The 12AX7 is not well known for its current drive capabilities but it's used ubiquitously in modern guitar amp cathode followers.
So my question is... generally speaking, can one half of a 12AX7 (noval) or 6SC7 (octal) provide sufficient 'positive' grid current to drive the second half well into the saturation region in a DC coupled cathode follower configuration, or does such a configuration only signal clip due to its lack of current drive when trying to push the second half deep into positive grid territory, but never actually making it to saturation? Both situations will generate significant distortion but via different mechanisms.
So my question is... generally speaking, can one half of a 12AX7 (noval) or 6SC7 (octal) provide sufficient 'positive' grid current to drive the second half well into the saturation region in a DC coupled cathode follower configuration, or does such a configuration only signal clip due to its lack of current drive when trying to push the second half deep into positive grid territory, but never actually making it to saturation? Both situations will generate significant distortion but via different mechanisms.
I think the answer to your question is that the anode current of a valve is almost always sufficiently larger than the grid current of that same valve type so there will not be a problem with supplying grid current if that is what you want to do. I don't think there is any saturation in this circuit; clipping is due to grid current loading.
The 12AX7 generally makes a poor cathode follower, but if a poor cathode follower will suffice (or is preferred for some reason) for your application then a 12AX7 can be used.
The 12AX7 generally makes a poor cathode follower, but if a poor cathode follower will suffice (or is preferred for some reason) for your application then a 12AX7 can be used.
Thanks DF - I might have mislead by using the term saturation (i.e. cathode thermal emission limitiation). Perhaps I should have used 'space-charge limited' region. I'm talking about the ~10v Anode and lower region in this graph, showing positive grid current (and voltage).
.
It suggests 5ma at 8 volts (above V2 cathode) might be required into the 12AX7 V2 grid on a DC cathode follower to achieve space-charge limiting for a load line with a gentle transition into positive grid distortion.
If V1 is unable to produce sufficient current to reach the space-charge limited region, the the signal on the grid will instead be limited by 'voltage division' - as the output impedance of the V1 source and the (low) input impedance of the grid under positive grid voltage creates a potential divider.
Another item that is not yet clear to me is the impact of the (large) cathode resistance on the V2 cathode follower and its impact on grid current flowing in V2, especially that required to attain space-charge limiting.
Notwithstanding, your answer might be the same.
.
It suggests 5ma at 8 volts (above V2 cathode) might be required into the 12AX7 V2 grid on a DC cathode follower to achieve space-charge limiting for a load line with a gentle transition into positive grid distortion.
If V1 is unable to produce sufficient current to reach the space-charge limited region, the the signal on the grid will instead be limited by 'voltage division' - as the output impedance of the V1 source and the (low) input impedance of the grid under positive grid voltage creates a potential divider.
Another item that is not yet clear to me is the impact of the (large) cathode resistance on the V2 cathode follower and its impact on grid current flowing in V2, especially that required to attain space-charge limiting.
Notwithstanding, your answer might be the same.
I am unclear what you are asking. The space charge limited region is the normal region of valve operation i.e. not thermally limited. Almost all valves in almost all audio applications operate in this region, however large or small is the current or voltage. My guess is that a 12AX7 only approaches the thermally limited region at currents of many tens of mA - unless you have reduced the heater voltage.
In the circuit you linked, distortion arises because the grid current of V2 loads the anode circuit of V1. As far as I can see, nothing to do with saturation (not happening) or the space charge limited region (always happening).
In the circuit you linked, distortion arises because the grid current of V2 loads the anode circuit of V1. As far as I can see, nothing to do with saturation (not happening) or the space charge limited region (always happening).
I'll take another shot at describing it...
V2 loading the anode of V1 is one method of distortion - the 'potential divider' method mentioned above. If V1 has very little current drive ability, it is unable to push V2's grid positive (where V2's input impedance drops dramatially) and a 'sharp' distortion or clipping is presented. i.e. distortion is down to the inability of V1 to drive V2.
The second method and the one I'm interested in, is where distortion is not due to V2 loading; because in this case V1 has the ability to supply sufficient current into V2's grid to maintain the positive grid voltage (whatever that might be). i.e V1 is behaving as a perfect voltage source. In this second case the distortion happens because V2 is no longer linearly responding (increasing anode current) to further increases in grid voltage from V1. In the grid curves shown this is where the curves start to bunch and eventually overlap in the low anode voltage area of positive grid voltage. I don't know how to best describe this area of distortion (perhaps there is a name for it?) but it does not appear to me as grid-current limited distortion and neither is it classic saturation (i.e. thermal limited).
From the graph it can take several mA for V1 to drive V2 into this 'non-linear' area - depending on where the load line falls... and where the load line falls determines the 'sharpness' of the clipping in this method.
V2 loading the anode of V1 is one method of distortion - the 'potential divider' method mentioned above. If V1 has very little current drive ability, it is unable to push V2's grid positive (where V2's input impedance drops dramatially) and a 'sharp' distortion or clipping is presented. i.e. distortion is down to the inability of V1 to drive V2.
The second method and the one I'm interested in, is where distortion is not due to V2 loading; because in this case V1 has the ability to supply sufficient current into V2's grid to maintain the positive grid voltage (whatever that might be). i.e V1 is behaving as a perfect voltage source. In this second case the distortion happens because V2 is no longer linearly responding (increasing anode current) to further increases in grid voltage from V1. In the grid curves shown this is where the curves start to bunch and eventually overlap in the low anode voltage area of positive grid voltage. I don't know how to best describe this area of distortion (perhaps there is a name for it?) but it does not appear to me as grid-current limited distortion and neither is it classic saturation (i.e. thermal limited).
From the graph it can take several mA for V1 to drive V2 into this 'non-linear' area - depending on where the load line falls... and where the load line falls determines the 'sharpness' of the clipping in this method.
I think what is happening there is just that the grid is stealing more of the cathode current, so the anode current is smaller than it otherwise would be.
Yes there is a replacement of grid current for anode current but (on paper) it also enables access to the 'bunching up', non-linear operational region of the positive grid curves.. similar but opposite to as to what we see on highly negative grid voltages (cut off). By definition, simple grid current limiting cannot drive the grid positive enough to get into this operataional region. Instead grid current limiting creates a variable voltage divider at the grid of V2 via the low impedance diode effect of its grid under positive voltage and high output impedance of the V1 source.
What I'm looking to do here is to force the tube to operate in that gradually bunching up operational region but significant grid current is required to do this and I'm not sure if a 12AX7, configured as a simple DC cathode follower, has the 'welly' to get there - hence the question... plus I'm still not sure what to call this area of operation / distortion.
Figured that someone else out there has already 'been there, done that' with DC followers, so could advise. However there is always opportunity for time at the breadboard tying to reinvent the wheel.
What I'm looking to do here is to force the tube to operate in that gradually bunching up operational region but significant grid current is required to do this and I'm not sure if a 12AX7, configured as a simple DC cathode follower, has the 'welly' to get there - hence the question... plus I'm still not sure what to call this area of operation / distortion.
Figured that someone else out there has already 'been there, done that' with DC followers, so could advise. However there is always opportunity for time at the breadboard tying to reinvent the wheel.
I think this is what is happening. However, the clipping of the output voltage is not ‘sharp’. It is rather smooth, because the grid current into V2 departs through the 100k cathode load, increasing the output voltage.
In an experimental test, a small grid-stopper (say 100 ohms) can be introduced between the V1 anode and the V2 grid so that the grid current can be measured.
In practice, the amount of grid current in this circuit varies widely among different samples of 12AX7 or ECC83.
- Status
- Not open for further replies.