Grid current characteristics

[Alan0354]

I am reading page 20 to 22 of chapter 2 in RDH4, it talks about the grid current. Is this important to learn? I am confused with the term "Grid Blocking" in Fig 2-11 in page 21. It talks about the grid current when it is driven +ve and the current change direction due to different reasons. I don't under why it said if the driving impedance is 100K, the load line intersect at A and B. It said when the grid voltage reach a little over 40V, it will just to over 80V and latched up until it return back down!!! I don't really get what the book said. Can anyone explain this?

The book said if the driving impedance is less than 40K, the load line don't intersect the loop, then it will not latch up. Is that the reason some of the Fender amps has 47K as the plate resistor in the differential stage( phase splitter) so the grid of the power tube see 47K?

Thanks

Alan

[Jazz P Bass]

In a nutshell, grid current should be avoided.

[exclamationmark]

When the grid of the tube goes positive it's impedance drops from several megaohms to perhaps only a few kilo-ohms. Thus the coupling capacitor discharges into the tubes grid fairly rapidly, and bias will be upset momentarily (this is determined by source/load impedances which affects the RC constant of the whole mess). Mostly occurs on bass notes where the RC constant is higher than the frequency the tube is attempting to amplify. You can avoid it easily on pre-amp tubes by sticking a huge 100k+ grid stopper on the grid so the coupling cap can't discharge (470k is the norm on high gain amps). A little more problematic with power tubes because they have a maximum impedance from grid to ground value (thermal runaway and all that...)

[Alan0354]

Actually from practical experience, grid stopping resistor is a must for cascade gain front end. The overdrive sound sounds so awful without it. When I looked at the wave form on the scope, it looked like pulses with low duty cycle due to the charging of the coupling cap.

But my question is why the latchup that the book described. In my observation, the grid conduct so much that it looks like the input has a clipping diode on it. You cannot drive pass the clipping unless you have low impedance drive like a cathode follower.........or a transistor.

[exclamationmark]

I just looked at the RDH4 book and it's evidentially not talking about blocking distortion. What I THINK it's talking about (I'm most probably wrong ), is the effects of secondary emission. A similar thing occurs in power valves where (usually in leaky tubes) as the electrons go about their business from plate to cathode, they dislodge positive ions from collisions with other particles (air!), which are immediately attracted to the most negative thing in the tube, which is more than likely the grid. This increases the grids voltage (as the reverse grid current develops a voltage across the grid leak resistor - that's why some tubes have maximum grid-ground resistances specified), which in turn increases plate current, which increases positive ion generation.... and so on. Perhaps the RDH4 book is talking about the situation where the plate current flowing is so great at positive grid potential, that it simply reaches a point of equilibrium, with the grid leak resistor as the load line.

[Gingertube]

I started to write a grid current "essay" but realised that I dont have the time right at this moment and would end up giving bad advice - so introductory "summary" ONLY.

Like many (I assume) I struggled with the grid current concept until I got as far as this in my understanding:

There are 3 types of grid current:
1) Grid Current to neutralise a negative charge build up on the grid due to space charge (electron cloud) coupling from the cathode to the grid. The voltage developed across Rg1 (grid leak resistor) effectively adds to the tube bias and is the basis for grid leak bias etc.
2) Grid Current to neutralise a positive charge build up on the grid due to positive ions collecting at the grid. As electrons accelerate "up" the tube from cathode to anode, some of them collide with residual gas atoms, stripping an outer orbit electron and turning them into positive ions, which then accelerate back "down" the tube and some of them ending up at the grid. This is a problem with old gassy tubes and tubes with high currents (such as output tubes). The voltage developed across Rg1 effectively subtracts from the bias and is why you see different maximum Rg1 values for power tubes with fixed bias (lower max Rg1) vs cathode bias (higher max Rg1 value since the cathode bias will partially oppose the bias shift). This is the root cause of "Run away" in power tubes which is reasonably common in guitar amps because guitar amp designers regularly abuse the max Rg1 values and run the tubes hard. It is also the reson why some guys have trouble with grid leak bias when using older "gassy" preamp tubes.
3) Grid rectification current from driving the grid positive with respect to the cathode. This is the root cause of blocking distortion.

RDH is talking about mostly about the second of those above which is why they give the "Rule of Thumb", for high mu preamp tubes (where bias drift is more of a problem) the grid leak resistor should be no more than 3 times the anode load resistor for cathode bias and no more than twice the anode load resistor for fixed bias.

The problem is that different sources/references use different terms for these grid currents so you need to decide from the context of the article which of these they are talking about.

I tend to use these terms ( I note that Merlin B uses these same terms in his Preamp Boook):
POSITIVE Grid Current
NEGATIVE Grid Current
GRID RECTIFICATION Current

Cheers,
Ian