DC Filament supply boards

[pmillett]

I've had a number of people ask me to make the DC filament supply boards I made for an earlier project available. I've finally gotten around to it...

Regulated DC filament supply

I made the PCB the same size/shape/mounting as the Tentlabs supply, so you can try this one (constant voltage) and replace it with the Tentlabs voltage-controlled-constant-current supply without any modifications to your amp.

Pete

[Wavebourn]

Hi Pete;

do you remember Dell & electrolytic capacitors issue?
The problem is, motherboard developers heated them up too much...

[ChrisA]

Originally Posted by pmillett

...
I've had a number of people ask me to make the DC filament supply boards I made for an earlier project available. I've finally gotten around to it...

I'm wanting to do something like this myself. One question. You say the CML drops some voltage. Why not have the regulator sample the voltage at the output terminals so as to compensate for the (load dependent) voltage drop. (Yes I know the heater load stays constant but some of use swap tubes around.) The cost and BOM remains unchanged.

Maybe there is a problem with this? Oscillation maybe?

In the past I've built power supplies for remote loads that had 50 or 150 feet of cable between the supply and load. I'd run one small gauge wire in the cable to sense the voltage at the load end of the cable so as to automatically account for voltage drop in the cable. Seems to me the inductor is exactly like a long power cable, only rolled up.

[pmillett]

Originally Posted by ChrisA

Maybe there is a problem with this? Oscillation maybe?

Yup. If not oscillation, at least some AC response issues. One of my mantras is that if you want to use a regulated power supply in audio, make darn sure it does no harm. To me, that means the AC response has to be carefully controlled.

IMO, that's why shunt regulators tend to work well in audio. Less likely to get in the signal path (not that they do not or cannot...)

Taking the feedback after the CML could result in some AC response anomolies that might not sound so good...

Pete

[Gordy]

ChrisA, sounds as if you may have been fortunate not to get oscillation with your long run. (Or may be just well designed ). I know that extending the negative leg (after / under the resistor) to the load can work well, but that does not normally apply to the upper / feedback arm.

Originally Posted by pmillett

Taking the feedback after the CML could result in some AC response anomolies that might not sound so good...

Too right Pete. Experience suggests that the only thing that should go in the feedback loop of a three terminal regulator is a resistor, and that it should sit right next to the regulator.


Nice board Pete, thanks for posting that; I am sure it will be very useful to many of us.

[ChrisA]

Originally Posted by Gordy

ChrisA, sounds as if you may have been fortunate not to get oscillation with your long run. (Or may be just well designed ). I know that extending the negative leg (after / under the resistor) to the load can work well, but that does not normally apply to the upper / feedback arm.

We were supping power, a handful of amps, to a thermoelectric cooling system. Bundled in the cable also was water tubing for the heat sink. If the load is not reactive and has a time constant of several seconds not much can happen. I think the "sense wire" was connected to a voltage divider that was in parallel to the load. You can do it. It's a pretty common technique.

[DualTriode]

Hello Pete,
I just received a pair of your heater PCB’s in the mail. They are fine. Good job. Now the job is to see how low the noise can go on a 307A SET circuit.
DT
All just for fun!

[regal]

Are you planning on making boards for the HF oscillator DHT heater design?

[pmillett]

Originally Posted by regal

Are you planning on making boards for the HF oscillator DHT heater design?

I'm working on a new design for an ultrasonic filament supply... it will take a little time, but it looks promising.

Shooting for 2 x 2.5A at 125kHz, transformer isolated sine wave, in the same form factor as these boards.

Stay tuned!

Pete

[Pano]

Originally Posted by pmillett

Stay tuned!

At 125Khz, I'll have to break out the old long wave radio to do so!