So, my tubelab SE is almost completed. I installed the Hasimoto OPT. I don't think I have gifted hears, but there is a big difference compared to the James.
I'll post a picture as soon as I get the covering plate (It has a funny cabinet
)
Going to business, I am using a separate power supply, feeding the amp with B+ , B- and AC 6.3V.
At the moment nothing is special in the unregulated PSU, just using good component.
I would like to try to put the Salas shunt regulator in the middle, and see what will happen. I wanted to do it with the 300B, but maybe I'll try first with the 45.
I drafted current and voltage requirements, and I would like to know if you think they are correct:
300B
400V (Could be a bit less)
190mA Idle 250 mA peak
45
275V
90mA idle
140 mA peak.
George, I remember a post where you were speaking about the benefit of regulated DC. Did you try this ?
Thanks,
Davide
I'll post a picture as soon as I get the covering plate (It has a funny cabinet
)Going to business, I am using a separate power supply, feeding the amp with B+ , B- and AC 6.3V.
At the moment nothing is special in the unregulated PSU, just using good component.
I would like to try to put the Salas shunt regulator in the middle, and see what will happen. I wanted to do it with the 300B, but maybe I'll try first with the 45.
I drafted current and voltage requirements, and I would like to know if you think they are correct:
300B
400V (Could be a bit less)
190mA Idle 250 mA peak
45
275V
90mA idle
140 mA peak.
George, I remember a post where you were speaking about the benefit of regulated DC. Did you try this ?
Thanks,
Davide
Davide,
I design and develop all of my amplifiers using lab type regulated power supplies, including the Tubelab SE. After the amplifier is working to my satisfaction, I try to design a power supply to result in similar performance. It is difficult to design a non regulated power supply with the same performance as a fully regulated power supply, so some compromises are usually involved.
In order to create a power supply for the Tubelab SE that works better than the on board uint some considerable complexity will be involved. In order to stabilize the output tubes, if the B+ is regulated the negative bias supply must be regulated too. If you are going this far, it would be good to provide another positive supply in the +50 to +150 volt range for the PowerDrive mosfets. Running the mosfet drains on a lower voltage than the B+ will allow the current through them to be increased without excessive heat generation. This supply doesn't need to be regulated, but should be clean.
You asked for B+ voltage recommendations, so I will offer mine, but you may have different preferences with different tubes and transformers. For 300B's I usually like 360 to 380 volts with about 75 to 80 mA through each tube. I usually use cheap tubes (Shuguang or Sovtek) and I have seen tendency toward runaway in older Sovteks when you get too close to 400 volts in some tubes. If I was designing the power supply (see below) I would make it adjustable from 350 to say 420 volts. Then you could find the sweet spot for your tubes.
On the 45 I tend to use 300 to 320 volts. Yes this is over spec, but I have never seen a tube that complains about it. I don't have any rare or expensive 45's though. I run the current at 25 to 30 mA per tube. Again, an adjustment range of 275 to 325 volts should work with most tubes.
I have not tried a shunt regulator. They tend to be a little more specific in their setup since the voltage is not usually adjustable. I have been working on a "universal regulated power supply" for use with vacuum tube amps. I will eventually make a board for one of these designs for use with the Tubelab SE and other amps. I have 3 different designs currently, 2 solid state and 1 tube. It's too soon to know when anything will be ready since I haven't had time to work on them.
I design and develop all of my amplifiers using lab type regulated power supplies, including the Tubelab SE. After the amplifier is working to my satisfaction, I try to design a power supply to result in similar performance. It is difficult to design a non regulated power supply with the same performance as a fully regulated power supply, so some compromises are usually involved.
In order to create a power supply for the Tubelab SE that works better than the on board uint some considerable complexity will be involved. In order to stabilize the output tubes, if the B+ is regulated the negative bias supply must be regulated too. If you are going this far, it would be good to provide another positive supply in the +50 to +150 volt range for the PowerDrive mosfets. Running the mosfet drains on a lower voltage than the B+ will allow the current through them to be increased without excessive heat generation. This supply doesn't need to be regulated, but should be clean.
You asked for B+ voltage recommendations, so I will offer mine, but you may have different preferences with different tubes and transformers. For 300B's I usually like 360 to 380 volts with about 75 to 80 mA through each tube. I usually use cheap tubes (Shuguang or Sovtek) and I have seen tendency toward runaway in older Sovteks when you get too close to 400 volts in some tubes. If I was designing the power supply (see below) I would make it adjustable from 350 to say 420 volts. Then you could find the sweet spot for your tubes.
On the 45 I tend to use 300 to 320 volts. Yes this is over spec, but I have never seen a tube that complains about it. I don't have any rare or expensive 45's though. I run the current at 25 to 30 mA per tube. Again, an adjustment range of 275 to 325 volts should work with most tubes.
I have not tried a shunt regulator. They tend to be a little more specific in their setup since the voltage is not usually adjustable. I have been working on a "universal regulated power supply" for use with vacuum tube amps. I will eventually make a board for one of these designs for use with the Tubelab SE and other amps. I have 3 different designs currently, 2 solid state and 1 tube. It's too soon to know when anything will be ready since I haven't had time to work on them.
George,
Thank you very much for your reply. The current you indicated are the idle currents. What should I consider as peak ? from my calculations with 75mA for each tube (300B) I get aroung 105 mA peak. The driver should stay at 17 mA as it has the CCS. So the total peak current should be 240mA. Am I correct ? something in the range of +30%.
For 45 (biased at 25ma) I should a total peak of 100mA.
By the way great amp, I am playing with it more to learn things that for a real improvement. I have a separate PSU as I do not want to spend lot of money on power transformer, as I will change country and voltage in few years. I asked to have a custom made transformer for the tubelab SE with double primary, and they asked me 270 $ !!!
D.
Thank you very much for your reply. The current you indicated are the idle currents. What should I consider as peak ? from my calculations with 75mA for each tube (300B) I get aroung 105 mA peak. The driver should stay at 17 mA as it has the CCS. So the total peak current should be 240mA. Am I correct ? something in the range of +30%.
For 45 (biased at 25ma) I should a total peak of 100mA.
By the way great amp, I am playing with it more to learn things that for a real improvement
. I have a separate PSU as I do not want to spend lot of money on power transformer, as I will change country and voltage in few years. I asked to have a custom made transformer for the tubelab SE with double primary, and they asked me 270 $ !!!D.
I made some experience with the salas regulator for another project, and I think I made up my mind on how to test them.
The dissipation of the shunt are very high for current over 100mA, so I will use one b+ regulator per channel, between the board and opt, with the 10 ohm resistor after the regulator. B- will be regulated with the Morgan Jones, mpsa92 based regulator. As George suggested a separate psu for the power drive will be provided.
The onboard psu will be used as a hv source for the regulators.
What I did not address yet is the b+ for the drivers. Is it i
important to have it regulated with the ccs? In case should i also separate one regulator per channel ?
I'll test first with the 45, then I'll raise the current.
Like this all the voltages will be adjustable.
BR,
Davide
The dissipation of the shunt are very high for current over 100mA, so I will use one b+ regulator per channel, between the board and opt, with the 10 ohm resistor after the regulator. B- will be regulated with the Morgan Jones, mpsa92 based regulator. As George suggested a separate psu for the power drive will be provided.
The onboard psu will be used as a hv source for the regulators.
What I did not address yet is the b+ for the drivers. Is it i
important to have it regulated with the ccs? In case should i also separate one regulator per channel ?
I'll test first with the 45, then I'll raise the current.
Like this all the voltages will be adjustable.
BR,
Davide
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.