Individual Cathode Bias Resistors and DC elevated heaters

[brilliantblue]

Hello and thanks for reading! I've been building an amp using Push Pull 6AQ5 output stage, cathode biased - currently each tube has an individual 2w 470 ohm resistor with 22uf/50v cap on each as well. This is the first amp I've done with individual cathode resistors/caps at the output stage, and it has brought up a few questions which I've been unable to find clear answers for.

1) is the 2w rating for each resistor high enough? I had them on hand. 265v plates, 15v cathode-to-ground. I have an old projector amp schematic showing a 2w SHARED cathode resistor at 200 ohms for similar plate voltage, and thought I was essentially doubling the rating when doing individual resistors, but I've run into different beliefs in that area.

2 and 3 are heater noise related:

2) I'd wanted to use DC elevation - is there still a way to steal it from the output cathodes?, or do I need to grab it from elsewhere? I attempted lifting it from only one of the output cathodes, but it caused that tube's cathode voltage to drop to 4v, and reduced plate voltage as well - and made a stink.

3) I've noted that my heater noise is coming from V1 (2 stages of preamp - still not settled on 12ax7 or 12ay 7) is a low, slowwwww pulsating 60hz hum, not very loud either. I pull the tube and the amp is virtually dead quiet. This amp is to be mated with a 10" alnico and small baffle/open backed cabinet. I'm wondering about dropping the value of the output section cathode bypass capacitors. And that brings up the question of the effect of the individual bypass caps in the output section and how to adjust their value for response eq. Would dropping to 10uf each roll off some lows?

Thanks for your time and insight!
Jeff

[Malcolm Irving]

When you go to two independent cathode resistors they each dissipate half the power of the shared version (assuming that they each have double the resistance of the shared version). If you have 15V and 470 ohms, the dissipation will be 15x15/470 = 0.5W, so a rating of 2W is fine.
To keep the turn-over frequency the same, the independent bypass caps should each have half the capacitance of the shared version.
For DC elevation, the voltage on the independent version is not as constant as the shared version (but you could make it more constant by increasing the C value). However, it should not have caused the problems you mentioned - I wonder if your heater circuit is already grounded somewhere else? Hopefully, sorting this out might fix the hum problem. Rolling off the lows would reduce the 60Hz hum, but would also reduce your 82Hz low E note.

[brilliantblue]

Thanks Malcolm!

When I removed the dc elevation, the tube returned to 15v, matching its partner 6aq5.

When disconnected from the transformer, the heater chain shows "open" in reference to ground on my multimeter at its highest setting (20M).

My heater wiring isn't great around v1 - and I suspect that is where my hum is coming in - it's all point-to-point and "finished" or else I'd correct it. At any rate, it is quite a low hum in both frequency and volume. There's a gentle slow "pulse" to it, about once per two seconds, which doesn't seem to degrade the tone.

[eschertron]

1) is the 2w rating for each resistor high enough? I had them on hand. 265v plates, 15v cathode-to-ground. I have an old projector amp schematic showing a 2w SHARED cathode resistor at 200 ohms for similar plate voltage, and thought I was essentially doubling the rating when doing individual resistors, but I've run into different beliefs in that area.

from the power law [P = IE, or P = E^2/R], the average power dissipated through the shared resistor is 15v^2/200R = 1.125W, so the 2W resistor is OK, not quite the safety factor of X2 that some go by. The power individually is 15v^2/470R = 0.479W (assuming the cathode voltage is about the same, you didn't say). This is kind of intuitive, isn't it? the power tubes are still drawing the same current to make the same power, so why wouldn't the cathode resistors each have half? So by using 2W resistors for the individual cathodes, you have indeed doubled the rating (the safety rating, that is!) of the resistors.

2) I'd wanted to use DC elevation - is there still a way to steal it from the output cathodes?, or do I need to grab it from elsewhere? I attempted lifting it from only one of the output cathodes, but it caused that tube's cathode voltage to drop to 4v, and reduced plate voltage as well - and made a stink.

The ground reference for your heaters is undoubtedly still connected somewhere in the amp if the cathode sinks to 4 volts. How was it referenced before? Do you have a CT on the heater windings at the PT? Did you just connect one leg of the heaters to the cathode? You have to find and disconnect the old heater reference point in order to proceed.

3) I've noted that my heater noise is coming from V1 (2 stages of preamp - still not settled on 12ax7 or 12ay 7) is a low, slowwwww pulsating 60hz hum, not very loud either. I pull the tube and the amp is virtually dead quiet. This amp is to be mated with a 10" alnico and small baffle/open backed cabinet. I'm wondering about dropping the value of the output section cathode bypass capacitors. And that brings up the question of the effect of the individual bypass caps in the output section and how to adjust their value for response eq. Would dropping to 10uf each roll off some lows?

