I am thinking about making a 2 stage single ended amp using the 71A as output tube.
I am wondering whether the 71A filament can run on AC with low noise.
I have high sensitivity speakers.
I am wondering whether the 71A filament can run on AC with low noise.
I have high sensitivity speakers.
A look at the data sheet shows a 5 V. filament. IMO, you should play safe and use current regulated DC for heating 71A filaments.
2A3s and #45s have 2.5 V. filaments and AC heating is frequently OK. However, the hum increases, when higher filament voltages are used.
2A3s and #45s have 2.5 V. filaments and AC heating is frequently OK. However, the hum increases, when higher filament voltages are used.
Thanks for the input.
With 45 tubes running AC filament, do you need to have separate filament transformer for each tube ?
With 45 tubes running AC filament, do you need to have separate filament transformer for each tube ?
The 45 filaments carry the cathode (music) signal-current, as well as the heating current, so if you share ac-windings across filaments, there will be cross-talk from channel-to-channel, as well as crosstalk of the filament heating current (50/60 Hz and all the mains noise).
A hum pot can balance-out the idle heating currents out of each side of the filament, and into the cathode return circuit, which reduces the 50/60 Hz hum, but does nothing about the hum harmonics (100/120Hz) which the DHT generates. Also the ac-heating current produces huge amounts of intermodulation (50/60 and 100/120 Hz sidebands on every note in the music) - this is not high fidelity!
For a long time, DC heating had a bad reputation, because raw rectified dc causes more problems than it solves - mixing rectifier forward and recovery pulses with the music (shared with a large electrolytic capacitor, which also presents itself in the music signal path).
As Eli mentions, properly-implemented current-control is the way to go. I have DIY kits, if you need a ready-to-go solution ("Coleman Filament Regulator").
A hum pot can balance-out the idle heating currents out of each side of the filament, and into the cathode return circuit, which reduces the 50/60 Hz hum, but does nothing about the hum harmonics (100/120Hz) which the DHT generates. Also the ac-heating current produces huge amounts of intermodulation (50/60 and 100/120 Hz sidebands on every note in the music) - this is not high fidelity!
For a long time, DC heating had a bad reputation, because raw rectified dc causes more problems than it solves - mixing rectifier forward and recovery pulses with the music (shared with a large electrolytic capacitor, which also presents itself in the music signal path).
As Eli mentions, properly-implemented current-control is the way to go. I have DIY kits, if you need a ready-to-go solution ("Coleman Filament Regulator").
Last edited:
I used a 71a preamp (mu follower gyrator load) with filament bias for about 8 months. Loved its tone but preferred 4P1L and 26 so changed after. As with all my DHTs I used Rod's regulator for a extreme quiet output. Hard to beat.
Ale
Ale
i have made two 2A3 sets running on ac filaments, one filament winding for each tube....
hum was never an issue....the sound was very good indeed....
hum was never an issue....the sound was very good indeed....
Same here, three stage amp with 46 driver and 300B outputs, all push-pull. All AC heated, less than 1 mV hum at the speaker outputs, and very nice sound. Low IMD, not "huge amounts". I don't believe it is true to say "this is not hi-fidelity" on the basis of heating method. I have tried direct current heating with both better and worse results. Properly implemented AC can be hi-fidelity.
The 2.5 V. tubes are OK SE on AC. Higher filament voltage tubes are not OK SE on AC. When PP topology is employed, hum cancellation (in the O/P "iron") is occurring. An AC heated SE 6B4G hums like a bee hive. OTOH, an AC heated PP pair of 6B4Gs can exhibit satisfactory hum levels.
Thank you guys for the input.
If I go with DC, is Rod's regulator better than a simple CRC filter ?
If I go with DC, is Rod's regulator better than a simple CRC filter ?
I suggest you read the following DHT filament regulator | Bartola Valves and Why filament regulators in DHT? | Bartola Valves entries and make your own conclusion if you are looking at implementing DC filaments 🙂
Ale
Ale
Is the Pope Catholic? 😉 Yes, current regulated DC is SUPERIOR to a simple CRC filter.
Start out by rectifying the AC with Schottky diodes. Zero switching noise, from the outset, is a good foundation on which the filament supply can be constructed.
So you think rectifying the AC with Schottky diodes and use Rod's regulator, with nothing else, is the way to go ?
Last edited:
Filter well before regulating. Don't use a huge capacitance at the filter's I/P, as that can be a source of "hash". A CRC filter, with the bulk of the capacitance in the 2nd position should get the job done. Using an inductive wirewound resistor is a good idea. You get a tiny bit of smoothing from the part's inductance.
I suspect that Rod uses a 2 stage setup. A 2 stage setup voltage regulates the filtered DC 1st, then it current regulates the feed to the tube.
I suspect that Rod uses a 2 stage setup. A 2 stage setup voltage regulates the filtered DC 1st, then it current regulates the feed to the tube.
i have resisted going dc in my builds so far.....my experience so far is that dc lifting heaters in the low level stages helps and that output tubes can use ac....but this is only me, YMMV...
- Status
- Not open for further replies.