No Grid Leak on Input Stage??

[wizard333]

Was working on an amp tonight and found no grid leak on the input stage (typically 1M), which I assume was a mistake. Got me thinking though:

Normally that 1M is in parallel with your guitar's volume pot, so without it, your volume pot alone becomes the grid leak. Is that input stage grid leak really even necessary? Without that 1M in parallel with your volume pot (~333k for a 500k pot, ~200k for a 250k pot), the tube grid sees a higher value grid leak (full volume pot value plus whatever the grid stopper is; cable resistance too small to make a dif), and the pickups see a higher input impedance for that stage.

Is that correct, or is this my brain throwing up at 2am?

While we're at it, say you put a .047uf input capacitor before your grid leak at the input. How does that affect the Rg your grid sees and the input impedance your pickups see?

[pdf64]

I think it's poor practice to omit a grid leak at the amp's input, eg if the instrument was unplugged from the cable, then the grid would be open circuit and the tube's operating point would drift off, may red plate and become damaged; a pedal may not have a dc path to ground on its output, causing the same thing to happen.
Yes, as the input grid leak value is increased, for a directly connected instrument with its vol and tone controls at max, pickup resonance will likely increase in magnitude and frequency. See http://www.buildyourguitar.com/resources/lemme/
With a grid leak in place, a suitably large dc blocking cap shouldn't have any noticeable tonal effect; if the particular tube / circuit is prone to high grid current, it should stop the instrument volume control making a scratchy noise when operated. In the case of a grid leak biased input stage (ie grounded cathode) an input blocking cap would be pretty much essential (though that's not to say that some cheaper designs won't do without them).

[wizard333]

I think it's poor practice to omit a grid leak at the amp's input, eg if the instrument was unplugged from the cable, then the grid would be open circuit and the tube's operating point would drift off, may red plate and become damaged; a pedal may not have a dc path to ground on its output, causing the same thing to happen.
Yes, as the input grid leak value is increased, for a directly connected instrument with its vol and tone controls at max, pickup resonance will likely increase in magnitude and frequency.
With a grid leak in place, a suitably large dc blocking cap should have any noticeable tonal effect; if the particular tube / circuit is prone to high grid current, it should stop the instrument volume control making a scratchy noise when operated. In the case of a grid leak biased input stage (ie grounded cathode) an input blocking cap would be pretty much essential (though that's not to say that some cheaper designs won't do without them).

Excellent points; thanks!

[R.G.]

Some older tube circuits were designed to run on - um, grid leak bias?? - I think it was called. No grid leak resistor, but cathode grounded. Leak electrons piled up on the grid and self biased it somehow. I'd have to go look it up. Big gain, intolerant of any funny conditions IIRC.

[KevinOConnor]

Hi Guys

The grid must have a DC path to ground, or more importantly, a DC path to its cathode. Without this tether, the tube will not be controlled and will conduct maximum current. For a power tube, this results in a "red plate" condition where the plate literally turns red with excess current producing excess heat - more heat than the plate was designed to dissipate.

For a preamp tube, red-plating is not an issue as the internal impedance of the tube is so high that current will be limited to a value (generally) within the plate dissipation rating. An uncontrolled 12AX7 might pull a few milliamps.

At the input of a guitar amp, it is important to have a grid-leak resistor of some sort, with 1M being the most common value. This value is not entirely arbitrary as it helps to provide a small bit of damping to the inductive pickup coils on passive guitars. because the tube grid is of an inherently much higher nominal impedance under normal conditions, one can increase the grid-leak value to 10M or higher to match raw piezo piclups and have full bass capture.

It is also important at the input to have a grid-stop of a reasonable value placed right at the tube socket. This forms a low-pass filter with the tube's internal capacitance which reduces risk of radio break-through.

"Grid bias" is an old method of biasing the tube where there is no cathode resistor. A very high r\grid-leak resistance is needed so that the charge built up near the grid can develop enough voltage difference between the grid and cathode to control the tube's idle current. This generates a negative voltage at the grid since the cathode is at ground potential. A DC locking cap is then required so that the signal source does not upset the bias of the tube.

