That won't increase current capacity. It might create diode noise though, because the two diodes are no longer isolated from each other by the electron stream within the tube.
Jack
That won't increase current capacity. It might create diode noise though, because the two diodes are no longer isolated from each other by the electron stream within the tube.
Jack
Pio1980
AK Member
Perhaps draw it out on paper then, the two diode strings function normally with the two tube sections paralleled after, in series with them, which shares the current between the sections. If one section is ca 2## mA, then both in parallel will be 4## mA, depending on the individual example.
Last edited:
I suppose the argument has to do with duty cycle. Its rated for 250ma (I think, don't quote me and it doesn't truly matter) running full wave but only one plate is conducting at a go. Parallel the plates and in theory its good for 500, but the plates are now conducting 100% duty cycle instead of 50% so I don't know that it tracks exactly the same. I guess its not precisely 50/100 because of zero crossings and all that but close enough for example purposes. I'd think it should be good for at least some increase because of effectively doubled plate and cathode surface area but maybe not double the rating.
Similar to how horizontal output tubes have extreme voltage and pdiss ratings with a * note that says not to exceed 15% duty cycle.
I've seen multiple things with parallel rectifiers and most of them seem to use each rectifier full wave, so one plate in each tube conducting at a time. I have seen a few things with each tube run as a half wave with the plates paralleled but both cases have the tubes at 50% duty cycle.
Similar to how horizontal output tubes have extreme voltage and pdiss ratings with a * note that says not to exceed 15% duty cycle.
I've seen multiple things with parallel rectifiers and most of them seem to use each rectifier full wave, so one plate in each tube conducting at a time. I have seen a few things with each tube run as a half wave with the plates paralleled but both cases have the tubes at 50% duty cycle.
Pio1980
AK Member
I've wondered about duty cycle effect in this case, but take @dcgillespie word on this. Certainly peak current handling is increased if not doubled.
We should be clear about what’s happening in this circuit. In both cases - anodes connected together and not - only pulsating DC is applied to the 5AR4 anodes by the solid-state rectifiers. When the anodes are connected together, each passes half the current all the time, rather than all the current half the time. On that basis, average current at each anode doesn’t change. Nor has peak current capability increased, relative to the specification of 3.7A. Peak current in this case is defined as a non-repetitive current draw not exceeding 0.2 second duration at the cathode. Installing solid-state rectifiers ahead of the tube and connecting the anodes together does not reduce that current or allow more to be drawn. In essence, nothing has changed by connecting the anodes together, other than the potential for diode switching noise. The original circuit that merely adds the diodes protects the tube from excessive PRV, and it does so without that risk.
Jack
Jack
Pio1980
AK Member
Dave @dcgillespie can take that up if he chooses to do so, he has far more experience and knowledge than myself.
Last edited:
Lavane
Lunatic Member
Post #618. Is this what your were referring to?
https://audiokarma.org/forums/index.php?threads/dynaco-st-70-base-line-testing.578485/page-31
https://audiokarma.org/forums/index.php?threads/dynaco-st-70-base-line-testing.578485/page-31
FWIW, I employed this technique in the design of my large SETs about 15 years ago. The only difference is that I used a solid-state FW bridge ahead of the rectifier, and the tube was a 5R4GYB, rather than the 5AR4. When it's connected this way, the rectifier tube only serves to provide a time delay. I felt that was important because B+ was over 1 KV. The point is, I'm not speaking theoretically on this topic. I've implemented it myself, and I understand the concept pretty well. What IS theoretical is whether the diode switching noise created by this technique is an issue. Some builders seem convinced it radiates into the signal path and degrades amplifier quality. This has led to claims that only Schottky or UF rectifiers should be used. I'm not one of those people, and in fact, I've never observed diode switching noise riding on the signal path of any tube amplifier. So, in that regard, the switching noise this technique creates might or might not create a tangible problem. I only intended to point it out as a side effect of connecting the two anodes together.
Jack
Jack
Pio1980
AK Member
One of them.
I see no valid basis for such a characterization of cathode area as it relates to current capability. Just as at each anode, half the current all the time does not create greater capacity than all the current half the time. Please consider this not only in terms of average current, but also of power dissipation within each cathode structure.
Jack
Pio1980
AK Member
DCG discusses the PS limitations of the Dyna Stereo 70 here, and mentions the optimal approach to be SS rectification with a 50 Ohm/10W series resistor to mimic that of the tube, but that produces an unloaded voltage peak on start-up.
.
https://audiokarma.org/forums/index.php?threads/dynaco-st-70-base-line-testing.578485/#post-7609452
Some folks want to keep the tube for delayed B+ ramping, the mod provides this, and the peak current charge limiting resistance for the tube rectifier rated PA 060 PT. It is still a stretch for the paralleled single 5AR4 at full output, both channels together @ 500mA +.
.
https://audiokarma.org/forums/index.php?threads/dynaco-st-70-base-line-testing.578485/#post-7609452
Some folks want to keep the tube for delayed B+ ramping, the mod provides this, and the peak current charge limiting resistance for the tube rectifier rated PA 060 PT. It is still a stretch for the paralleled single 5AR4 at full output, both channels together @ 500mA +.
Last edited:
Pio1980
AK Member
True, it's the total capacity of the tube in either condition in question to carry it.
Pio1980
AK Member
Cathodes, plural, in parallel. Not singularly. The resistance will be halved as would any identical resistances in parallel, but likewise, current carrying capacity and wattage rating would be double a single example. And yes, voltage drop should halve and be even comparatively less under hard loading. A slight increase in power would be expected at higher B+.
Pio1980
AK Member
Bob has retired and passed tubes4hifi back to Roy Mottram. Dunno who's running the forum there.
Last edited:
It's not the same as two resistors. Voltage drop VS current isn't a linear function. Charts for a single section of the tube show this. For example, whereas 100 mA creates a drop of 10 volts, 400 mA does not create 40 volts. Instead, the drop is slightly less than 25 volts. The net result is that operating the sections in parallel at half the current each does not halve the voltage drop. It would be interesting to test this on the bench, with and without the jumper, just to see how well the real-life circuit follows the charts.
Yes, but not double that of a single section conducting only half the time.
Jack