grid stopper resistors on power tubes

18watt-specific Tech Talk - Building, Fixing, Parts, Mods...

Moderators: CurtissRobin, colossal, zaphod_phil, Daviedawg, Graydon

Post Reply
MrMike
Occasional poster
Occasional poster
Posts: 330
Joined: Tue 12/19/06 2:00 am
Location: Manatuba

grid stopper resistors on power tubes

Post by MrMike »

Why the value of 8.2k on the grid stopper resistors on the power tubes?

Is it to form an RC filter to filter oscillations of some kind? My calcs for a cut-off frequency don't make much sense:

If the total input capacitance of the EL84 is 35pF (got this from Valve Amplifiers by Morgan Jones p453), the freq = 1/(2*pi*R*C) gives a value of 554kHz.

Can anyone please clarify?


thanks
mm
0 x

User avatar
SonnyReVerb
Newbie
Newbie
Posts: 71
Joined: Mon 09/27/04 2:00 am
Location: Indianapolis

Post by SonnyReVerb »

Check out Aiken's explanation:

http://www.aikenamps.com/InputRes.htm
0 x
No, I´m not an amp tech, but I did stay at a...

User avatar
zaphod_phil
Builder, Admin
Builder, Admin
Posts: 15208
Joined: Wed 03/19/03 2:00 am
Location: YYZ

Post by zaphod_phil »

Interestingly the EL84 design guides recommend a minimum value of 10k, but almost all the amps you see use less than that,
0 x
Nature abhors a clean tube amp

Mooreamps
Occasional poster
Occasional poster
Posts: 225
Joined: Wed 02/21/07 2:00 am
Location: Silicon Valley

Post by Mooreamps »

What it means is; one could use the EL-84 in radio circuits up to the IF level, like as a mixer stage in an AM radio reciever.
-g
0 x

MrMike
Occasional poster
Occasional poster
Posts: 330
Joined: Tue 12/19/06 2:00 am
Location: Manatuba

Post by MrMike »

So what Aiken is saying:

(a) used as a filter for high freq oscillations
(b) used as a lo pass filter on inputs to filter radio stations
(c) limit grid current to prevent blocking distortion.

Since the cut-off freq is 554Khz, they 8.2k in the 18W are obviously cannot be for (a)

Since this is a power stage, he says option (b) does not apply.

So are these just to limit current to prevent blocking distortion?
0 x

User avatar
markh
Charter Member
Charter Member
Posts: 1503
Joined: Wed 12/18/02 2:00 am
Location: SE PA, USA
Contact:

Post by markh »

The raw math doesn't seem to support it, but I know from experience that some tubes (e.g. the KT66) sing like a banshee without grid stoppers.

--mark
0 x

MrMike
Occasional poster
Occasional poster
Posts: 330
Joined: Tue 12/19/06 2:00 am
Location: Manatuba

Post by MrMike »

thanks Mark, I have seen the same. Like you say the math does not seem to support it. I just see such wide ranging values with EL84 output stages, it just has me confused. Perhaps a lot of it is "let try this value" and "gosh it works - lets stick with it!".
0 x

Mooreamps
Occasional poster
Occasional poster
Posts: 225
Joined: Wed 02/21/07 2:00 am
Location: Silicon Valley

Post by Mooreamps »

MrMike wrote:So what Aiken is saying:

(a) used as a filter for high freq oscillations
(b) used as a lo pass filter on inputs to filter radio stations
(c) limit grid current to prevent blocking distortion.

Since the cut-off freq is 554Khz, they 8.2k in the 18W are obviously cannot be for (a)

Since this is a power stage, he says option (b) does not apply.

So are these just to limit current to prevent blocking distortion?
Well, for a guitar amp, lets see.

a. yes is used to roll off the high end responce.
b. yes, but this is more for the pre-amp gain stages.
c. yes, absolutly to limit grid current to midigate blocking distortion.
-g
0 x

MrMike
Occasional poster
Occasional poster
Posts: 330
Joined: Tue 12/19/06 2:00 am
Location: Manatuba

Post by MrMike »

Mooreamps wrote:
MrMike wrote:So what Aiken is saying:

(a) used as a filter for high freq oscillations
(b) used as a lo pass filter on inputs to filter radio stations
(c) limit grid current to prevent blocking distortion.

Since the cut-off freq is 554Khz, they 8.2k in the 18W are obviously cannot be for (a)

Since this is a power stage, he says option (b) does not apply.

So are these just to limit current to prevent blocking distortion?
Well, for a guitar amp, lets see.

a. yes is used to roll off the high end responce.
b. yes, but this is more for the pre-amp gain stages.
c. yes, absolutly to limit grid current to midigate blocking distortion.
-g

OK... back to my original question. Then why the value of 8.2k on the power tube grid?
0 x

User avatar
markh
Charter Member
Charter Member
Posts: 1503
Joined: Wed 12/18/02 2:00 am
Location: SE PA, USA
Contact:

Post by markh »

Undoubtedly because the WEM Dominator has 8k2s. Now, why did Watkins choose 8k2? :lol:

I note that Mullard (in "Tube Circuits for Audio Amplifiers") used 4k7 stoppers in their example P-P EL84 amp. To quote Mullard:
"The stopper resistors...have been included as a normal measure to prevent parasitic oscillations. These resistors should not be bypassed, and a close connection should be made to the tags on the valveholders [sockets]."

