My Mini Moog Voyager. Click to go to my web site...

Wednesday, December 22, 2010

VC Trigger, S&H, and Trigger Converter

The VC Trigger Generator, Sample & Hold, and Trigger Converter work great. There are some footnotes however:

VC Trigger Generator:
The first go around just wasn't working right, so I gave it to Nyle to look at. There were a couple of changes Nyle made. He moved where the Gate and Run switch connect and a de-bounce circuit. The push button switches I'm using bounce a lot. The good thing is it prompted Nyle to add some de-bounce circuitry to the modules so they should be more robust for everyone. He also added an LED for visual feedback of the rate and width.

Sample & Hold:
Worked first time. I was concerned that the high impedance nodes weren't guarded, but I spread out the legs of the FETs and kept the other traces as far away as practical to minimize leakage. It seems to hold for several seconds with no droop.

Trigger Converter:
Works fine first time, too. Although, the way I laid out the panel is opposite to how the PCB is laid out. I ended up rotating the PCB by 180 degrees so the jacks were close to the connectors.

I was playing around with the sequencer and EG and realized the sequencer clock was not triggering the ENVDT section; EG3. I took the EG and sequencer down to Nyle and we spent an afternoon trouble shooting. This resulted in changing two resistor values. The problem was related to my doubling the timing capacitor C3 because I had to use a 1 Meg pot instead of a 2 Meg pot. The clock output pulse of the sequencer is pretty short, about 0.0001 second. Using a doubled value for C3 prevented C3 from discharging while the trigger was active, and so the primary flip flop wouldn't stay "flipped". This is only a problem for short triggers and I could have just as easily used a smaller capacitor. But Nyle likes to have more robust designs. So, changing the two values ensures the EG will work with short triggers, too.

Last, UJTs:
The sequencer clock was saturating at about mid-span for the rate control. Nyle suggested tweaking a couple of resistors, but it didn't change anything. I was using a 2N2646 UJT, listed as a substitute for the 2N4891 originally used. I have several other UJTs, so I thought I'd try a different one, NTE6409 to start. That fixed the saturation. So, I tried a 2N4871 and it worked, too. The best thing is the 2N4871 is the cheapest, about $1 or $2 ea as opposed to $6 or $7.

With that new information, I swapped out the UJT in the VC Trigger Generator and that fixed some residual erratic behavior I was seeing. I also used the 2N4871s in the Sample & Hold in the random voltage sections.

OK, I've also made a couple of videos, but they really stunk. Nyle also suggested I do some videos for YouTube. I will make some, really.

No comments:

Post a Comment