AB763 in a pedal, output problem

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
A huge output? Like the signal coming out is too hot?

The schematic shows a master volume, are you saying it's not effective to just turn it down there?

Keep in mind, the AB763 preamp was designed to drive a power amp to full volume, which needs on the order of 5-10V peak signal in the 20W models. This is WAY too much signal for a pedal, which even if driving a line-level input only needs around 1V peak.

IMHO, the best way to cut gain in these common-cathode tube stages is to split the plate load resistor (R8, R4). For example, if you split R8 into two, 50K resistors, and then tap the signal from the middle you would theoretically reduce the output by half, without effecting the voicing of that stage.

If you want to experiment, you can substitute in a 100K pot, and take the output from the wiper, and adjust it until you get the maximum expected signal you need with the master volume at full. Then, you can measure it and substitute in two fixed resistors. Be forewarned: the pot will have DC across it, and will scratch LIKE HELL, but for testing it's ok.
 
Last edited:
You got it!

However my original question remains: if you turn down the master volume, the signal should be cut as well?

It does, but it doesn't much and doesn't mute when all the way to the left...
The original AB763 schem has a 1M resistor to GND after C9 and before the output, so I thought of replacing it with a 1M log to have an output control, since when the Gain is hot the volume id huge, but someone here told me to use a 220k Resistor + a 10k log. But that doesn't work very well. It does turn the volume a bit down but not a lot and not all the way to silence. Should try the 1M log or split the plate load resistor?

I did a search and found a couple of very interesting papers aboput splitting plate load resistors, especially one by Rob Robinette :)
Thanks a lot for your help
Sono
 
Got it.

I had assumed the master was 1M, but looking closely I see indeed it's a 10K. A 10K won't do anything, as it's too small relative to the the 220K.

I would replace the 220K/10K with a single 1M pot, and see what kind of adjustment range you get. If you find that it only 'works' between 0 and 1 (or 2) on the master (my guess), you can split the plate resistor and find a ratio that gives you full adjustment between 0 and 10 on the master volume.

If I had to guess, even a 50/50 split will be too much: more like 1/10 ratio would work (10K on the bottom, and 90K on the top).
 
At idle, you have roughly 100V across the 100K resistor (~280V B+4 average, with ~180V sitting on the plate). Even if the tube was a dead short, you'd be dropping 280V across 100K which is 3/4W. If you have two resistors, 1/2W should be fine.
 

Latest posts

Back
Top