Grid-bias is not as useful as normal cathode bias for guitar amp preamp stages, as cathode bias affords a means of voicing the preamp.

Have fun

[Mike Sulzer]

For a preamp tube it is the plate resistor that limits the current through the tube and assures that when the current is at its maximum, the voltage across the tube is small. Thus the dissipation is limited.

The 1 meg resistor has a small, but noticeable, effect on the Q of the guitar pickup circuit. The 250K or 500K volume pot is a bigger effect; the tone pot has a nearly identical effect since on ten the tone cap is essentially a short near the resonant frequency. Thus the two pots together are effective either 125K or 250K, a lot smaller than 1 meg. Also the dc resistance of the coil lowers the Q, although in a somewhat different way. If a pickup has steel in the magnetic circuit, eddy currents can have a signifiant effect on the Q.

The grid stop resistor at the input is usually the major contributor to noise in the input stage. It can be much reduced or eliminated. Since most rf pickup is on the shield of the cable, it can be much reduced with ferrite instead.

Hi Guys

The grid must have a DC path to ground, or more importantly, a DC path to its cathode. Without this tether, the tube will not be controlled and will conduct maximum current. For a power tube, this results in a "red plate" condition where the plate literally turns red with excess current producing excess heat - more heat than the plate was designed to dissipate.

For a preamp tube, red-plating is not an issue as the internal impedance of the tube is so high that current will be limited to a value (generally) within the plate dissipation rating. An uncontrolled 12AX7 might pull a few milliamps.

At the input of a guitar amp, it is important to have a grid-leak resistor of some sort, with 1M being the most common value. This value is not entirely arbitrary as it helps to provide a small bit of damping to the inductive pickup coils on passive guitars. because the tube grid is of an inherently much higher nominal impedance under normal conditions, one can increase the grid-leak value to 10M or higher to match raw piezo piclups and have full bass capture.

It is also important at the input to have a grid-stop of a reasonable value placed right at the tube socket. This forms a low-pass filter with the tube's internal capacitance which reduces risk of radio break-through.

"Grid bias" is an old method of biasing the tube where there is no cathode resistor. A very high r\grid-leak resistance is needed so that the charge built up near the grid can develop enough voltage difference between the grid and cathode to control the tube's idle current. This generates a negative voltage at the grid since the cathode is at ground potential. A DC locking cap is then required so that the signal source does not upset the bias of the tube.

Grid-bias is not as useful as normal cathode bias for guitar amp preamp stages, as cathode bias affords a means of voicing the preamp.

Have fun

[wizard333]

"one can increase the grid-leak value to 10M or higher to match raw piezo piclups and have full bass capture."

Isn't that true only if something is plugged into the input? Without something plugged in to be in parallel, isn't that 10M too high for a 12AX7 grid leak, at least according to the tube manual? I thought the max was around 2M.

The amp in question does not have a grounded cathode and does have a grid stopper, so I an assuming the lack of a grid leak was an oversight. I think the input stage on the 12AX7 did suffer for it with nothing plugged in at some point, so I replaced the tube and added a 2M grid leak and the amp sounds a lot better for it.

[Leo_Gnardo]

Without something plugged in to be in parallel, isn't that 10M too high for a 12AX7 grid leak, at least according to the tube manual? I thought the max was around 2M.

1960-70's Ampegs commonly used a 5.6M grid leak R, and I often find their preamps running fine with original pre tubes in place. No reason you couldn't go at least halfway to 10 Megs.

[Merlinb]

"one can increase the grid-leak value to 10M or higher to match raw piezo piclups and have full bass capture."
Isn't that true only if something is plugged into the input? Without something plugged in to be in parallel, isn't that 10M too high for a 12AX7 grid leak, at least according to the tube manual?

The 12AX7 has very low grid current, so you can in fact go up to 10M provided the anode current is <1mA and the bias is greater than 1V, which it often is in a 12AX7. Some 12AX7 datasheets do confirm this.

The amp in question does not have a grounded cathode and does have a grid stopper, so I an assuming the lack of a grid leak was an oversight.

It wasn't uncommon in early, budget amps, where even cent counted. The grid would usually be grounded when nothing was plugged in.

