Faux ground locations for 6.3v heaters

[pjd3]

Hello,

Finally down to the last mile on the stereo tube amp build (Plexi preamps with Deluxe reverb type output sections, each with their own PT and rectifier caps/diodes,etc...).

I realized that the two 6.3vac heater sources may or will need a couple of 100 ohm resistors to ground since neither heater coils have a center-tap.

Since this amp has isolated inputs and isolated speaker jacks, where in the topology of things would you expect to be the best place to land these 100 ohm resistor pair? Or, since these "two amps in a chassis" is at least attempting to live with isolated inp and outputs, will these 100 ohm resistors even be an issue?

Whats not clear in my head is if the presence of these two heater 100 ohm resistors is to address the potential of ground loops OR, are these 100 ohm pairs meant to address an imbalance in the heater wires that could lead to hum in the system due to magnetic fields emanating from the wires?

Please feel free to educate me a bit on this - that may be what I need to better decide where to ground the 100 ohm resistors, if they are even needed at all in a floating inp/out stereo amp.

Thanks everyone!
Best,
Phil D

[martin manning]

Whats not clear in my head is if the presence of these two heater 100 ohm resistors is to address the potential of ground loops OR, are these 100 ohm pairs meant to address an imbalance in the heater wires that could lead to hum in the system due to magnetic fields emanating from the wires?

You definitely need a ground reference of some kind. If left floating the heater winding will drift up due to capacitive coupling with the HV winding, and that will inject hum into the cathodes. Most commonly the CT or faux CT (2x 100R) is grounded with the HV CT, but some people like to ground it at the input.

[pjd3]

Thank you for that Martin.

I'll send it to the HV CT to see how that works out. I admit I was initially afraid to do that as I thought since the 6.3vac is a separate coil from the HV CT that it might introduce some issue. I have in the past grounded those 100 ohm resistors right to the chassis straight from the Fender Jewel lamps, and that seemed to work but, since this amp is a bit different from all the others I've build, it seemed like something worth worrying about!

Thanks again, Martin,

Best
Phil D (pjd3)

[lonote]

I borrowed an idea from Sluckey that I have use on my last few projects with good results & it is now standard practice.

Just a few extra components gives you a reference point for an elevated heater CT (either real or artificial) & also bleeds-off the caps on the power rail after shut-down, making it safer to work on.

Heater Elevation-Bleeder.jpg

[R.G.]

Per the dogma... er, advice... in Guitar Amp Wiring Notes, the idea is to decide what currents run through the "ground" wires, and what you want the resulting connection to do to .. er, for you.

Two 100 R resistors across 6.3(ish) volts should have not much current running into or out of the connection - it's a reference potential, not a current carrier. Any current that goes into/out of the resistors is only any unbalanced leakage from the tubes and wiring. So it ought to go somewhere quiet-ish, but not critical. Bad idea to connect it to the signal ground wire in the signal path, for instance.
The One True Ground at the negative of the first filter cap is never a bad choice, as any leakage currents are prevented from contaminating any signal grounds. Chassis is a crap shoot, as always; maybe good, maybe bad, depending on random other things that might be attached to chassis.

An interesting point is whether with isolated ins and outs you have any ground connection between the two signal paths. Do they share a One True Ground at a single first filter cap, or are the signal grounds isolated entirely, perhaps with dual first filter caps?

[pjd3]

Thanks everyone for coming along.

Martin seemed pretty confident in his comments on how to go about landing the two 100 (or in my case, 4 on each amp because I decided to use one of the 6.3vac coils for preamp filaments and the other 6.3vac for the 6V6 filaments - each of the two PT's have 2 6.3vac coils, 2A and 3A, no CT's).

Martin suggested to bring the 100 ohm resistors to the main power ground so, thats what I went forward with. I should mention that I'm keeping as much away from the input and preamp sections as possible.

About an hour ago I brought the 100 ohm resistors from their respective heater wire to the main power ground that is directly connected to the first filter cap and PT CT. To clarify R.G."s question, this stereo amp is truly composed of two totally individual amps, each with their own power transformer and their own rectifier with filter caps. You could literally band saw the chassis straight down the middle and be left with two fully featured amps in every respect. The only thing that is common to both amps is a little Type A DPDT on-on-on switch that grounds the right input inp/grid, the left inp/grid or no grounding at all middle position so that you can A/B each channel for comparison. Each input grounding stays with its own preamp channel.

