cold heaters

[merlinb]

Merlin, what is the vertical scale on these plots?
If your "10mA" is per-division, you show (for your 9.7V-heat graph) 55mA, which seems impossible.

It's 2mA/div, and its a 12AU7, not a 12AX7

Your curves show no strong sign of flat-topping. Maybe a wee bit at 6.3V.

I'd say saturation was pretty obvious at 6.3V. It flat tops at a little over 20mA somewhere off screen. I chose that value exactly because it starts to exhibit saturation while occupying the whole screen with the same axes as previously.

[slacker]

Is the saturation where the traces start to curve over to the right? Where it stops being linear?

[samhay]

Thanks for running the 9V data Merlin - looks like it knocks the input resistance of the grid down a fair bit. I assume this holds right down to starved plate conditions?

[merlinb]

Is the saturation where the traces start to curve over to the right? Where it stops being linear?

Yes, eventually they will all level out (off screen) which is total saturation, something like this:

Thanks for running the 9V data Merlin - looks like it knocks the input resistance of the grid down a fair bit. I assume this holds right down to starved plate conditions?

Um, not sure what you're getting at here. With 9V on the heater I wouldn't expect the input resistance of the grid to be much different from normal...

[samhay]

The I/V curves seem to be much steeper in the low V region of the 9V vs 12.6V curves. At least with the lower grid biases.
Is the grid input resistance not the reciprocal of the I/V slope?

[merlinb]

Is the grid input resistance not the reciprocal of the I/V slope?

No no! The axes are anode voltage and anode current, so the slope is the input resistance of the anode, not the grid. In otherwords, the slope is the internal anode resistance.

[samhay]

Of course - sorry, another case of not paying enough attention.
So the anode resistance is as much as halved at 9V vs 12.6V heater. That seems fairly significant.

[merlinb]

So the anode resistance is as much as halved at 9V vs 12.6V heater. That seems fairly significant.

Still not sure what you're seeing... The 9V curves are only slightly less steep than the proper 12.6V curves at small bias voltages. This sample actually seem to be a bit hotter than the published characteristics at larger bias voltages, although my curve tracer is also marginally innaccurate.

[samhay]

Nevermind. Sorry Merlin, I'm not having a good day - I mistook the 6.3V for 12.6V data.
So I think I've got it now - things get a little funky at half the heater supply, but 9V is probably ok(ish).