VFE White Horse Optical Compressor not squishin'

[bowanderror]

Hi guys,

I recently built a RoughCut Optical Compressor, VFE Bumblebee Swell-Compressor, & a VFE White Horse Optical Compressor. The first two work great, but I'm struggling to get the White Horse to get it's squish on. Even with the Sustain knob cranked, I get only subtle compression. I also can only get the blue LED to illuminate with the VFE switching board 20k trimpot at it's highest setting, and even then it's quite faint.

I've tried a few different colors/types of LEDs. My blue LED is not the exact one listed on the built doc, but has a Vf in with the same specs. Red/yellow/green LEDs illuminate well at lower trimpot settings & are much brighter. Changing the type of LED and/or the trimpot setting doesn't seem to effect the amount of compression.

I went through and confirmed that the resistor & pot values are correct. I am using the photo of the v5.0 board from the build doc.

I took voltage measurements to compare with those from the build doc. I've attached a photo with my voltage readings, taken with the pedal on & the switching board trimpot at MAX:

• V+/V- are OK coming from the switching board & on the quad IC.
• I also took voltage readings of the positive power rail that feeds the LED anode of the NSL-32SR3 vactrol, through 2.2k R8, then A50K response pot, and 2.2k R17, where it feeds the collectors of the 2x 5088s & the cascaded BS170s.
• My test voltages on the White Horse's 2N5088s & the first BS170 are lower than those listed on the build doc (4.73V vs 7.28V on the build doc). Keep in mind, they are listed for the v4.2 PCB, whereas my board is v5.0.
• The gate of the last BS170 (which feeds the LED anode) is a bit lower (2.34V vs 2.53V on the build doc), and is right in the middle of the forward voltage range listed for the LED.

I've kinda run out of ideas, so any help would be greatly appreciated! If anyone has answers to these questions, that would great as well:

• I soldered in the BS170 & 2N5088 transistors, should I go back and socket them to see if that helps?
• Does anyone have voltages for the v5.0 board, or are they the same as v4.2?
• What should the knob/trim pot positions be when taking voltage readings?

[madbean]

Shouldn't be subtle at all so yeah you have a problem.

- Your parts values look correct. I think you might have heated up a couple of caps (just kidding you, I do it too sometimes).
-Voltage readings b/w 4.2 and 5 ver should be about the same. The are very nearly identical.
-Your voltages actually look pretty good. The 2n5088 will depend on where the Response pot is set at the time of readings. Based on my list, I must have had the Response pot turned most of the way down since the collector voltage is close to V+.
-Your BS170 voltages look different and I wonder if that comes down to LED choice here. As in despite your effort to choose the correct LED it may be what is causing the problem of muted compression.

I started to type up an analysis of the anti-pop circuitry but it was getting a bit TMI so I'll hold off on that. I think there are two things you could try right away to see if your LED choice is the problem:
1) Defeat the anti-pop circuitry by either removing Q3 or clipping out the Drain lead (do it in the middle of the lead so you can solder the two halves back together easily). That's the far left lead of the BS170 on the left side.
2) Don't do #1, leave everything as built but temporarily put a lower Vf LED in parallel with internal LED of the NSL32 to see if it lights up when you play. This is actually more awkward because if you have a 16mm pot soldered in under the NSL it's going to be really hard to get to. But, honestly, I don't think your optocoupler is the issue. I've bought and used dozens upon dozens of them and never had a problem with a single one.