Burnedfingers,
Rather embarassed that I only found time now to further on my post #6. In his post #16 Kmaier suggested the same thing I had in mind, but my component values seem to differ somewhat. We have 50 Hz mains compared to your 60 Hz, which would make a slight difference. But here goes:
Firstly I took the original ST70 bias circuit as basis, i.e. 10K from common to two 10K bias pots in parallel, followed by another 10K to a diode (1N4007, whatever). From either side of the second 10K two 47uF/63V electrolytics go to common. (Apology for not being able to post a schematic at this time.)
I find that if I use 0,066uF from 435V to that diode, with a 47K (1W) from the junction of [diode - 0.066uF] to common, I get -41V and -27V either side of the bias pots, which should include proper bias for the EL34s. I am not sure why I had to use double the series capacitor value compared to Kmaier, but there it is. Incidentally the series capacitor had better be 630V rating (I used two 0.033uFs in parallel, being the only 630V types I have in that capacity range. You could very well use a single 0,068uF/630V).
That then my tested suggestion; it should not take all that much space. Should you find that with 60Hz the bias voltage is too high, take down the 47K/1W to 39K. (Data shows EL34 pp to require -37Vg1 in this design.)
There is a possible disadvantage. The full bias voltage is only reached after 4 - 5 seconds, but as the tubes take longer to get to full emission it is probably academic.
Good luck.
Rather embarassed that I only found time now to further on my post #6. In his post #16 Kmaier suggested the same thing I had in mind, but my component values seem to differ somewhat. We have 50 Hz mains compared to your 60 Hz, which would make a slight difference. But here goes:
Firstly I took the original ST70 bias circuit as basis, i.e. 10K from common to two 10K bias pots in parallel, followed by another 10K to a diode (1N4007, whatever). From either side of the second 10K two 47uF/63V electrolytics go to common. (Apology for not being able to post a schematic at this time.)
I find that if I use 0,066uF from 435V to that diode, with a 47K (1W) from the junction of [diode - 0.066uF] to common, I get -41V and -27V either side of the bias pots, which should include proper bias for the EL34s. I am not sure why I had to use double the series capacitor value compared to Kmaier, but there it is. Incidentally the series capacitor had better be 630V rating (I used two 0.033uFs in parallel, being the only 630V types I have in that capacity range. You could very well use a single 0,068uF/630V).
That then my tested suggestion; it should not take all that much space. Should you find that with 60Hz the bias voltage is too high, take down the 47K/1W to 39K. (Data shows EL34 pp to require -37Vg1 in this design.)
There is a possible disadvantage. The full bias voltage is only reached after 4 - 5 seconds, but as the tubes take longer to get to full emission it is probably academic.
Good luck.
Johan,
Thank you for the help. I am also thinking about using a rectifier tube to get my negative voltage for the bias. There is a schematic of the bias circuit of a Grommes 215BA and it shows a 6X5 being used. I have enough room to shoe horn a 6X4 between the two bias pots the stock location on a Dynaco Stereo 70 chassis. Do you have any thoughts on this? I realize this is also old school tech but I was looking for something just a little bit different to play with. Another thought would be to use a regulator tube to obtain a bias voltage.
Thank you for the help. I am also thinking about using a rectifier tube to get my negative voltage for the bias. There is a schematic of the bias circuit of a Grommes 215BA and it shows a 6X5 being used. I have enough room to shoe horn a 6X4 between the two bias pots the stock location on a Dynaco Stereo 70 chassis. Do you have any thoughts on this? I realize this is also old school tech but I was looking for something just a little bit different to play with. Another thought would be to use a regulator tube to obtain a bias voltage.
Burnedfingers,
I see a few problems with a 6X4 in that role in your case.
Firstly, you will require a separate heater 6.3V source for the 6X4. When the 6X4 is non-conducting the peak positive cathode-heater voltage is going to be >630V when using a general grounded 6V winding - the spec is only 450V peak.
Then using a 5U4 directly heated rectifier, you are going to have more than full h.t. after about 2 seconds until the power valves heat up in about 11 seconds. The 6X4 better heat up before the EL34s or you are going to have a few seconds of low bias while the 6X4 is reaching full cathode temperature. That will probably happen, but narrowly so.
