Express heater elevation.

[jem]

For an Express, is there any reason why using the bleeder resistors to elevate the heater voltage shouldn't be used? See attached image, should give about 43 volts to refernce heaters.

Just looking for ways to try to tame the Express with minimal altering the audio path, gain structure.

Also, I read somewhere (can't remember now) where the goal for heater elevation was to match the heater reference with the cathode reference. So question is, if this is the case (and I probably have that all wrong), could you tie the heater common to the cathode resistor on the first preamp tube and track it exactly?

[rhinson]

For an Express, is there any reason why using the bleeder resistors to elevate the heater voltage shouldn't be used? See attached image, should give about 43 volts to refernce heaters.

Just looking for ways to try to tame the Express with minimal altering the audio path, gain structure.

Also, I read somewhere (can't remember now) where the goal for heater elevation was to match the heater reference with the cathode reference. So question is, if this is the case (and I probably have that all wrong), could you tie the heater common to the cathode resistor on the first preamp tube and track it exactly?

sure you can do it if you think you have a filament hum problem, it helps eliminate that noise, and in the 40-60v range is generally what most guys use to float dc on the filaments. by "tame" you must have meant taming some noise issues. except i would take the voltage off the screen supply instead of the plate supply, giving it one more section of filtering and ripple reduction. rh

[jem]

i would take the voltage off the screen supply instead of the plate supply, giving it one more section of filtering and ripple reduction. rh

I ended up using the circuit I drew; I didn't want to bypass the standby switch, and only wanted to add one part. To get some more filtering, I used a 47uF 100v instead of 22uF.

It did take some of the hum out of it.

[FUCHSAUDIO]

i would take the voltage off the screen supply instead of the plate supply, giving it one more section of filtering and ripple reduction. rh

I ended up using the circuit I drew; I didn't want to bypass the standby switch, and only wanted to add one part. To get some more filtering, I used a 47uF 100v instead of 22uF.

It did take some of the hum out of it.

I suggest 1-M over 3.3-K (no wattage issues there) a small filter (1-UF is fine, you don not want a time constant), and add a hum balance control, as it's simple and easy. You ideally want the cathode and the heater about the same potential, which means 1.5 or 2.0 volts or so. It's "differential" you need to minimize.

[jem]

I suggest 1-M over 3.3-K (no wattage issues there) a small filter (1-UF is fine, you don not want a time constant), and add a hum balance control, as it's simple and easy.

So would it make sense to add this to all of these amps?

You ideally want the cathode and the heater about the same potential, which means 1.5 or 2.0 volts or so. It's "differential" you need to minimize.

So back to the original question. Can the heater common be tied to the cathode? Then there would be absolutely no differential at all times.

[paulruby]

If the cathode voltage is higher than the filament, you can get thermionic from filament to cathode which you'll hear as hum. You won't hear this in anything but a high gain amp since it's a very tiny amount of injected noise and I doubt a wreck suffers from this. I've never heard it in a wreck. The solid state recto and large 1st filter cap is a much bigger problem in a wreck for hum and buzz.

That said, elevating the filament doesn't hurt, so can always be done. The peak voltage swing on the filament is just over 9V, so you want to elevate to something that is at least 9V higher than the cathodes of the early stages, so 11V or so and add some voltage just to make sure... 30 to 50V is all good, too.

[lastwinj]

I suggest 1-M over 3.3-K (

going to be dropping more voltage across that 1m that it is designed for unless you use a 5w.

could be mistaken, though. 350v is what most 1w metal/carbon film are rated at.

germ

[lastwinj]

for any heater elevation, i use a 220k and 47k 1w metal film, with a 47uf 100v across the 47k works perfect for everything up to about 450v, elevates the heaters nicely.

germ

[FUCHSAUDIO]

First:

Jem, I would not use the cathode to raise the filament voltage. It's asking for trouble, and you want to keep them separate.

As for component wattages, there is little to no dissipation for the 1-M nor 3.3-K, however it is assumed a high quality resistor able to handle the voltages present would be prudent.

For that matter, you could replace the 3-3K with a 5-K pot, reduce the 1-M to 500-K and set the pot as a variable resistor to adjust the lift voltage to where you want it. In some amps (depending on grounding) there may be a "null" on both the hum balance and lift pots. Honestly, the 40-60 volt number seems high to me, especially when the cathodes on most of the preamp tubes is maybe 2 volts or so above ground.

