That circuit is correct. I can not find any published limits on capacitance, any reasonable amount should be fine.
What you have to look out for is the load you are going to place on the rectifiers. While the data sheet says 125ma per plate, it also says max of 5.0 plate watts for each plate in the tube pair. At 950 volts this would limit this supply to 10mA.
What you have to look out for is the load you are going to place on the rectifiers. While the data sheet says 125ma per plate, it also says max of 5.0 plate watts for each plate in the tube pair. At 950 volts this would limit this supply to 10mA.
Hi Felipe,
The 6AX4 was never characterized for mains rectifier use. So no solid data exists for the max capacitance. You could sim it with PSUD2 and make sure the peak currents stay below the limits. But I would apply some substantial derating since the peak currents in the datasheet are given for damper use.
I'd say you should be save with 20uF. However choke input would be the better solution also usually in terms of obtainable sound quality. Or low value capacitances. It is a common mistake to load the rectifiers with the maximum possible capacitance they can withstand. Maybe this saves some cost in the filter chain, but quality is compromised.
Some other remarks: Only one heater winding is needed in this configuration. Connect it to the cathodes. Make sure the power transformer has a sufficiently isolated heater winding. With the voltage in your schematic, use at least 2 silicon diodes in series for each leg, otherwiese PIV rating will be exceeded. Otherwise your schematic is correct.
Best regards
Thomas
The 6AX4 was never characterized for mains rectifier use. So no solid data exists for the max capacitance. You could sim it with PSUD2 and make sure the peak currents stay below the limits. But I would apply some substantial derating since the peak currents in the datasheet are given for damper use.
I'd say you should be save with 20uF. However choke input would be the better solution also usually in terms of obtainable sound quality. Or low value capacitances. It is a common mistake to load the rectifiers with the maximum possible capacitance they can withstand. Maybe this saves some cost in the filter chain, but quality is compromised.
Some other remarks: Only one heater winding is needed in this configuration. Connect it to the cathodes. Make sure the power transformer has a sufficiently isolated heater winding. With the voltage in your schematic, use at least 2 silicon diodes in series for each leg, otherwiese PIV rating will be exceeded. Otherwise your schematic is correct.
Best regards
Thomas
Hi!
This is not really correct, the plate dissipation calculates as current multiplied by the voltage drop within the tube. Not the supplied voltage. The voltage drop within a 6AX4 is quite low, depending on current draw maybe 25V.
A supply like this can easily deliver 200mA at 950V
Best regards
Thomas
This is not really correct, the plate dissipation calculates as current multiplied by the voltage drop within the tube. Not the supplied voltage. The voltage drop within a 6AX4 is quite low, depending on current draw maybe 25V.
A supply like this can easily deliver 200mA at 950V
Best regards
Thomas
Hi!
They seem fairly close to the 6AX4, with probably more current handling capability. I wouldn't have a specific preference. I'd expect them to work quite well like most TV dampers.
Thomas
They seem fairly close to the 6AX4, with probably more current handling capability. I wouldn't have a specific preference. I'd expect them to work quite well like most TV dampers.
Thomas
I use a pair of 6AX4GT tubes with a pair of silicon diodes in series before each anode, and they quite happily drive 150uF of capacitive loading at 370 volts. The voltage drop across the 6ax4 tubes is around 25 volts in this circuit which is supplying 200mA
Hi Peter,
why the silicon diodes in series? They can only provide some extra peak inverse voltage capability. The 6AX4 has more than enough PIV for a 370V supply.
While it maybe can handle a large cap, I wouldn't want that. If I have to use a cap input filter I try to keep that cap as small as possible. This always yields better results.
Best regards
Thomas
why the silicon diodes in series? They can only provide some extra peak inverse voltage capability. The 6AX4 has more than enough PIV for a 370V supply.
While it maybe can handle a large cap, I wouldn't want that. If I have to use a cap input filter I try to keep that cap as small as possible. This always yields better results.
Best regards
Thomas
I use around 1000uf cap after the silicon diodes to make the PSU as stiff as possible and as much capacitance after with low esr electrolytics afterwards for a rock solid PSU with a soft start. I follow this with a set of valve regulators down stream for the output and preamp stages each with at lest 100uF of filter on them to lower the impedance of the regulator at HF. This prevents nasty power supply sag even prior to the regulator and after the regulator the filter cap keeps the regulator on track even at high output hi frequency material (eg up to 20KHz). This provides the amplifier the best possible sag-less PSU for ultimate sound quality and performance, and it really works the improvement is amazing. So you get all the slow start properties of a valve rectifier with all the benefits of really steady PSU,
Hi!
So this is probably a high power amp, possibly Class AB? I never had an issue with PSU sag in a pure Class A design. So you have a 6AX4 rectifier followed by 150uF and then another 1000uF after the silicon diodes? This will still stress your 6AX4s. But they are some tough tubes and cheap
Best regards
Thomas
So this is probably a high power amp, possibly Class AB? I never had an issue with PSU sag in a pure Class A design. So you have a 6AX4 rectifier followed by 150uF and then another 1000uF after the silicon diodes? This will still stress your 6AX4s. But they are some tough tubes and cheap
Best regards
Thomas
The 1000uF is between the silicon diodes and the Damper tubes.
The amp is mostly class A up until the last 10% of output, using 4 pairs of 6AQ5 tubes in parallel push pull.
By having a regulated supply for both the output stage and the preamp stages fed from a rock solid psu it separates the amp from line voltage drift and makes sure your amp and preamp are running precisely at the voltages you designed them for.
It does take over 20 tubes of varying types per channel but its worth it!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.