I was once confronted with a decision between all tube, or power supply and protection circuits using modern i.e. semiconductor circuitry. My necessary extra layout for tube technology there would have been at least as complex as the amplifier itself. In the end I let wisdom rule, i.e. use semiconductor circuitry for rectification and safety circuits. 'Solid' rectifiers were already used 'in tube days', even if only by way of those ghastly selenium rectifiers. So in this case I would personally opt for the previous mentioned series capacitor+silicon-diode circuit mentioned. The series 68K in the 6X5-circuit will have to be at least 2W for safety.
Thus for my money - not the 6X4 alternative unless you are over-partial to tubes. I know tube folks would like it all-one-way, but as said in the previous paragraph. (Oh, and with respect, let us just forget about 'preferred tube rectifier sound'. That is a lot of nonsense, but another subject.)
Regards
I see a few problems with a 6X4 in that role in your case.
Firstly, you will require a separate heater 6.3V source for the 6X4. When the 6X4 is non-conducting the peak positive cathode-heater voltage is going to be >630V when using a general grounded 6V winding - the spec is only 450V peak.
Then using a 5U4 directly heated rectifier, you are going to have more than full h.t. after about 2 seconds until the power valves heat up in about 11 seconds. The 6X4 better heat up before the EL34s or you are going to have a few seconds of low bias while the 6X4 is reaching full cathode temperature. That will probably happen, but narrowly so.
I was once confronted with a decision between all tube, or power supply and protection circuits using modern i.e. semiconductor circuitry. My necessary extra layout for tube technology there would have been at least as complex as the amplifier itself. In the end I let wisdom rule, i.e. use semiconductor circuitry for rectification and safety circuits. 'Solid' rectifiers were already used 'in tube days', even if only by way of those ghastly selenium rectifiers. So in this case I would personally opt for the previous mentioned series capacitor+silicon-diode circuit mentioned. The series 68K in the 6X5-circuit will have to be at least 2W for safety.
Thus for my money - not the 6X4 alternative unless you are over-partial to tubes. I know tube folks would like it all-one-way, but as said in the previous paragraph. (Oh, and with respect, let us just forget about 'preferred tube rectifier sound'. That is a lot of nonsense, but another subject.)
Regards
Johan,
Thanks for the reply.
I guess I should have mentioned that I have a 6DW4 damper diode after the 5U4 to slow the warm up on the high voltage. I was going to use a 6.3 winding for the damper tube and the 6X4 that I planned to use. Your correct I am going extra tubes on this amp just for grins. I have to try this and if it doesn't cut it I will take out the 6X4 and go the cap resistor route.
Thanks for the reply.
I guess I should have mentioned that I have a 6DW4 damper diode after the 5U4 to slow the warm up on the high voltage. I was going to use a 6.3 winding for the damper tube and the 6X4 that I planned to use. Your correct I am going extra tubes on this amp just for grins. I have to try this and if it doesn't cut it I will take out the 6X4 and go the cap resistor route.
Right then; your choice. As long as you keep in mind the heater-cathode ratings of the 6X4. Average value=+/-100Vmax. Peak value(for short periods)=+200V, -450V. That includes the peak value of any ac present.
I made an error in the previous post. As a bias rectifier, the maximum heater-cathode voltage will occur with the cathode negative to the heater if the heater winding is earthed. That value is 200V, not 450V. The latter is the maximum in the opposite direction, i.e. with the 6X4 used normally as in a rectifier supplying a positive voltage from the cathode. Apology.
I made an error in the previous post. As a bias rectifier, the maximum heater-cathode voltage will occur with the cathode negative to the heater if the heater winding is earthed. That value is 200V, not 450V. The latter is the maximum in the opposite direction, i.e. with the 6X4 used normally as in a rectifier supplying a positive voltage from the cathode. Apology.
Johan,
Trying to understand this. Heater to cathode rating -450 on a specification that I saw. Does this mean I cannot exceed 450volts on the cathode?
Do I have to have a separate heater winding for the 6X4 or can I share it with my damper tube?
If I have to have a winding just for the 6X4 then can I have the 5U4 and the 6DW4 share the 5 volt winding?
Starting to believe that I should not have punched the 5/8'' hole for the 6X4's socket. But wait!!! If the 6X4 won't work I could always stick in a 0C2 regulator as sort of a night light you see.
I could adjust the resistor value in your bias schematic to come up with enough to light the regulator tube and then tone down the bias voltage from there to something that I could use.
