Grid stoppers, clamping, blocking and crossover distortion.

JJman

Member
Messages
994
In my homebrew I’m using a LTPI to drive 2- 6v6 tubes in PP. They are cathode biased running AB at 12watts idle each. The plates are at 350VDC. I’m wondering about the grid stoppers and if they are adequate at 1.5k. I sometimes think I’m hearing a bad distortion in addition to the good distortion. No doubt the PI is driving them very hard. The PI doesn’t clip on it’s own until after the output is clipping. I recently added NFB but I read that this can actually increase the chance of “blocking distortion.”

http://www.aikenamps.com/BlockingDistortion.html

http://www.freewebs.com/valvewizard/gridstopper.html

I definitely see crossover distortion on the speaker output when driven past the onset of the output clipping, but is that normal? (I once read that it is.) I understand the “clamping” that happens but I’m not clear on whether the grid stoppers curtail clamping. Do they?

I have a signal generator and scope. Can blocking distortion be recognized on a scope? Is the crossover distortion, that’s caused by extreme clamping, the same thing as blocking distortion?

I could throw 10k resistors on the grids as a test but I’d like to fully understand 1st.
 

MKB

Silver Supporting Member
Messages
9,316
If you are seeing (and hearing) too much crossover distortion in your distorted output, look at your power tube and screens B+ and make sure it isn't sagging too much under load. What sometimes happens is under load your B+ drops and the power tubes basically become unbiased. But the bias is then correct after the signal goes away and the B+ returns.

I can't recall what power tube blocking distortion looks like, but know it instantly when I hear it; your tone gets very muddy and the attack goes away. It sounds very bad. I wouldn't be surprised if it would be hard to see blocking distortion on a scope as it is more of a dynamic signal issue and you would have to have a pulsed input into the amp to see it. It probably wouldn't be too hard to set up such a test but I haven't tried it yet.

There is a Mesa patent where he uses rather large grid stoppers (220k) and 2.2meg resistors from the grids to ground that seems to help this quite a bit, especially in EL84's. I can't recall the patent number, but he uses them in the Studio .22:

http://www.schematicheaven.com/boogieamps/boogie_22cal.pdf

You might try those. I've had issues cleaning up EL84's where they sound good overdriven, and those tricks in the Studio .22 help a lot. I haven't looked very much into why the mods work so well, but I assume it keeps the grids from being driven positive, which seems easy to do in EL84's. Hope this helps you.
 

xk49w

Member
Messages
222
You may notice that flat topping of the driver plate voltage occurs at
the same point as the output starts to flat top and that this plate
voltage is well below the plate supply voltage. The driver is being
driven into cutoff but the voltage at the plate is being clamped by
the forward conducting output tube's grid. This is an AC coupled
circuit, usually a 0.1uF or something. Under a no signal conditon the
voltage across the coupling cap might be 200V. Consider what would
happen if the driver were subsequently driven into cutoff and kept in
the cutoff condition -- the plate voltage on the driver would rise to
equal the plate supply voltage as the coupling cap charged. The
output tube grid would now be at ground potential and not drawing grid
current. Now restore an AC signal to the driver. As the driver
started to conduct and as the driver's plate voltage decreased the
voltage on the output tube's grid would be very negatively biased
until the coupling cap discharged to its steady state condition. The
output tube would be cutoff until the coupling cap recovered.

You can see that the frequency of the input signal can determine
whether this occurs and to what degree. A low frequency signal will
keep the driver cutoff for a longer time than a high freqency signal,
so there would be more time for the cap to charge. Depending on
component values and the amount of overdrive of the driver, crossover
distortion is a possibility.

(I have not tested this but) one way you might detect the presence of
blocking distortion is to feed the driver with a low frequency signal
into cutoff. During the flat topping phase you should see the plate
voltage rise as the cap charges.

The other point by the OP regarding NFB, when the output stage clips
the NFB loop loses control. The loop difference amp (usually PI), will
drive the output stage even harder during this time, providing more
opportunity for minor bias shifts due to the AC coupled drive signal.
 
Last edited:




Top Bottom