On 3/3/08, James Elliott put forth:
>Mark, I haven't forgotten your request. It's going to take some time
>before I can answer your questions.
No problem :) I still haven't built mine. At this point, I'm
waiting for the weather to get warmer. Then I'll build it, and JH's
Tau Phaser pedal. I'm also interested in envelope followers and
twin-T filters, which I will probably end up building on perfboard
and putting in project boxes, since I can't seem to find satisfactory
PCB's/panels.
>Why? Well because I have no EE experience other than the little I've
>gained building full KITS from MOTM. I've just recently started
>working from "scratch". ie buy a PCB and then source all the parts
>and troubleshoot issues on your own because of little to no
>documentation.... I spent about an hour and a half reading about
>vacuum tubes yesterday so I could get a little perspective on how
>they work before crafting a response to your message.
Imho, reading the "theory of operation" in the MOTM manuals is a
great way to learn how things work. The kits also provide very good
examples of grounding, shielding, parts selection, etc.
>With all of that being said, on the latest version of Ken's CGS65
>PCB, Ken has implemented the "bias option" on the board itself.
I just noticed that. I also have version 1.1 of the PCB. Since it's
a single-sided board, it's fairly easy to hold it in front of a light
and see where the traces go.
>In order to choose which option you want to implement you have to
>set two different components: 1) A jumper between one of two points
>2) AND a resistor between one of two points. With the DPDT switch,
>one pole switches the jumper position, the second pole switches the
>resistor position. Now, it very well may be possible to use a SPDT
>switch to do this, however upon a novice's first (or second, third,
>fourth) glance it seems that the jumper switches between to pins on
>the tube - so I don't know that a SPDT will work without some other
>circuit voodoo magic. I'm writing this at work so I can't
>necessarily verify what I've just said at this moment....
The way to do it is to not install the 330K resistor on the board,
and solder it directly to the switch. The best way is to use heat
shrink tubing, which I guess is sort of like magic :)
This way the switch either connects the grid through a 330K bias
resistor to one of two volatge points on the board. If one of those
points ends up being ground, or if I can find a good way of doing it,
I would use shielded cable to help keep noise off the grid.
Notice where the board can be cut in half, there are extra holes for
things such as ground, and the negative and positive power supply
rails.
>I'll try to do some more research sometime this week so I can tell
>you EXACTLY how everything is connected (unless someone beats me to
>it). I'll also try to get that measurement for you once I figure out
>what you were asking me to measure :)
OK :) What I waould like to know, is what is the voltage on the
signal input grid, when there is no input signal, for each switch
position. The grid is pin 6 on the tube.
Also, where did you connect the suppression grid?? While you are at
it, could you measure its voltage?? That's pin 4 on the tube. On
the CGS website, it says "it may be appropriate to connect pin 4 of
the tube to pin 1 of the tube instead of pin 2 as well, if making
this modification." I don't know how much of a difference that would
make, or how that could help. The suppression grid deflects "loose"
electrons back towards the plate. Generally it's connected to the
cathode, but this is a rather unique circuit. It should be more
negative than the the plate (which is a given) and screen grid (which
carries the control voltage of the VCA) on either side of it, so that
it does not draw current.