My Tubelab SE unanswered question

[Nikon1975]

Dear All,

I'd like in my new Tubelab SE to reduce the voltage for the power drive to 200V to limit the heat consumption.

The question is: how much current should I make available ? I tried to measure, but it looks like the current drain is almost zero.

Thanks,

Davide

 

[Tubelab_com]

The average current drain is set by the resistor from the source of the fet to the negative supply. I usually aim for 5 to 10 mA to limit dissipation in the mosfets. If your positive voltage supply is lower, then you can increase the current to improve the dynamics a bit. I have found no benefit to going beyond 20 or 25 mA. These current numbers are per channel.

For my test amps, I use an isolation transformer with two 120 volt secondaries which I put in series. With a bridge rectifier and filter I get a +/- 160 volt supply for PowerDrive. That works well.

 

[Nikon1975]

Thanks!!! I will be using the regulator from diyhifisuppy, that has -120v supply. For the positive suppy of the powerdrive i will use a secondary winding of the main transformer. As I will have it custom made, i will request 50mA 150v. That should make.

We will see what will come out.

Davide

 

[Nikon1975]

George,

Another question: what filament voltage are you aiming for the 5842 tubes ? I notice that with the current arrangement it is below 6.3 V. Do you think there is any advantage to run a separate filament supply for the drivers ?

Thanks,

Davide

 

[Tubelab_com]

Another question: what filament voltage are you aiming for the 5842 tubes ? I notice that with the current arrangement it is below 6.3 V

I aim for 6.3 volts. The 5842 is the most variable tube that I have ever used and that includes the heater current. I have seen 5.4 to 6.7 volts with the stock configuration. The 5842 heater voltage depends on the power line voltage and tube current. I usually just tweak R3 as needed.

If you were building the ultimate amplifier with off board power supplies a good DC source for the 5842's might be a good idea. I did that in my 845SE amp which used a Tubelab SE for the driver.

 

[Nikon1975]

Will do like this.

Then I think it makes sense also to lower the AC for the 300B filament, so I reduce the heat production of the regulator. Am I correct ?

Thanks,

D.

 

[Nikon1975]

As I am pretty flexible with the power supply at this stage I was thinking something like this.

The module that I am using needs the rectifier, so I am using the rectifier on board, mainly to hold the tube in place.
The output of the module will be +400V and -120V. The B+ from the module should be mainly ripple free.
I will have a separate supply CRC for the power drive.

Now I can feed the B+ directly to the OT and feed the board with the power drive supply of around 200V. Like this the drivers and the power drive will share the supply, and I will reduce the heat also on the CCS.

The disadvantage is that the driver will not have a supply as clean as the power tube, but I noticed that the CCS makes his own work to reduce ripple.

Am I correct ? Does it make sense ?

For the filament, I will use the on-board supply for the 300B, feed with something around 5.5VAC and a separate PSU feeding 6.3 V 0.8 A regulated to the drives. I will have an ACof around 7VAC.

Please comment on my plan. Maybe I am missing something.

Thanks,

Davide

 

[Nikon1975]

And the other disadvantage is that the drivers will be feed with a supply done with diodes rectification, not tube.

D.

 

[Nikon1975]

Any input?

Thanks,

Davide

 

[Nikon1975]

So I think I will use the separate supply only for the mosfet. I set on 160 VAC 220uF 270R 220uF. Does it make sense here to have something like PP caps at the output, let's say 10uF or something smaller.

Additionally, I have two ammeter to connect instead of the resistors. Now they will be directly in the signal path. Should they bother me ?

I plan to connect one pp cap for each channel right at the OT. I think this should mitigate any possible negative effect of the meter. Am I correct ?

Thanks,

Davide

 

[Nikon1975]

Sometime you think you understand things, and experience shows you that you did not.

The question is general on the CCS: I put the plate of the driver at 175 V with 10mA flowing through the tube. From the specs of the semiconductor I understood that it needs few volts of headroom to work, so I thought 200V would be enough.

But it is not the case, as if I don't raise the supply voltage up to almost 300 V I get visible positive clipping, well before a 175 V swing.

Can somebody explain this ? I thought that in this configuration the plate could swing between 0 and 175 V.

Thanks and sorry for my lack of basis.

D.