Adding a grid stopper to smooth overdrive...

[playonit]

I was searching mods for the 5e7 circuit on Google when I came across a mod removing the wire between V3/7 to the P/I 1m pin on the board and replacing it with a grid stopper anywhere between 100k and 1m mounted direct from socket to board.. It's benefit being the smoothing of the overdrive removing harshness at higher volumes and a little added compression. I put a 220k resistor in to replace the wire but have been unable to really turn the amp up because the family has been shut in because of the weather...

What are the thoughts here about this.... Is there trade off..... pick something up here loose something there?? my biggest gripe about this circuit has been the gain at higher volume. [/b]

[echuta13]

Read the last paragraph regarding the grid-stopper: http://www.freewebs.com/valvewizard/cathodyne.html

AFAIK there really isn't much of a trade-off.

[tubeswell]

That cathodyne stage can take 1M grid stopper without any noticeable difference. If you are worried about it, use 470k. The input capacitance of a cathodyne stage is really low (about 3pF for a 12AX7 triode), so you can get away with a large grid stopper without losing any guitar high end.

[Structo]

It's interesting to experiment with things like this.
You can read all the books and forums about this stuff and do the calculations, but until you try it, it is just words.

A great device to have is called a decade or substitution box.
This can be resistive or capacitive.

It's a box with two clip leads that has a bunch of resistors or capacitors on rotary switches
so you can substitute different values in circuit while the circuit is hot.

For grid stopper or coupling resistors this can really help you to understand how the different values affect tone and stability.

For instance, we are taught that when coupling stages that a higher resistance will start to attenuate the high frequencies.
But at what point will it start degrading signal strength and volume.

That is where this little box comes in handy.

For interstage coupling I find that going over 200K in a lot of instances will start to lower the volume,
which can be a good thing or bad thing given the amplifier.

And a capacitor box can demonstrate how a larger cap will allow more bass through or how a smaller cap will allow
more treble through.

Anyway, another tool for your box.

You can spend anywhere from $15 on up to hundreds.

Here is a good one for $25.
http://www.amazon.com/Elenco-Watt-Resis ... B0002KX76M

http://www.amazon.com/Elenco-K37-Resist ... im_sbs_t_2

As with anything, read the reviews.

[mike9]

Thanks for those links Tom - I just ordered one of each

[rdjones]

That cathodyne stage can take 1M grid stopper without any noticeable difference. If you are worried about it, use 470k. The input capacitance of a cathodyne stage is really low (about 3pF for a 12AX7 triode), so you can get away with a large grid stopper without losing any guitar high end.

Yep, you can look at it as a first order RC low pass filter.
But to describe the total effect the tube's effective capacitance is multiplied by the gain.

http://en.wikipedia.org/wiki/Miller_effect

This is why different types of tubes and even different tubes within a type will change the corner frequency, possibly into the audible range, when a large(r) stopper is used.
More gain x higher input capacitance = lower rolloff frequency.
This will also contribute to different tubes sounding more alike in the absence of a grid stopper, especially when driven by a lower source impedance.

In the case of the cathodyne (no gain) the stopper resistance can be quite large before audible frequency rolloff.
It serves mainly to restrict grid current and it's effect on clipping behaviour.

RedDog Steve

[jjman]

I recently rigged a 1meg pot as a variable grid stopper, but to a LT type PI. I couldn't hear any loss in highs until around 400k or so. Not a huge result even at 1meg. It did not change the clipping characteristics as it's purported to do in a cathodine type. My guess is the cathodine clipping "problem" isn't there on a LTPI. I think I left a 330k or 470k in there to remove a little bite.