At the request of those who are helping me with this issue, and for the benefit of all members of the community who may someday experience a similar problem, I am going to gather the details of this "case" in this new thread...
The circuit in question is the TSE-II SE amplifier with 5842 input tubes and 300B power tubes.
The schematic is attached to this post.
The power transformer in this case is:
https://edcorusa.com/products/xpwr178-660v330-0-330-200ma-6-3v3-15-0-3-15-6a-5v2-5-0-2-5-3a
I measured DCR as follows:
I did use this choke in place of R4:
https://edcorusa.com/products/cxc125-10h-200ma-10h-200ma-choke
Its DCR is 75 ohm (vs. the 150 ohm R4).
C4 is a quality 47uF 450V electrolytic capacitor.
(A 5AR4 should be able to handle a 60uF input capacitor, according to the datasheet.)
C5 is a quality 150uF 450V electrolytic capacitor.
I have the following motor run capacitor in parallel with C5:
https://www.digikey.com/en/products/detail/kemet/C870CG36100AA0J/6556371
(100 uF Film Capacitor 470V Polypropylene)
The circuit in question is the TSE-II SE amplifier with 5842 input tubes and 300B power tubes.
The schematic is attached to this post.
The power transformer in this case is:
https://edcorusa.com/products/xpwr178-660v330-0-330-200ma-6-3v3-15-0-3-15-6a-5v2-5-0-2-5-3a
I measured DCR as follows:
- Primary winding: 1.75 ohm
- HV secondary: 35.3 ohm and 36.8 ohm
I did use this choke in place of R4:
https://edcorusa.com/products/cxc125-10h-200ma-10h-200ma-choke
Its DCR is 75 ohm (vs. the 150 ohm R4).
C4 is a quality 47uF 450V electrolytic capacitor.
(A 5AR4 should be able to handle a 60uF input capacitor, according to the datasheet.)
C5 is a quality 150uF 450V electrolytic capacitor.
I have the following motor run capacitor in parallel with C5:
https://www.digikey.com/en/products/detail/kemet/C870CG36100AA0J/6556371
(100 uF Film Capacitor 470V Polypropylene)
The story continues...
I have had problems with 3 rectifier vacuum tubes: Sovtek 5AR4, a Linlai 5U4G and most recently a Psvane 5AR4. All new issue units. The Sovtek flash during the first few hours. The Linlai I don't recall. The Psvane lasted 4-5 months.
@Francois G first helped me with this post:
And then this post:
@6A3sUMMER gave me this advice:
I have had problems with 3 rectifier vacuum tubes: Sovtek 5AR4, a Linlai 5U4G and most recently a Psvane 5AR4. All new issue units. The Sovtek flash during the first few hours. The Linlai I don't recall. The Psvane lasted 4-5 months.
@Francois G first helped me with this post:
And then this post:
@6A3sUMMER gave me this advice:
The story continues...
I put the amplifier on the bench and I am trying to figure out the best course of action to keep rectifier tubes alive longer. 😉
The input capacitor, at 47 uF, should be OK, since a 5AR4 is meant to work with up to 60 uF.
The 5AR4 datasheet gives:
So, I am guessing 330V RMS at either plates should not be a problem.
What is clear, however, is that I should have at least 90 ohm of limiting resistor at each plate, and I do not, since the DCR of my HV winding is ~35 ohm and the reflected impedance of the primary would not fill the gap between 35 ohm and 90 ohm.
I put the amplifier on the bench and I am trying to figure out the best course of action to keep rectifier tubes alive longer. 😉
The input capacitor, at 47 uF, should be OK, since a 5AR4 is meant to work with up to 60 uF.
The 5AR4 datasheet gives:
So, I am guessing 330V RMS at either plates should not be a problem.
What is clear, however, is that I should have at least 90 ohm of limiting resistor at each plate, and I do not, since the DCR of my HV winding is ~35 ohm and the reflected impedance of the primary would not fill the gap between 35 ohm and 90 ohm.
The story continues...
Finally, here comes the part where "solutions" are considered...
I see 5 possibilities:
1) Adding power resistors in series between each tap of the HV winding and the rectifier plates. 55 ohm should do. It would be wasted energy, but it would keep the 5AR4 happy, if the datasheet is to be believed. It would also drop some B+, which I may not want.
2) Adding a CL-90 thermistor on the primary winding, and that would probably let inrush currents pass without tickling the rectifier the wrong way. @6A3sUMMER explained he is not a fan of that "solution".
3) Adding a CL-140 thermistor in series with the HV winding CT. I don't know anything about that solution.
4) Adding silicon diodes in series with the plates.
