gap.
I know which sockets you mean, i think they are called SK509 by Svetlana St petersburg but could be mistaken.
I had a crazy idea to use a thyristor behind a high voltage diode bridge to chop up the high voltage and regulate it at the same time. This produces some harmonics, but im quite certain i can get these down to the single digit mV range at which point a 8K primary transformer is going to step down the noise down to unnoticable levels.
If you make an enable pin on the control electronics for that you can disable the high voltage alltogether before the tubes are warmed up. You can also pull this pin low in case of overcurrent in the outputs.
There are a couple really fun IC's if you need to do a floating current measurement for some reason. You can use a HCPL7800 insulation amplifier for current sense. You can feed this HCPL7800 with a 1W DC/DC
Another possibility is the HCNR200 this 2 dollar part and 50 cents of transistors allows you to send analog signals with great accuracy across a 1.5KV potential gap.
Thanks for the heads up on the insulation amplifiers - new to me, and could be very useful in the 1KV, thoriated tungsten filament world, where measuring cathode current might be harder (but safer!) than measuring anode current.
Damper diodes can be surprising when compared to conventional rectifiers. Their intended operation has their anodes to ground and heaters near ground. Cathodes swing way up into the several KiloJolt positive range, but can't go negative (because then they conduct).
Their rated peak cathode to heater voltages are really big when the cathode is positive of the heater, but only a coupla hundred Volts when negative. This allows some surprising things like full wave bridge rectifiers with only two, or possibly only one, transformer heater winding, for positive voltage power supplies.
YOS,
Chris
Damper diodes can be surprising when compared to conventional rectifiers. Their intended operation has their anodes to ground and heaters near ground. Cathodes swing way up into the several KiloJolt positive range, but can't go negative (because then they conduct).
Their rated peak cathode to heater voltages are really big when the cathode is positive of the heater, but only a coupla hundred Volts when negative. This allows some surprising things like full wave bridge rectifiers with only two, or possibly only one, transformer heater winding, for positive voltage power supplies.
YOS,
Chris
I dont mind solid state in my amps if you are switching the mains of the high voltage transformer anyway. i did a KISS relay board for that with a 555.
Chokes are to heavy for my tastes, and id rather use the heater winding for a tube regulator. So i can feed the output with a well regulated HT.
There are some insane sweeps suitable for that.
I have a weird schematic in my head for a tube regulator controller that drives a tube-transistor cascode . just use a voltage doubler on the heater, good reference like 1N823 and a matched BJT like the MAT01
Il get behind my laptop in a bit and see if i can find the article. Edit found it This is about a vacuum tube regulator that uses a transistor-tube cascode for a 0-2500V 10mA power supply. But the principle could also be used to regulate 1KV @150mA for most SE amps
PS: The 7800 is differential output, if you need a single ended signal, INA105 or something like that is better than discrete s in most cases.
Chokes are to heavy for my tastes, and id rather use the heater winding for a tube regulator. So i can feed the output with a well regulated HT.
There are some insane sweeps suitable for that.
I have a weird schematic in my head for a tube regulator controller that drives a tube-transistor cascode . just use a voltage doubler on the heater, good reference like 1N823 and a matched BJT like the MAT01
Il get behind my laptop in a bit and see if i can find the article. Edit found it This is about a vacuum tube regulator that uses a transistor-tube cascode for a 0-2500V 10mA power supply. But the principle could also be used to regulate 1KV @150mA for most SE amps
PS: The 7800 is differential output, if you need a single ended signal, INA105 or something like that is better than discrete s in most cases.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.