Grid Leak Resistors

[hesamadman]

What are the affects of different value gird leak resistors? In my 2203 with an added gain stage I have been playing around with some modifications. One thing I added was 330k grid leak resistor (on a switch) in parallel with the already existing 1m grid leak resistor. This in my thoughts is an obvious gain reduction as I have just opened up the the path to ground a little bit. But what my ear is hearing is a bit of a reduction of the muddiness that can come with a lot of gain. I dont want to say that its a reduction in bass but I guess some of the bottom end is being reduced? This is a topic that I would like to gain more knowledge. It would further help my modification process on this amp.

[SILVERGUN]

If you've got Merlin's preamp book (if not, get it), look up and read into 'coupling'...specifically C-R, AC or interstage coupling.
Basically,,,the coupling cap and grid leak resistor form a high pass filter, where the value of the coupling cap is used to attenuate low freq (helpful in tightening bass response and stopping blocking distortion)
 
Here's a cool calculator to help you see that what your hearing is the result of a predictable equation:
https://www.ampbooks.com/mobile/amplifier-calculators/coupling-capacitor/

[dennyg]

+1 on Merlin's pre-amp book!  In addition to the high pass filtering Silvergun mentioned, the grid leak is also the lower leg of a voltage divider of some sort, depending on the local circuit, so when you drop the value of the grid leak, you’re also reducing the signal gain going into the following grid which would reduce overdrive/clipping in the following stage. 

[SILVERGUN]

+1 on Merlin's pre-amp book!  In addition to the high pass filtering Silvergun mentioned, the grid leak is also the lower leg of a voltage divider of some sort, depending on the local circuit, so when you drop the value of the grid leak, you’re also reducing the signal gain going into the following grid which would reduce overdrive/clipping in the following stage.

Picture your guitar signal as an AC voltage at a variable frequency
 
Use a voltage divider to drop a certain amount of signal between stages
 
Commonly, you'll see 470K in series with the signal (prior to the grid leak - R1) and 470K to ground (grid leak -  R2)...this results in a 50% signal reduction ( V in = 1V  -  V out = .5V  - see pic)
-
http://www.raltron.com/cust/tools/voltage_divider.asp

[PRR]

> 2203 ...added gain stage ...parallel with the already existing....

Completely un-clear where we are here.

> reduction of the muddiness

IF there is a C feeding this R, you have moved the bass-cut up two octaves. If it was 2Hz and is now 8Hz, no change to the ear. If it was 50Hz and is now 200Hz, BIG change in bass and "mud".

You could get a similar bass-cut change with a smaller C. And in mass production, save a penny each. 

[hesamadman]

Hi guys. Thanks for your reply. I do have merlins book and I will read this section. I have an understanding of how caps affect frequency. I drew out the schematic of the specific section I am talking about. The coupling between these two stages has a 330k resistor switched. When it is on I notice a drop in lower frequency. If my math is correct, it takes the grid leak from 1m to 250k? I guess im just experimenting to see what different values do here. So heres the schem if you guys could explain the science here a little more. 


EDIT: The 470k and 500p paralleled components are actually 1m and 500p

[sluckey]

You should also hear a noticeable drop in volume with that resistor switched in.

[jjasilli]

It may simply be the effect that hi gain also boosts noise of all sorts. So lowering the gain causes noise to drop out.  IMHO the point of a hi gain stage is to clip signal to square waves which are noisy sounding.  Cutting input signal to a hi gain stage may at first seem to be at cross-purposes.   But it's one option to tame a hi gain stage when desired.