DC Filaments

[bgkyt1]

Hi all,

I've read topics on this subject, with different ways to do it, as well as people saying it's not necessary...

Some background:

I've had success building Fender circuits, Plexi, and JTM circuits...
with no noise problems,
but a recent TW Liverpool and JCM 800 had hum issues.
JCM was great, but 60 hz hum the more the master was cranked,
and the Liverpool was ridiculous.
I do the headphone test for transformer position and do a decent job with layout, lead dress, etc.
I am not a tech (just a hack) and have no pro gear like oscilloscopes, etc...

I took the Liverpool to a tech, who said any amp with 3 or more gain stages
needs DC filaments on the preamp.
I've seen and heard indy builders with DC filaments and they seem to work great.

But when the tech added it to my TW-style, the amp was still unacceptable to me.
I tried the same circuit in the JCM and afterward it hummed all the time, but seemed higher in frequency.
120 hz maybe? and jumping off the 3rd tube, i was only getting 2.7v per lead.
(a big fail on both amps. they are now scrapped)

I saw an interview with Soldano where he said the only difference b/w a new version of the SLO
was now he has DC filaments b/c modern tubes are so noisy.
It's interesting how I've had no problems with 1-2 stage amps, but 3 is a problem.

the question is:

Is there a definitive way to do DC filaments for a certain amp style (JCM..., Express/Liverpool)
as far as the supply voltage (where/how to get the 6.3v AC: a separate tap, parallel off the first node, etc.),
how much filtering (cap values, how many) ?

is there an existing production amp circuit to use as the benchmark?

How can one approach a new amp build with moderate to hi gain
and not end up with filament hum, or have to roll 50 tubes to find one quiet one,
that can be done consistently and planned for?

thanks...

[pompeiisneaks]

Whether or not it's needed is a big debate.

The original trainwrecks didn't use DC heaters.

Hum issues are almost always one of the following:

Poor lead dress
poor transformer alignment between OT and PT.
poorly soldered connections allowing more signal bleed

etc.

Using DC filaments can and does get a quieter heater circuit, in theory, but poorly designed ones can add more noise than they're trying to remove.

If the noise in the amp is coming from the above listed issues, then DC heaters won't always solve them. If your OT leads are too close to some of the preamp circuitry in the wrong place etc, they can inject hum that's not heater based.

If your heaters are connected to the DC circuit and the hum is being injected before rectification the DC line won't help either.

For HIFI, heater hum can be a much more serious issue, but for almost all guitar amp work, heaters should never be a cause for hum. I've even built an amp with non twisted lines, just running in at direct angles away from the signal wires so they didn't cross at all, except at right angles and had almost 0 hum. This was on a marshall dual with both a JCM800 and 2204/2204 preamp in it. No hum issues at all.

I built a Dumble ODS #124 and even at higher gain, no hum. There's typical gain noise, like hiss, but no hum.

I've built a Liverpool, with typical twisted wires, no hum at all.

finding where hum comes from CAN definitely be a PITA. One way to try and find out where the hum is coming from is to remove one tube at a time starting at V1 and see if it disappears anywhere except at the power tubes. If it is still present even with the PI removed, then it's something coupling into the power tubes, and heaters are too low level signal to actually inject noise into power tubes as I understand it.

There are a ton of amps out there that use DC though, and if done right, doesn't cause any negative impacts and can make higher gain circuits cleaner.

I had a lot of hum issues on one build, a Vox AC30/4 and found it was lead dress of my OT being too close to the pots on the front, and a simple realignment of the OT leads removed all hum.

One of the better articles I remember reading about this is here: http://valvewizard.co.uk/heater.html

he goes over a lot about the myths/mysteries and good ways of handling heaters, including DC heaters and how to supply them correctly.

~Phil

[xtian]

Grounding SCHEME is hugely important. PT and Heater CTs should be grounded near PT, and first gain stage grounded near input jack, far away from the noisy PT. Et cetera.

http://www.valvewizard.co.uk/Grounding.pdf

[bgkyt1]

thanks.

on the liverpool, the hum went away when i pulled the pi tube.
and had less hum if i pulled v2...

if i follow what i think are best practices from forums, valve wizard, etc
and look at other chassis pics for a reality check when i doubt placement of a part or wire,
i thought i was doing fine.
and on fender and other 2-stage circuits, that seems to be the case.

[mirage_indigo]

Figure out where the hum is coming from, then figure out what to do about it.

There's a bunch of low hum techniques any or all of which may or may not help depending on what is causing the hum.
What xtian said, proper star grounding makes a huge difference.

Isolating the input jacks (input and effects return) which is also part of proper star grounding.
There are various heater schemes, not just DC (also not usually it's DC just for the preamp tubes, not the power tubes).
Shielding the wire from the jack to input to the first tube input.

[stephenl]

In addition to above;

• Add a hum balance pot.
  Add a negative dc reference to the AC heater supply - this can be connected to the wiper of the hum balance pot.
  Supply the preamp heaters from a separate small tranformer. Go with 12v heaters for the preamp.
  Rectify>Zener>Lots of capacitance filter and do DC only on the first preamp tube.
  Use SiC Schottky rectifier diodes for the B+ and short wire runs to the reservoir cap. Schottkys have little to no reverse recovery...

[bgkyt1]

thanks, everyone, for the info.