Artificial centre tap for heaters

[TIMBO]

Hi guys, I just need some help with the heater centre tap. Most schematic show the two 100ohm resistors going to ground, i have also noted them being tied to the cathode of the power tube. Could anyone give me some guidance. Thanks

[tubenit]

You looked in Hoffman's Library of Information under heater wiring and grounding schemes ............. correct?  

Geezer tied them to power tube cathodes per his Tweed Overdrive schematic. Reportedly, this can in some instances reduce hum/noise.

With respect, Tubenit

[Fresh_Start]

The virtual center tap can be "grounded" at any available AC ground.  sluckey finally got the notion through my thick head that you can have an AC ground with positive (or negative) DC voltage.  IOW all notions of voltage are relative.

I don't remember the theory right now, but "grounding" the heaters' virtual center tap at a DC voltage between 40 and 70 volts can help reduce noise.  Even the moderately elevated DC voltage found at a power tube cathode can be used. 

In some circuits with a DC-coupled cathode follower, the tube's maximum cathode-to-heater voltage limit get's pushed in the cathode follower triode.  Elevating the center tap for the heaters reduces this problem.

From the AX84.com FAQs:

47.Does DC heater bias really help with hum?
This question arose in another discussion: how many db does raising the heater voltage by about 40 volts dc reduce the hum? These measurements (db) are from a Modified HO I am working on:


Frequency                                          60 120 180 240
Background (G1 up, G2 down, MV up): -70 -52 -63 -62
G2 adj. for good hum, 0 v. dc              -50 -22 -42 -23
Same G2, 40V dc                               -55 -36 -62 -52

Yep, simple, elegant, and it works.

(These are measured hum voltages, and subject to the usual limitations of measurements. The actual improvement is probably considerably better.)

-Mike Sulzer

Here is Merlin's discussion of heaters, including using an elevated DC voltage as a reference for the virtual center tap: http://www.freewebs.com/valvewizard/heater.html

See... I don't have to remember anything except where I put my bookmarks!

Cheers,

Chip

[TIMBO]

Thanks guys, i was just want to know which was the better way, as i was not sure that using the cathode would work.I'll try just grounding the resistors first. Thanks

[bluesbear]

I've never been able to hear the difference EXCEPT with single-ended amps. It'll make all the difference with a noisy Champ.
Dave

[HotBluePlates]

I don't remember the theory right now, but "grounding" the heaters' virtual center tap at a DC voltage between 40 and 70 volts can help reduce noise.  Even the moderately elevated DC voltage found at a power tube cathode can be used. 

The first tubes did not have a heater and cathode arrangement, they had a filament. It emitted electrons when hot.

All tubes, from diodes on up, work on the basis of having a structure emits electrons when heated, which are then collected by an element more positive than the electron emitter (filament or cathode).

The heater in a tube is usually connected to a d.c. voltage near ground; the cathode is often positive of ground by the amount of bias (in a cathode-biased stage). That sets up a condition where current could be emitted from the heater (the emitter) to the cathode (the "more positive" collector). This is like a half-wave rectifier circuit, where the heater and cathode form a diode.

If the heater is referenced to a d.c. voltage higher than the bias voltage at the cathode, the "diode" is reverse-biased, and no current can flow. This can kill hum if there is hum in the circuit due to rectified current flow from heater to cathode.

And as Chip noted, it reduces the heater-to-cathode voltage difference when you have tubes with an elevated cathode (like cathode followers and long-tail inverters).

It is not a cure-all, because there are other causes of heater-induced hum that this does not address; nor is it always needed.

[TIMBO]

Thanks again, posting on this forum is a great learning tool. I would like to add that the original schem (champ) showed that the heater tap used the ground as part of the circuit, can this be a possible cause of hum to an amp?Thanks

[FYL]

I would like to add that the original schem (champ) showed that the heater tap used the ground as part of the circuit, can this be a possible cause of hum to an amp?

Fender used the chassis as a heater return on a number of vintage amps. Hum was quite high, but the speakers used had a very poor low frequency response - nothing really useful below 140 Hz or so for a 6" mounted to a vintage Champ 600 or wide panel tweed, slightly better with 8-inchers as fitted to the Champ 800 and narrow panel tweeds.

[HotBluePlates]

And just saying "Champ" makes things a little uncertain. Fender made amps named "Champ" (or Champion 600/800) since the late-40's. So there are a lot of different individual amps you could be talking about.

One type used only 1 heater wire running between tubes; the other heater pin was connected to the chassis, and used that as a "wire". It was done in other Fender amps of the time. It might work out okay, ar you might get a lot of hum. It is best not to do this. [I had a '55 Tremolux that was wired this way. There was no obvious hum, and no lack of low end. I considered myself very lucky.]

There are a lot of possible sources of hum when dealing with heaters, and a lot of tweaks to cure each source. I notice that London Power attacks nearly every source in their amps, but experience shows that approach is overkill (except maybe in production amps where you can't fine-tune each one). In practice, twisted leads and a center-tap (on the PT or an artificial one made of 100 ohm resistors) prevents problems in well over 95% of all amps.

And some possible source of heater hum can't be solved with the heater circuit, but require a signal circuit design change (or pre-screening to select good, quiet tubes).

[TIMBO]

Thanks HBP, The Champ i referred to is a kit 5F1 that has printed circuit board which i assume uses the earth trace as part of the heater circuit. I did not purchase the kit only the transformers and not know that the PT did not have the CT on the heaters. So thanks for the guidance.

[PRR]

> printed circuit board

There is NO good way to get low hum with PCB heater wiring.

Live with it, or fix it. First verify that 6VAC flows in PCB "wires", simple circuit tracing. If so, cut the heater lines AT the socket pins, leaving only the pin solder-blob. Get some hookup wire and run heater power OFF the PCB, in twisted pair cable AWAY from all audio points (basically everything except the heaters). Study some of the excellently-wired amps posted here.

[RicharD]

>There is NO good way to get low hum with PCB heater wiring.

DC.

Anything I've ever done PCB had DC filaments.  This still leaves the concern of current capabilities of PCB traces.  Big ol honkin 1/4" traces ought to do it.  The quick fix is to dremel through the heater traces and wire up and away with a twisted pair.  AC filaments are fine as long as long as they are kept away from signal paths, especially when run in parallel routes.