If I manage to get this running then no more holes in the chassis for additional tubes. Just wanted something different.
Trying to understand this. Heater to cathode rating -450 on a specification that I saw. Does this mean I cannot exceed 450volts on the cathode?
Do I have to have a separate heater winding for the 6X4 or can I share it with my damper tube?
If I have to have a winding just for the 6X4 then can I have the 5U4 and the 6DW4 share the 5 volt winding?
Starting to believe that I should not have punched the 5/8'' hole for the 6X4's socket. But wait!!! If the 6X4 won't work I could always stick in a 0C2 regulator as sort of a night light you see.
I could adjust the resistor value in your bias schematic to come up with enough to light the regulator tube and then tone down the bias voltage from there to something that I could use.
If I manage to get this running then no more holes in the chassis for additional tubes. Just wanted something different.
OK.
Firstly, the heater-cathode voltage ratings. The heater is obviously insulated from the cathode, and working at the close spacing and high temperatures they do, a limit on the voltage difference between the two must be expected.
Rectifiers like 6X4 have a higher h-c rating than normal tubes, because the cathode is positive by several 100V in a normal power supply, wrt the heater 6V which usually also feeds other tubes and is normally earthed. Where there is a difference in value depending on which is positive wrt which (h-c), one must understand that the cathode can also emit electrons on its inside. Thus a heater positive to the cathode can despite its insulation draw whatever electrons can creep 'through' microscopic holes in the heater insulation and cause trouble. On the other hand heater insulation is such that it gives off no useful electrons to the outside world (which is the cathode round it), thus [heater negative to cathode] safe voltage rating can be higher. So, using the 6X4 as example, the heater can be 450V negative to the cathode but only 200V positive. These voltages should include the peak value of any ac present there.
You will note that all the time we are discussing heater-cathode conditions. So re your question, you can have maximally +450V on the 6X4 cathode when its heater is earthed. It has nothing to do with absolute voltage on the cathode per se in a circuit. This is also not applicable in your bias circuit; there you have a cathode negative to the heater (when using normally earthed heater voltage also for other tubes in the amplifier).
Coming to your application, the 6BW4 has an unusually high rating of 900V (heater negative wrt cathode). So yes, you can connect that to the heater supply of the 5U4. It will get only 5V instead of 6,3V on the heater, but without too much effect.
But now other things begin introducing themselves. I presume you already have these tubes. Otherwise, consider that the 6BW4 has a maximum continuous plate current rating of only 250mA - it does not help that your 5U4 can 'tolerate' more (as required by the ST70). So, as you use two tubes anyway, why not consider 2 x GZ34/5AR4 (each with plates tied together, per ac phase)? This will solve the problem of using a damper to delay application of ht while other tubes warm up, as well as giving you the advantage of a higher available total current and lower voltage drop, less heat etc. than a single 5U4. (This apart from the bias business.) I do not know whether the economics will play a role in your part of the world.
Finally, then you have a free 6V winding which you can tie to the 6X4 cathode beating that part of the problem. I hope all this helps - please feel free to post further questions. (When working with tubes, it is wise to understand the basics. High voltages are involved, and faults can be spectacular, to say the least!)
Firstly, the heater-cathode voltage ratings. The heater is obviously insulated from the cathode, and working at the close spacing and high temperatures they do, a limit on the voltage difference between the two must be expected.
Rectifiers like 6X4 have a higher h-c rating than normal tubes, because the cathode is positive by several 100V in a normal power supply, wrt the heater 6V which usually also feeds other tubes and is normally earthed. Where there is a difference in value depending on which is positive wrt which (h-c), one must understand that the cathode can also emit electrons on its inside. Thus a heater positive to the cathode can despite its insulation draw whatever electrons can creep 'through' microscopic holes in the heater insulation and cause trouble. On the other hand heater insulation is such that it gives off no useful electrons to the outside world (which is the cathode round it), thus [heater negative to cathode] safe voltage rating can be higher. So, using the 6X4 as example, the heater can be 450V negative to the cathode but only 200V positive. These voltages should include the peak value of any ac present there.
You will note that all the time we are discussing heater-cathode conditions. So re your question, you can have maximally +450V on the 6X4 cathode when its heater is earthed. It has nothing to do with absolute voltage on the cathode per se in a circuit. This is also not applicable in your bias circuit; there you have a cathode negative to the heater (when using normally earthed heater voltage also for other tubes in the amplifier).