[Mike Sulzer]

1960-70's Ampegs commonly used a 5.6M grid leak R, and I often find their preamps running fine with original pre tubes in place. No reason you couldn't go at least halfway to 10 Megs.

You can use a very large resistor, much bigger than 10 meg, if you have a method of stabilizing the cathode current, for example, adding a negative supply and using a cathode resistor on the order of 100K (bypassed, of course).

[wizard333]

The 12AX7 has very low grid current, so you can in fact go up to 10M provided the anode current is <1mA and the bias is greater than 1V, which it often is in a 12AX7. Some 12AX7 datasheets do confirm this.


It wasn't uncommon in early, budget amps, where even cent counted. The grid would usually be grounded when nothing was plugged in.

This is a recent amp from a medium sized manufacturer. Their other amps I've drawn schematics for all had grid leaks, so I'm guessing this was an oops.

Very interesting that you can go that big on the grid leak (on a cathode biased tube), I had no idea.

Assuming you use an input capacitor so your pots/pickups don't see grid current, the input impedance the pickups see from an AC perspective could be quite a bit higher than the 1M standard, so you could use that as a tone shaping tool.

[Mike Sulzer]

This is a recent amp from a medium sized manufacturer. Their other amps I've drawn schematics for all had grid leaks, so I'm guessing this was an oops.

Very interesting that you can go that big on the grid leak (on a cathode biased tube), I had no idea.

Assuming you use an input capacitor so your pots/pickups don't see grid current, the input impedance the pickups see from an AC perspective could be quite a bit higher than the 1M standard, so you could use that as a tone shaping tool.

How is increasing the resistance above 1 meg going to shape the tone of a guitar pickup?

[daz]

This is a recent amp from a medium sized manufacturer. Very interesting that you can go that big on the grid leak (on a cathode biased tube), I had no idea.

Assuming you use an input capacitor so your pots/pickups don't see grid current, the input impedance the pickups see from an AC perspective could be quite a bit higher than the 1M standard, so you could use that as a tone shaping tool.

But ya gotta figure that if there really was a advantage to it, wouldn't a lot more amps use it instead of the much more common input circuit? These seem to be the kind of things i experiment with, think it's a asset, then later realize it was best left the way it was before. Then i realize once again fender/marshall/etc got it right the first time. Like in marshal MV amps i always find certain things like those treble peakers to be a necassary part of that circuit. I may think it sounds good w/o them, but later i realize there are issues that only putting them back in cures. It happens over and over again. Certain design aspects that you see are the same in most amps or most amps of a given type are usually there ior of a certain value because they just work where nothing else works quite as good. Thats what i find anyways.

[Leo_Gnardo]

How is increasing the resistance above 1 meg going to shape the tone of a guitar pickup?

Some folks can hear that last wisp of hi frequency response gained by having - what looks like a load resistor to the guitar pickup - above 1 meg. And some find that little bit of extra hi frequency response bothersome. For instance I had a guitarist who saw the value in using an MXR MicroAmp as a buffer, but insisted the tone was too brittle with the MicroAmp in use. Changed the 5.6M input resistor to 1M and he got happy once again & pronounced the Micro to be sonically transparent. To him anyway & that's what counts. What effect the grid leak resistor has on the low end, I haven't heard any difference and can't imagine there is any unless one is playing special basses with B or lower strings thru a system that really brings thru the fundamental of the lowest notes, not your everyday rig by far.

[wizard333]

How is increasing the resistance above 1 meg going to shape the tone of a guitar pickup?

The input resistor, in combination with your guitar's pots, affects the sharpness and width of the resonant peak of the pickups. The higher the value, the more narrow and larger the peak.

I measure every pot I buy (which is hundreds a year) and label them, then, if necessary, use a resistor across them to achieve the exact response I want given the tone of the instrument in question and the pickup.

If you raise the value of the grid leak, it's like raising the value of the pots on the guitar, but only for the amp in question. So if the amp seemed a little dull (or a little sharp) across multiple guitars, changing that value might be all the tailoring one needs without other circuit changes, or as part of a recipe of other changes.