I suspect these values were all derived experimentally, as opposed to by rigorous equations. ;-)

--mark
0 x

MrMike
Occasional poster
Occasional poster
Posts: 330
Joined: Tue 12/19/06 2:00 am
Location: Manatuba

Post by MrMike »

Thanks Mark, although I appreciate your answer, I was curious as to the response from Mooreamps. Sounds like he has some details I am missing.

mm
0 x

Mooreamps
Occasional poster
Occasional poster
Posts: 225
Joined: Wed 02/21/07 2:00 am
Location: Silicon Valley

Post by Mooreamps »

Well, lets do an electrical analysis. According to the data sheet for an EL-84, I see the input capacitance from grid to cathode at 10.8 pF, but if it's listed as 35 pF, then we can use that value.. and I show the Amplification Factor of the tube at 19.5. The cutoff frequency of the gain stage is calculated when the impedance of the grid circuit equals the impedance of the grid stop resistor, thus half the value of the input signal is dropped across the grid stop resistor and the other half is dropped across the impedance of the input grid circuit, otherwise referred to as the "3 db down point". So to look at this we will touch on some rf theory...

The short answer is ; plug 35pF for the input grid capacitance times the amplification factor [ = 682.5 pF ] into the reactance formula, and we come up with a reactance of 8.338 K ohms at 28 kHz cycles. Therefore, with a 8.2K grid stop for an EL-84, we can calcuate the 3 db down point where the gain of the tube will begin to roll off the high frequency responce just shy of 28 kHz.

-g
0 x

User avatar
SonnyReVerb
Newbie
Newbie
Posts: 71
Joined: Mon 09/27/04 2:00 am
Location: Indianapolis

Post by SonnyReVerb »

Found an excellent treatise on grid stoppers. He discusses power tubes in the lower half of the page (look for "Other reasons for grid-stoppers...").

http://www1.freewebs.com/valvewizard/gridstopper.html

This page shows some of the calculations pretty clearly (look down around Eq. 3, Fig. 10, and Eq.4).

http://www.eecs.umich.edu/~mmccorq/dive ... index.html

Notice the gain has an effect on the calculation. Also, the effective capacitance is the grid to plate (or anode) capacitance.
0 x
No, I´m not an amp tech, but I did stay at a...

User avatar
markh
Charter Member
Charter Member
Posts: 1503
Joined: Wed 12/18/02 2:00 am
Location: SE PA, USA
Contact:

Post by markh »

The problem is (for this EL84 discussion), all the example cited are for triodes. One of the main impetus' for inventing the tetrode and pentode was to eliminate/greatly reduce Miller capacitance -- they do so by having the intervening grid(s) act as a screen/shield (hence the name, perhaps) between the control grid and anode.

I don't know that the stock triode calcs map directly to the EL84 and other greater-than-three-electrodes tubes.

--mark
0 x

User avatar
SonnyReVerb
Newbie
Newbie
Posts: 71
Joined: Mon 09/27/04 2:00 am
Location: Indianapolis

Post by SonnyReVerb »

That's true, Mark. The datasheets show a mere 0.5pF capacitance from grid to plate. Referring to the Valve Wizard page:
While the input of the amplifier will usually have a grid-stopper to remove radio interference, other stages in the amp- particularly the power valves- will often have grid-stoppers for different reasons. The first is that the actual wiring inside the amp will have stray inductances. In combination with the input capacitance of a valve, this will produce a resonant circuit which can cause parasitic oscillation, particularly in high gm valves like power valves. This is cured by damping the resonance with a grid-stopper, fitted directly to the valve socket if possible. Data sheets will usually list a recommended value of grid-stopper, and typical values are around 1k to 10k on power valves. If in doubt, make it bigger.
The Brimar datasheet says:
Due to the high slope of this valve, trouble may be experienced due to parasitic oscillation, and it is advised that a resistor of 10,000 ohms minimum value be connected in series with the grid, at the valve holder grid contact.
Since stray inductance is not easy to measure or calculate ;), I would go with the datasheet recommendation or the schematic - 8.2k sounds reasonable. Experiment to your heart's (or ear's) content...
0 x
No, I´m not an amp tech, but I did stay at a...

Mooreamps
Occasional poster
Occasional poster
Posts: 225
Joined: Wed 02/21/07 2:00 am
Location: Silicon Valley

Post by Mooreamps »

My calculations were based for the EL-84. Maybe it's not clear to me the total grid to cathode capacitance, thus using the value that was suggested.
-g
0 x

Merlinb
Occasional poster
Occasional poster
Posts: 462
Joined: Mon 11/06/06 2:00 am
Location: Shropshire, England
Contact:

Post by Merlinb »

Good treatment of this subject is given in Radio Engineering, Terman, 1937 p231-239.
In brief, oscillation can result from the combination of stray inductance in the grid circuit, and stray capacitance across the output transformer, plus interelectrode capacitance. In other words, there are stray resonances in both grid and anode, and if they coincide the sh*t hits the fan.

p238:
"If trouble is encountered from this residual grid-plate capacity... place a resistance of a few thousand ohms in series with the grid.... such a resistance introduces proportionately more loss at the higher frequencies...and thereby tends to counteract the lower input admittance as the frequency is increased."
0 x

User avatar
zaphod_phil
Builder, Admin
Builder, Admin
Posts: 15208
Joined: Wed 03/19/03 2:00 am
Location: YYZ

Re: grid stopper resistors on power tubes

Post by zaphod_phil »

The purpose of the grid stoppers is also to help prevent blocking distortion.
2 x
Nature abhors a clean tube amp

Post Reply