Coming to your application, the 6BW4 has an unusually high rating of 900V (heater negative wrt cathode). So yes, you can connect that to the heater supply of the 5U4. It will get only 5V instead of 6,3V on the heater, but without too much effect.
But now other things begin introducing themselves. I presume you already have these tubes. Otherwise, consider that the 6BW4 has a maximum continuous plate current rating of only 250mA - it does not help that your 5U4 can 'tolerate' more (as required by the ST70). So, as you use two tubes anyway, why not consider 2 x GZ34/5AR4 (each with plates tied together, per ac phase)? This will solve the problem of using a damper to delay application of ht while other tubes warm up, as well as giving you the advantage of a higher available total current and lower voltage drop, less heat etc. than a single 5U4. (This apart from the bias business.) I do not know whether the economics will play a role in your part of the world.
Finally, then you have a free 6V winding which you can tie to the 6X4 cathode beating that part of the problem. I hope all this helps - please feel free to post further questions. (When working with tubes, it is wise to understand the basics. High voltages are involved, and faults can be spectacular, to say the least!)
I gave up on trying to use a VR tube in the bias circuit and on trying to use a 6X4 to obtain my negative voltage.
I settled on the circuit that used the .033/630 cap and resistor to ground as stated in post #16 I believe. My local parts supplier didn't have any 43K resistors so I settled for a 39K. I hooked this up to the bias circuit on my SDS power supply board and changed the value of my resistor to ground on that board from a 10K to a 20K. According to my meter this should give me about what I need to bias up all 4 tubes and adjust my 4 individual bias pots. I hope
I settled on the circuit that used the .033/630 cap and resistor to ground as stated in post #16 I believe. My local parts supplier didn't have any 43K resistors so I settled for a 39K. I hooked this up to the bias circuit on my SDS power supply board and changed the value of my resistor to ground on that board from a 10K to a 20K. According to my meter this should give me about what I need to bias up all 4 tubes and adjust my 4 individual bias pots. I hope
Johan Potgieter said:
Ok, Johan, I'm asking cause I want to be sure I understand. I'll focus on the 6X5GT, as the Grommes 215BA schematic shows. See if I've got this right...
The data sheets I have (RCA, GE, Sylvania) show max heater +ve with respect to cathode: 450V, max heater -ve with respect to cathode: 450V.
The 215BA schematic shows 395VAC on the cathode. Assuming the values printed are correct, the heater looks raised from ground (slightly): (10/230)*470 = 20V, giving 395-20 = 375V heater +ve with respect to cathode. This seems comfortably within the 6X5GT's rating.
I see where there's a problem if the 6X5GT warms up slower than the 5U4G (or fails during normal operation): the output tubes will likely fail shortly thereafter. It seems curious the designer chose this route as reliability would be an issue. (Perhaps this is why the 216BA employs a diode rather than a tube in this role!) It would seem this is an amp you would want to keep your eye on!
How's my math/analysis?
Update
I finished my "Spare Parts Dynaco Stereo 70" amplifier Saturday and fired it up on the bench. It worked fine so I started with the listening test. It biased up good and remained stable.
I would like to thank those who took the time to teach me how to go about getting a bias supply without hanging a transformer off the back side of the chassis. I still may use the 6X4 rectifier with the bias voltage and circuit that I have now just to plug up the hole in the chassis that I drilled and punched.
At any rate it turned out to be a good sounding amplifier with a different look and flair. Its kind of interesting to see a 5U4 and a 6DW4 damper tube in the line up. The damper works good taking about 15 seconds before it sees 1 volt out. To top things off the amplifier doesn't have any noise or hum which surprises me.
I finished my "Spare Parts Dynaco Stereo 70" amplifier Saturday and fired it up on the bench. It worked fine so I started with the listening test. It biased up good and remained stable.
I would like to thank those who took the time to teach me how to go about getting a bias supply without hanging a transformer off the back side of the chassis. I still may use the 6X4 rectifier with the bias voltage and circuit that I have now just to plug up the hole in the chassis that I drilled and punched.
At any rate it turned out to be a good sounding amplifier with a different look and flair. Its kind of interesting to see a 5U4 and a 6DW4 damper tube in the line up. The damper works good taking about 15 seconds before it sees 1 volt out. To top things off the amplifier doesn't have any noise or hum which surprises me.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.