5) Replacing the tube rectifier with a device made of 2 silicon diodes and a thermistor. (Weber WS1T) @Francois G is not a fan of that solution, as it eliminates the nice slow start provided by the 5AR4 rectifier.
If it did the job, I would lean towards "2)" as it seems to be the simplest "brew" to drink.
Finally, here comes the part where "solutions" are considered...
I see 5 possibilities:
1) Adding power resistors in series between each tap of the HV winding and the rectifier plates. 55 ohm should do. It would be wasted energy, but it would keep the 5AR4 happy, if the datasheet is to be believed. It would also drop some B+, which I may not want.
2) Adding a CL-90 thermistor on the primary winding, and that would probably let inrush currents pass without tickling the rectifier the wrong way. @6A3sUMMER explained he is not a fan of that "solution".
3) Adding a CL-140 thermistor in series with the HV winding CT. I don't know anything about that solution.
4) Adding silicon diodes in series with the plates.
5) Replacing the tube rectifier with a device made of 2 silicon diodes and a thermistor. (Weber WS1T) @Francois G is not a fan of that solution, as it eliminates the nice slow start provided by the 5AR4 rectifier.
If it did the job, I would lean towards "2)" as it seems to be the simplest "brew" to drink.
Ouf!
I am sorry for the long-winded monologue.
I have not decided which way I will go yet.
If anyone else has dealt with a similar issue before, please reply. I am sure there is more I could learn about this before I sacrifice another 5AR4. 🙂
I am sorry for the long-winded monologue.
I have not decided which way I will go yet.
If anyone else has dealt with a similar issue before, please reply. I am sure there is more I could learn about this before I sacrifice another 5AR4. 🙂
since i design and build all transformers used in my builds i m happy to use FWB or full wave doublers instead...more information here: https://www.valvewizard.co.uk/bridge.html
by using the hybrid type, arc overs can be minimized if not altogether avoided sins the tribe rectifier will never see the kind of ac inputs that the full-wave center tapped arrangements have, tube stress is mitigated..
by using the hybrid type, arc overs can be minimized if not altogether avoided sins the tribe rectifier will never see the kind of ac inputs that the full-wave center tapped arrangements have, tube stress is mitigated..
Attachments
This one is my favorite now, as it would not impact other secondary windings. 🥸
The CL-90 on the primary winding would impact all secondaries.
Now, there’s the issue that 330V exceeds the voltage rating of a CL-140. Although, the datasheet doesn’t specify a max voltage rating.
https://www.mouser.com/datasheet/2/...mometrics_NTC_Inrush_031814_web_p-1315885.pdf
Last edited:
I agree. I can solder one end to the transformer wire and shove the other into the PCB terminal block. Oops, that sounds bad. I would put a piece of wire on the other side too. 🤣
Purists will hunt me down and say I am diluting the magic properties of tube rectification.
But I’m sure I can live with that.
If the loaded primary is 330VAC, then the peak voltage is 1.4 x 330V = 462V peak.
Use 500V caps there; if the output tubes do not load the power supply, then the 5AR4 rectifier voltage drop is far less, and the secondary puts out more than 330VAC.
Tube rectification?
I gave that up for solid state rectifiers that connect to Choke Input filters.
One power supply uses HEXFREDs, a little expensive, but no more expensive than a very good tube rectifier, and no 5V winding needed.
Your choice may vary
Use 500V caps there; if the output tubes do not load the power supply, then the 5AR4 rectifier voltage drop is far less, and the secondary puts out more than 330VAC.
Tube rectification?
I gave that up for solid state rectifiers that connect to Choke Input filters.
One power supply uses HEXFREDs, a little expensive, but no more expensive than a very good tube rectifier, and no 5V winding needed.
Your choice may vary
Last edited:
what could be simpler than a single secondary winding to feed 5ar4 rectifier plates and add to 1n4007's to complete?
with FWCT psu, half the time or every 180 degrees of the electrical cycle, that half of the winding is not delivering currents, waste of electrical cycles....
transformer utilization is greater by about 30% more so lesser heating of the cores...
purist do what purists do, miss out on the good things...
Last edited:
i choose psu ecaps on the basis of all but the rectifier tube plugged in, what is the resulting B+? will it exceed the wvdc of the ecap?
a rule of thumb I observe, 80% ecap's wvdc, working volts dc spec for safety...
B+ will not exceed the 450V rating of the ecaps, in fact it won’t exceed 400V.
But I think @6A3sUMMER is trying to cover the brief power-on period where things could get close.
it is written down on the tube datasheet all along, just follow it and you are home free.....