5 VDC heater question

[sgerus]

Here is a simple C-RC-RC 5vdc power supply for a 300b filament.
(see attachment)

I would like to use the 6.3vac 6A from T1, which would be about 7vac at 1.3A.

According to the Power Supply Designer II s/w there will be 3mv p-p ripple.

Is 3mv of ripple low enough?
Do you recommend a filament choke?

Thanks

 

[Colt45]

3mV of ripple is better than 5V

 

[Gerrit Boers]

Is 3mv of ripple low enough?
Do you recommend a filament choke?

If 3mV is low enough depends on the sensitivity of your loudspeakers.

You can find my story here: Heater supply

John Atwood also performed some research on this topic:
DC filament supply test

 

[sgerus]

Gerrit, Thanks for that info.

So the choke is the way to go if you have the room for it and a few more $.

According to PUS Designer II s/w, a RC-RC has about the same ripple voltage as a LC filter.

For my PUS Designer II comparison I used:
C-RC-RC 10,000uf - 1R - 10,000uf - 1R - 10,000uf
Vrs
C-LC 10,000uf - 60mH – 10,000uf

It would be interesting to see how the RC-RC filter looks on the Audio Precsion System 2 that John Atwood used on his test.

 

[zigzagflux]

It would be interesting to see how the RC-RC filter looks on the Audio Precsion System 2 that John Atwood used on his test.

I suspect not as good, especially in the higher order harmonics. While you may be able to produce the same 120Hz ripple with the RC-RC, the choke input is much friendlier with high order harmonic content, drawing a more sinusoidal current from the transformer.

I suggest comparing in PSUD not only the output, but take a look at the input waveforms. True rms values will be different, due to the harmonics. Also compare ripple current requirements for the capacitors; they will be very different. Lower ripple current results in longer life for an electrolytic capacitor.

Fewer high order harmonics makes for less noise. Aside of cost and weight, there are only good things to be had with choke filters.