So my decision for starters was to send the 100 ohm pairs straight to the main power ground, where the first main filtercap meets the PT CT.

Ya think thats a worthy starting point?

Thanks guys, lts great to have this kind of mentoring!
Best,
Phil D.

[wpaulvogel]

I wouldn’t reference the heaters to the point of the highest current return. I’d reference it to the lowest current, the input. That’s the point of the lowest noise and it should be the quietest possible point. That should keep the input triodes quiet.

[R.G.]

I wouldn’t reference the heaters to the point of the highest current return. I’d reference it to the lowest current, the input. That’s the point of the lowest noise and it should be the quietest possible point. That should keep the input triodes quiet.

With respect, I believe that would be a mistake. There is a long discussion of why this is so in my "Guitar Amps Wiring Notes" article, linked in this thread. The negative terminal of the first filter cap would present far fewer ways to introduce hum. Give the article a read, and we can talk about it.

[pjd3]

Without knowing for certain, I would have been worried about any wiggling on the 100 ohm faux heater ground being translated into the sensitive input section, if, that were something that could happen.

Phil D.

[wpaulvogel]

I wouldn’t reference the heaters to the point of the highest current return. I’d reference it to the lowest current, the input. That’s the point of the lowest noise and it should be the quietest possible point. That should keep the input triodes quiet.

With respect, I believe that would be a mistake. There is a long discussion of why this is so in my "Guitar Amps Wiring Notes" article, linked in this thread. The negative terminal of the first filter cap would present far fewer ways to introduce hum. Give the article a read, and we can talk about it.

Not to start a fuss but I’ve referenced the true center tap of over 50 amps at the input jack reference point of the chassis and it’s mouse quiet. Being that this is an artificial center tap, I can’t speak for this 100% but I’d create the tap as close to the power transformer as possible then run the wire along the corner of the front panel and to the input jack chassis reference point because the center tap carries no current and the input jack carries the least current.

[trobbins]

I'm with wpaulvogel on where to connect the humdinger. My reasoning is that it is the heater wiring to grid capacitance that is likely the dominant hum injection mechanism, and so the voltage waveform from humdinger to grid should be as pure as practical and not include residual voltage signals that may occur between an input stage gnd node and say a power supply gnd node.

The other hum injection mechanism is through the heater to cathode interface, and coupling into the cathode circuitry. Again, that leakage loop of current can be separated as a resistive current and a capacitive current, of which the resistive current can be minimised by heater dc elevation, and/or tube rolling. The impact on the cathode circuitry can be minimised by bypass decoupling of the cathode to the input stage gnd node.

[Helmholtz]

Not to start a fuss but I’ve referenced the true center tap of over 50 amps at the input jack reference point of the chassis and it’s mouse quiet.

Did you verify that your method gives better results than grounding the heater CT with the reservoir cap?

Actually I think it shouldn't make much difference because of the extremely low heater to ground current.
But from theory I'm with R.G..

[wpaulvogel]

I wish we could stop calling it grounding. What we’re doing is referencing the heater circuit to chassis potential. The chassis is grounded. The heater circuit has no current that is entered the chassis. We’re referencing it to this point so that it remains set here and isn’t floating. The other circuits with the amplifier may use the chassis metal as a conductor and return current to their source but none of the current in the chassis is moving through the safety earth ground connection that leaves the chassis unless there is a fault.

[trobbins]

The 'performance' difference is likely not measurable if the 0V distribution and ground link are done reasonably, as the topic relates to residual signals interacting with secondary effects that may be likely masked by other forms of circuit design that minimise hum.

It may be worth noting that having a humdinger pot in one location, and then pedantically trying to run a link wire from the wiper to another distant part of the amp may by itself introduce other forms of noise/hum. Also, amps with ss diode rectification may introduce noticeable glitches in various waveforms and parts of the circuit, and may be a higher priority to suppress than the topic here. I recall Merlin posting a scope plot of heater humdinger related residuals many years ago, and the residual can be a combination of many influences.

[pjd3]

OK,

Lets say that there is a little truth in all of these thoeries.

Since these amps have two 6.3 coils on each PT, and that I did utilize both coils on each transformer - the 2A coil feeding the preamp tubes filaments, and, the 3A coil feeding the 6V6 pair, might it make sense to bring preamp tube 2A filament coild 100ohm resistors to the input ground and the 3A filament coil to power ground?

Thanks everyone, theres a new education in this for me, once again.

Best,
Phil D (pjd3)