Making any cathode bypass cap smaller will filter out 'more' LF. Treat the cathode R and C as a first-order HPF to see where the knee of the filter is. That kind of engineering is ALWAYS OK but may not get you the results you want (it could kill all the oomph from the amp before it makes the hum disappear). Look first at the power supply filtering. if you have a 450v or 500v electrolytic cap available you can temporarily (and carefully!) parallel it across each PS node cap to see how the hum is affected. With good results, make the change permanent. Also a possibility is that your input stage lead dress could be massaged to reduce EM coupling.

Of course, if the problem goes away with an effective heater DC offset (see point #1), then that's even better!

[Malcolm Irving]

When disconnected from the transformer, the heater chain shows "open" in reference to ground on my multimeter at its highest setting (20M).

Assuming that you have a dedicated heater winding on the power transformer, the heater chain should also read 'open' relative to ground when the heater chain is connected to the transformer (everything powered off and drained, of course).

[brilliantblue]

Actually now that I think about it, I'd only checked the wirito ensure there wasn't a short in the heater wiring. I'll check in reference to ground tonight. Thanks for waking me up.

[brilliantblue]

Thanks for the post - you helped me revisit and recognize an important detail that I'd missed - the heater winding and the HV winding share a center tap - I'd wrongly assumed the ct was only for the high side.

Soooo - does that, then, rule out dc elevation?

Transformer is from hammond ao-44 reverb amp.

Yes, both cathode voltages were 15v

Thanks for your help!

[Malcolm Irving]

- does that, then, rule out dc elevation?

Unusual arrangement for a PT. I think it does rule out heater elevation, as the shared centre tap will have to carry the power supply return current from the HT.

[Jazz P Bass]

Hammond AO-44 Schematic

Thanks for the post - you helped me revisit and recognize an important detail that I'd missed - the heater winding and the HV winding share a center tap - I'd wrongly assumed the ct was only for the high side.

I do not believe that the heater winding should share the center tap of the high voltage winding.
There should not be a connection from the heater winding to ground.
It floats.
Unless you have the 100 ohm balancing resistors.
Attached AO-44 schematic.

[brilliantblue]

I've not even listened to it without an artificial ct or dc elevation at its current design. Removing the artifical ct may kill the hum as I've apparently created a ground loop - as I didn't use the combined ct grounding point.

Sorry for the runaround! Beware Hammond AO-44 Transformers! Thanks for the help with the resistors/caps!

[Jazz P Bass]

Hammond AO-44 lLater Schematic

Well it appears that there is more than one schematic.
Does your amp have the vacuum tube rectifier?
If so, this schematic does tie both CT's to chassis ground.
Does your heater winding have three green wires?

[brilliantblue]

Yes, originally tube rectified.

[eschertron]

I don't know how you feel about popping the PT's end bells off and digging around for the CT connection... could be easy to find and 'snip', could be catastrophic

I'm always reminded to "don't fix what ain't broken". There are multiple possible cures to the hum that you are pursuing right now, if one cure becomes exceedingly hard then it's OK to step away from it.

[brilliantblue]

I don't know how you feel about popping the PT's end bells off and digging around for the CT connection... could be easy to find and 'snip', could be catastrophic

I'm always reminded to "don't fix what ain't broken". There are multiple possible cures to the hum that you are pursuing right now, if one cure becomes exceedingly hard then it's OK to step away from it.

I'll disconnect the aforementioned ground-loop when I get home tonight and take it from there. It wasn't a troubling hum to begin with - just a late start on "how quiet can I get". I've used dc elevation in the past so was just "programmed" to wire up resistors to the filament and take them to the cathode after not "seeing" a heater ct.

Walking away is great advice. I'm learning that better with this build as I don't keep as electronically active as I used to. You build a few amps for yourself and then you've got plenty of amps! This one is for a friend.
Thanks again! Jeff

[brilliantblue]

Well it appears that there is more than one schematic.
Does your amp have the vacuum tube rectifier?
If so, this schematic does tie both CT's to chassis ground.
Does your heater winding have three green wires?

Sorry - I missed a detail in here that may benefit someone else to know - the combined CT wire is white!

I removed the artificial CT-to-ground connection - the pulsing stopped, but I still have a low volume 60hz - quite possible from an untwisted wire reaching around the socket to pin 9 on v1 (12a*7)