Elevated heater worry: loads circuit

[Roe]

I want to try elevated heater in some of my low-medium gain guitar amps but have a worry:

Won't the elevated heaters load down the preamp and phase inverter, by changing the ground reference?

Say you add a 390k/56k voltage divider. This will load the circuit with 446k ohms in addition to the ground reference of the original circuit. But this can in some cases load down the preamp somewhat - can't it?

[martin manning]

Won't the elevated heaters load down the preamp and phase inverter, by changing the ground reference?
Say you add a 390k/56k voltage divider. This will load the circuit with 446k ohms in addition to the ground reference of the original circuit. But this can in some cases load down the preamp somewhat - can't it?

I don't think it's a problem. With a 446k load on the screen node say, at 450V, there will be just 1 mA added to the current draw from the rail.

[sluckey]

Sure it puts an additional load on a B+ node, but it is a very light load. For example, a 446K only loads a 300v node with .67mA. Not to worry.

[Roe]

I don't think it's a problem. With a 446k load on the screen node say, at 450V, there will be just 1 mA added to the current draw from the rail.

Sure it puts an additional load on a B+ node, but it is a very light load. For example, a 446K only loads a 300v node with .67mA. Not to worry.

yes, the lower consumptions seems negligible in most cases. but an extra 446k load on a low gain preamp could reduce the gain right?

If the load on the first gainstage is 1m to ground, an extra load of 446k seeems significant

[martin manning]

Put the divider on the screen node. That additional 1 mA will never be missed.

[turbofeedus]

Put the divider on the screen node. That additional 1 mA will never be missed.

+1, have done exactly this on a couple builds, albeit with a 500K/100K divider. No problems whatsoever.

[Mark]

What about having the voltage divider off the main node supplying the output transformer?

In a Vox AC-30, the physically idea spot is near the valve rectifier.

[pdf64]

I don't think it's a problem. With a 446k load on the screen node say, at 450V, there will be just 1 mA added to the current draw from the rail.

Sure it puts an additional load on a B+ node, but it is a very light load. For example, a 446K only loads a 300v node with .67mA. Not to worry.

yes, the lower consumptions seems negligible in most cases. but an extra 446k load on a low gain preamp could reduce the gain right?

If the load on the first gainstage is 1m to ground, an extra load of 446k seeems significant

But heater elevation load is directly on a HT node, not a preamp anode.
Hence from the perspective of the signal, elevating the heaters is immaterial, transparent.

[roberto]

an extra 446k load on a low gain preamp could reduce the gain right?
If the load on the first gainstage is 1m to ground, an extra load of 446k seeems significant

You are confusing the load on the power supply node with the load of the gain stage.
Elevated heaters' voltage divider won't be connected to the coupling cap of the gain stage, but to the (screen or main) voltage supply node.

[Roe]

an extra 446k load on a low gain preamp could reduce the gain right?
If the load on the first gainstage is 1m to ground, an extra load of 446k seeems significant

You are confusing the load on the power supply node with the load of the gain stage.
Elevated heaters' voltage divider won't be connected to the coupling cap of the gain stage, but to the (screen or main) voltage supply node.

well, I was worried that loading a power supply node with e.g. 100k (or 56+56k as in most 100w and 200w marshalls) may affect the gain stage considerably. Consider a gainstage with a 1m to ground powered by 100k plate resistor and a node with 100k to ground

[sluckey]

I would use the screen node. There can be several volts of ripple on the plate node. The screen node is much smoother. The load of your voltage divider is nothing to worry about.

[pdf64]

...well, I was worried that loading a power supply node with e.g. 100k (or 56+56k as in most 100w and 200w marshalls) may affect the gain stage considerably. Consider a gainstage with a 1m to ground powered by 100k plate resistor and a node with 100k to ground

Bear in mind that the HT node is ground.
As far as AC is concerned.
That’s what the node decoupling caps do.

[Stevem]

In amps in the gain range like your talking about its the first gain stage where most all of the unwanted hum levels come from so why don't you you check out something like this to power that first tubes filaments .

Note that if your heater winding already has a center tap you can do away with the 2 100 ohm resistors.

[Roe]

...well, I was worried that loading a power supply node with e.g. 100k (or 56+56k as in most 100w and 200w marshalls) may affect the gain stage considerably. Consider a gainstage with a 1m to ground powered by 100k plate resistor and a node with 100k to ground

Bear in mind that the HT node is ground.
As far as AC is concerned.
That’s what the node decoupling caps do.

Still, a gainstage with 1m to ground differs greatly from one with 100k to ground even if both use a 100k plate resistor.

[roberto]

Hi Roe, you are still confusing the supply node with the node at the plate of the tube.
Please look once more to the schematic. What you said in the last post about loading a gain stage is true, but it doesn't apply to the amount of current a supply node has to guarantee.