DHT filament circuit layouts ?

[PierreMonteau]

Hi All

my first post here ...looking for advice.

I've got reasonable experience building up many guitar amps and some DIY SETs , and have become hooked on bread-boarding tube amps.

I have built a couple of 300B SE, Gary K's 6EM7, and a C3g/PT15 SE amp too, but I'm keen to learn more about the circuits I'm building and eventually design from scratch.

I'm reading furiously to gain knowledge wherever I can find it.

My current problem though is that I'm trying to understand the best way to heat the filaments on a Shishido LW amp - paralleled Ecc83 into 6C4C.
My initial attempt sounds pretty good but the hum levels are too much - so I tried DC heating with simple Bridge, Cap supply. Still too much hum.

So now I've been reading up on constant current supplies using LM318, and LM338 Regulators.

I understadn the principles but I'm struggling to find good examples of construction techniques and suggested layouts to use to build up these supplies.

I just wondered whether anyone can point me in the direction of some good examples of physical layouts ( pcbs / heat-sinking techniques etc )

I'd really welcome the advice.

for this DHT amp I want to separately run about 1.0A at 6v for each of the these 2 6C4C's

Regards
Pete

 

[Rod Coleman]

Current-drive is certainly the way to go. The IC regulators can be hooked up to work as current sinks, but they do not work well with DHT heating, mostly because the IC tries to track the anode-current variations (in effect, it's supply-rejection is not good enough for this application). ICs also add an unnecessary noise source.

With ICs, the sound is degraded quite remarkably compared to using a current drive method that is effectively open-loop to the anode currents. On this forum, since 2004 we have discussed the best way to heat DHTs on this forum, using Tentlabs commercial ready-built module, or the circuit I published on the 'New DHT Heater' thread. This circuit has been refined greatly over the years, and is finally available as a PCB/component kit. See what some of the first constructors think:

http://www.diyaudio.com/forums/tubes-valves/38248-new-dht-heater-5.html#post2336382


The Tentlabs works somewhat differently, but is still current-drive heating:

http://www.tentlabs.com/Products/Tubeamp/Tubefilament/index.html

 

[tomchr]

You'll want DC for the heaters as AC will induce hum. Yeah, there are 'humdingers' and other band-aids to combat this, but I find it better to just not generate the 50/60 Hz interference in the first place.

Some people swear by constant current heating, others by constant voltage. For myself, I have never been able to hear the difference between the two. Nor have I been able to measure any difference. I haven't seen anybody back up their claims of one versus the other with hard data. Hence, I use constant voltage as it is the least expensive and easiest to implement. You can see my schematics here.

If you don't like to use an LM317 or similar for a constant current source, you can build one using an op-amp and a few external components. See a sketch here.

~Tom

 

[astouffer]

Having a floating heater supply can cause hum. Try biasing your ECC83 a few volts above ground or connecting a .1uF cap from the supply to ground. Having too small a filter cap on your DC supply can be bad as well. Have you tried a hum balance pot?

 

[tomchr]

Fix the hum issue first. Then modify the heater supply (if desired). You should be able to get a quiet amp with any type of DC heater.

How much ripple is on the B+ supply? And on the heater supply?

~Tom