Grid current

[lowell]

Am I right in that your typical AC coupled cathode biased 12ax7 preamp stage, with Vk = 1.5v, will begin grid conduction and grid current limiting (and blocking distortion), when a grid signal of 1.5v Peak or higher is applied? I know it's a hair above 1.5vPeak but we can ignore that for sake of discussion. But I think when Vgk reaches 0v or higher the grid will begin conduction. Just making sure I have this right.

[uneumann]

Am I right in that your typical AC coupled cathode biased 12ax7 preamp stage, with Vk = 1.5v, will begin grid conduction and grid current limiting (and blocking distortion), when a grid signal of 1.5v Peak or higher is applied? I know it's a hair above 1.5vPeak but we can ignore that for sake of discussion. But I think when Vgk reaches 0v or higher the grid will begin conduction. Just making sure I have this right.

That's basically correct... Yes, grid current begins to flow in the region of Vgk=0. That happens regardless of the bias method, but the grid behavior varies slightly from tube to tube, and with tube type (12ax7 vs 12at7, for example). As for blocking and limiting that depends on many other factors - mainly source impedance, coupling capacitance, degree of overdrive, etc.

[g1]

Aren't those both the same? 1.5V peak with respect to ground, or 0V with respect to cathode (Vgk).
Not sure if that is your question.

[lowell]

Yes they're the same. Was just trying to be clear in regards to AC signal applied. Thanks guys.

[66 Kicks]

Am I right in that your typical AC coupled cathode biased 12ax7 preamp stage, with Vk = 1.5v, will begin grid conduction . . . when a grid signal of 1.5v Peak or higher is applied?

No.

Unlike a pentode, there is nothing magical about 0Vgk for a 12AX7. With your typical setup, grid current starts at a negative Vgk and increases exponentially with increasing Vgk. The amount of grid current at any given Vgk is dependent upon the plate voltage. The amount of grid current at 0Vgk at one plate voltage can also be achieved at a negative Vgk by deceasing the plate voltage.

[tubeswell]

... a grid signal of 1.5v Peak or higher is applied?...

You mean 3V peak to peak?

[mhuss]

As 66 pointed out, it's not an on/off switch or sudden change like a Zener diode, but a gradual change, as shown by the tube curves. Grid current will start to flow when the grid gets N volts higher than the current cathode voltage, in a manner similar to a tube diode. Like a tube diode, there is a certain positive potential between (positive tube element) and cathode that needs to be there before the grid will start to conduct. It will not conduct significantly when Vg = Vk.

[66 Kicks]

I kinda think that grid current is going to exist as long as the grid is positive in relation to ground. At what point you are going to consider it significant depends on your application and your tolerance.

[Enzo]

And it might serve to calculate the amount of current. It isn't just a G and a K, it is resistive paths as well.

[lowell]

You mean 3V peak to peak?

Yes

[mhuss]

Another factor - is the hypothetical cathode bypassed? If not, as the grid gets more positive, so will the cathode, as more current flows through the tube. If it is bypassed, the grid can move faster than the cathode, so grid conduction is easier to accomplish. IME the grid voltage does not have to get much more positive than the cathode voltage for the conduction to start in earnest.

[lowell]

Yes I had learned in MerlinB's awesome book that grid conduction cause rather edgy clipping which must mean grid conduction onset is more sudden and sharp edged than subtle. I have done scope tests myself on this but couldn't get conclusive results. I remember pulling my hair out because it seemd the grid wasnt clipping even a volt or more above cathode voltage. Never did figure out why or if it was operator error. So all I really have to go by is Merlins book which has tons of solid data on the subject with scope images to boot. Thought posting this would clear it up as I lost my book years ago...but think my memory serves.

[loudthud]

If you drive the grid with a 50 Ohm signal generator direct with no stopper, you won't see much clipping. If you have any kind of reasonable stopper like 10K or more, the clipping will be quite sharp.

[eschertron]

typical AC coupled cathode biased 12ax7 preamp stage

In a typical preamp circuit, you won't see the grid rise above cathode voltage because the mA of current (uA?) available are consumed quite readily. Should be easy to spot in a properly set up test rig (or in a representative amp channel).

A warm-biased preamp triode has a distinctive sound when grid-limiting kicks in. It is typically a soft-clipping and it generates a bunch of 2nd harmonic distortion. So not only is the bias 'warm', some people refer to the distortion as 'warm-sounding'.

As LT suggests, enough drive can overcome grid limiting. I'd think too much of that would overheat a 12AX7?

[lowell]

I thought grid stoppers REDUCE the effect of grid current. ? But LT you're saying the opposite. I was using a scope and was actually using a pot as a stopper to see the effect...if remembering correctly. Guess I should try again with solid variables. I was also using a Lodestar oscillator from Tubes and More for signal.