Screen and screen resistor failure modes and other concerns?

peteb

Friend of Leo's
Joined
Apr 25, 2003
Posts
4,891
Location
Cascadia
I often read of concern with the failure of a screen or a screen resistor, but I have never understood the risk or the failure modes.

what are the failure modes and what are the risks?

the tungsol spec sheet shows a max plate dissipation of 12 W and a max screen dissipation of 2 W. I don’t see a problem there.


I have never heard or read of an actual screen failure, but I have read about screen grid resistors burning up.

i kind of like this line of reasoning as the screen current is relatively low:

“Fender used carbon comp screen resistors and carbon in that high-heat area is prone to breaking down,same with the grid stopper resistors they used tied from pin 1 to pin 5.
There is really no need for overkill as far as wattage goes,but 3-5 watt is really not needed,as it's the resistor TYPE,not the wattage that is the real issue here.That and the fact Fender put them right on top of the socket where they get really hot.So 5 watt ones should last forever.”
from: https://www.thegearpage.net/board/i...s-in-a-vintage-fender-what-to-expect.1173784/


one thing I do notice about the screen is that when signal on the grid goes positive, and the plate and screen current increase, the signal on the plate is inverted and the plate dissipation would drop as compared to idle, where as the signal voltage on the screen holds steady, with the current going up, increasing screen dissipation above idle level.
 

schmee

Telefied
Silver Supporter
Joined
Jun 2, 2003
Posts
20,231
Location
northwest
My guess on the failure is simply they put CC right on top of the hot tube socket. Had they been metal oxide even at the same wattage I think they would have held up. I bought a BF Bassman once "as is" whose only issue was those screen resistors were broke in half from heat.
What I don't understand is why the amp didn't work...? Earlier amps didn't have those resistors on them at all....
🤔
 

Bendyha

Friend of Leo's
Silver Supporter
Joined
Mar 26, 2014
Posts
3,187
Location
Northern Germany
1652997225860.png

Have a look at this 6V6 graph. Plate voltage along the bottom, plate current on the left, screen current on the right, the marked curves are for different screen voltages - 50V, 100V, 150V......... 300V
As can be seen, the plate curves are the solid lines rising from the 0V / 0mA, up along a straight-ish upward rising diode line, before developing a knee, to curve off to the right and flatten out, showing a relatively constant current to voltage relationship.
The dotted lines, relating to the screen-grid currents shown on the left-hand scale, are, as can be seen, like a horizontal mirror image of the plate lines, starting high before dropping to a more constant, low current region.

As can also be seen in this chart, not only do the screen grid current curves move more and more to the right at the higher voltages, (like one would expect in a guitar amplifier), the knee of the screen grid current curves get continually softer.
If the plate load resistance is high - then in conditions of driving the tubes hard, the screen grid current can rise to dangerous levels.
Have a look at the attached PDF. They explain it well.

Also, If for some reason (supply sag?) the plate voltage momentarily drops below the screen voltage, then dynatron oscillations can happen, with secondary emissions from the plate, and the electrons using the screen to return to ground.

A note in favour of carbon compound screen resistors, is that they are better suited to limiting current surge than other sorts. This is because the load is evenly spread across their whole short-fat mass, rather than just being limited by a thin cross-sectioned band or wire that uses its length to spread out the heat load. Thick-Film / Planar resistors are the best modern solution.
 

Attachments

  • TP-122 - Screen-Grid Current, Loading, and Bleeder Considerations.pdf
    42.1 KB · Views: 11
Last edited:

dan40

Friend of Leo's
Joined
Aug 19, 2015
Posts
2,733
Location
Richmond Va
I have witnessed the screen grid glowing cherry red on a 50 watt Marshall build that did not have screen resistors. Under heavy overdrive, the screen grids would begin to glow until some 1k screen resistors were added. These were some robust NOS tubes so i'd imagine that some of the newer production tubes wouldn't be able to survive this kind of abuse for too long.
 