I agree somewhat with Paul Ruby (hey Paul), but (IMHO) to wring every last ounce of hum out of any tube amp, lifting and balancing is not a bad idea. It's obviously most manifested in high gain amps, which is why I use DC filaments in that case anyway. For the cost of the parts, I say "why not" ?

[rhinson]

I suggest 1-M over 3.3-K (no wattage issues there) a small filter (1-UF is fine, you don not want a time constant), and add a hum balance control, as it's simple and easy.

So would it make sense to add this to all of these amps?

You ideally want the cathode and the heater about the same potential, which means 1.5 or 2.0 volts or so. It's "differential" you need to minimize.

So back to the original question. Can the heater common be tied to the cathode? Then there would be absolutely no differential at all times.

you can tie the heater ct point to the power tube cathode of cathode biased amps and float that dc on them-----lots of vintage cathode biased amps did this---ampegs esp. the express is fixed bias so that doesn't apply. and with most el84 cathode biased amps it doesn't really apply either because there's not enough voltage there (ususally in the 10-12v range) to make any difference. with some 6l6 or el34 cathode biased amps (depending on operating voltage) you can get in the 25 - 35 v range which makes it worth a shot in those type of amps. rh

[jem]

So back to the original question. Can the heater common be tied to the cathode? Then there would be absolutely no differential at all times.

you can tie the heater ct point to the power tube cathode of cathode biased amps and float that dc on them-----lots of vintage cathode biased amps did this---ampegs esp. the express is fixed bias so that doesn't apply. and with most el84 cathode biased amps it doesn't really apply either because there's not enough voltage there (ususally in the 10-12v range) to make any difference. with some 6l6 or el34 cathode biased amps (depending on operating voltage) you can get in the 25 - 35 v range which makes it worth a shot in those type of amps. rh

How about tying it to the cathode of the first preamp tube where it's most sensitive? Fuchs is saying 1.5 to 2.0 volts is all that's needed, but you're saying 25-35 is needed. I'm confused here??

[FUCHSAUDIO]

The correct voltage is the one where the amp is most quiet. No correct voltage really. I forgot about the power tube cathode lift circuit. That's a source of immaculate clean DC for free ! One HH Scott tube amp used 12AX7 filaments as the cathode resistor I believe, which is really slick.... You could easily tie a hum balance pot to the power tube cathodes (cathode biased amp only), and get a free lift with basically no additional parts to speak of.

I always felt it was "differential" you were trying to eliminate, so you'd want the filament close-to the average preamp cathode voltage in the amp, but I have seen some lifted 30 to 50 volts over ground.

[lastwinj]

Honestly, the 40-60 volt number seems high to me, especially when the cathodes on most of the preamp tubes is maybe 2 volts or so above ground.

look at directly coupled cathode followers. alot of them run well over 90v on the cathode. this exceeds the heater cathode breakdown voltage in 12ax7 type tubes. going up to 90v for heater elevation can help increase tube longevity.

example, look at the dynaco st70 and some of the monoblocks using 7199. almost every 7199 used by them had a 100% failure rate due to heater cathode V issues.

I agree somewhat with Paul Ruby (hey Paul), but (IMHO) to wring every last ounce of hum out of any tube amp, lifting and balancing is not a bad idea. It's obviously most manifested in high gain amps, which is why I use DC filaments in that case anyway. For the cost of the parts, I say "why not" ?

AMEN!!!!!

[FUCHSAUDIO]

In the Trainwreck style circuit, the one and only cathode that "high in the air" is the PI, and it's not a noise concern at that point in the circuit.

Cathode followers (especially DC coupled ones), are indeed pushing the heater cathode differential quite a bit, and should have an elevated heater reference. But that's not what we're dealing with here....it's mainly noise reduction that we're trying to achieve.

Not to belabor the point, but we're talking two different issues here imho.

I do want to point out that the Tung-Sol reissues seem quite prone to this failure. We saw a high failure rate in the FX loop and Driver locations in our amps, while the Ruby 12AX7's worked fine for years. This seems isolated to that one type and brand btw.

[d95err]

The schematic posted at the top of the thread references the heaters using the first B+ node, directly after the rectifier. This is a very noisy voltage, so it seems that it would introduce a lot of 100Hz rectifier noise into the heaters.

It seems like a better idea to use the screen or splitter voltage as a reference as it is much better filtered.