Paul G.

Friend of Leo's
Joined
Mar 17, 2003
Posts
4,185
Location
Rhode Island
My guess on the failure is simply they put CC right on top of the hot tube socket. Had they been metal oxide even at the same wattage I think they would have held up. I bought a BF Bassman once "as is" whose only issue was those screen resistors were broke in half from heat.
What I don't understand is why the amp didn't work...? Earlier amps didn't have those resistors on them at all....
🤔
The amps with no screen resistors still have the screens connected to B+. Your broken resistors did not pass DC, so no voltage on screens. Amp should made sound, but low volume and distorted without screens.
 

peteb

Friend of Leo's
Joined
Apr 25, 2003
Posts
4,891
Location
Cascadia
What I don't understand is why the amp didn't work...? Earlier amps didn't have those resistors on them at all...
I agree that they are not necessary.

if the resistor shorted, that would be one thing.

if the resistor went open, then it would be a triode and should pass signal.

I am going to look inside my amps. I think I have a couple with stock CC on top of the tube.


I am seeing evidence that the AC signal current is more than the DC current, possibly five times more.
 

2L man

Tele-Afflicted
Joined
Nov 23, 2020
Posts
1,437
Age
62
Location
Finland
Screen resistor(s) value what are installed between B+_ and Screen(s) is one way to tune amp overdrive sound and usually quite easy. Dropper resistor between B+1 and B+2 also has effect but it is not as "dynamic" because B+2 has a filter capacitor.

Screen idle current effect to g1 too which effect to bias and if tube is cathode biased it comes more complicated when bias effect sound too.

I have not measured all power tube Screen peak currents but 6V6GT seem to draw high Sceen current when driven to hard distortion.
 

Bendyha

Friend of Leo's
Silver Supporter
Joined
Mar 26, 2014
Posts
3,187
Location
Northern Germany
Screen resistor(s) value what are installed between B+_ and Screen(s) is one way to tune amp overdrive sound and usually quite easy. Dropper resistor between B+1 and B+2 also has effect but it is not as "dynamic" because B+2 has a filter capacitor.

Screen idle current effect to g1 too which effect to bias and if tube is cathode biased it comes more complicated when bias effect sound too.

I have not measured all power tube Screen peak currents but 6V6GT seem to draw high Sceen current when driven to hard distortion.

Ah yes, the wonderful and complex world of sliding screens.
Bloom.🌊🌈💥🌺
Shifting interactions of the ratios through the supply sag affecting g1/cathode bias, and the voltage and current at all the tubes pins, in different ways.
Complex and interesting stuff. Here though, is not the thread for such serious delights. Oldmantele could explain it well..
 

peteb

Friend of Leo's
Joined
Apr 25, 2003
Posts
4,891
Location
Cascadia
Here is how I mount my screen resistors.

not a pro job, but functional

EDIT: wrong picture, this shows carbon composite control grid resistors on a 6G2, mounted across the side of the tube socket.
 

Attachments

  • F5398B60-7476-4F36-8EA6-AE028F2FCC22.jpeg
    F5398B60-7476-4F36-8EA6-AE028F2FCC22.jpeg
    306.1 KB · Views: 14

peteb

Friend of Leo's
Joined
Apr 25, 2003
Posts
4,891
Location
Cascadia
here are CC screen grid resisters on the tube socket with CC control grid resistors across the center of the tube socket on the Bassman
 

Attachments

  • 47C872FF-3F43-4665-8015-08F2BB7C21EE.jpeg
    47C872FF-3F43-4665-8015-08F2BB7C21EE.jpeg
    243.8 KB · Views: 14
  • C9C242D2-BF07-4EB2-9CED-905F54ADBD62.jpeg
    C9C242D2-BF07-4EB2-9CED-905F54ADBD62.jpeg
    255.5 KB · Views: 15




Top