SOLVED General Tso's comp - not 'flashy' enough?

[jamesm]

Hello everyone.

Just 'finished' a General Tso optical compressor and I'm now testing it before putting it in an enclosure. I've checked all the voltages. The audio signal flows through from input to output. When I cover the LDRs in tape, the volume increases. The audio signal is making it to the bases of Q2 and Q4. So far, so good...

The thing that's giving me pause and stopping me from shrinkwrapping the LED/LDRs is that the LEDs don't seem to be illuminating in time with the peaks in the music. If I connect the cathode of LED 1 to ground, it illuminates strongly; if I connect the anode of LED 2 to Vcc, the same. Just nothing when the guitar is playing. I was expecting flashes of light.

Is this to be expected? Is the LED illumination supposed to be so small as to be imperceptible?

These are the LEDs: https://www.mouser.co.uk/datasheet/2/427/tlhg640-1766969.pdf

Looking again at that datasheet, I think I may have picked the world's dimmest green LED...

 

[zgrav]

Glad you could confirm they are working before you seal them together with the LDRs.

 

[jamesm]

Yeah, I've made other designs in the past using Vactrols and you're never quite convinced, even when it sounds condensed...

 

[fancypete]

I know I'm late to the party but I had a similar experience with the green LEDs from Tayda. Very dim indeed. I ended up using some red LEDs since they were brighter and the LDRs I had were sensitive in that wavelength. The pedal sounds great, but I'm always wondering "what if?" The problem is there are way too many variables to deal with on this design. Which LED, LDR, distance between them, etc. Also the fact that it's super hard to test before boxing because of light pollution. Probably everything makes a difference.

A cool idea would be to assemble a few different LED/LDR combos in a factorial design with switches so that one could A/B/C/... compare. One could put the whole protoboard in a light tight box and put the issue to rest, forever.

That said, whether or not it's functioning as intended, it's definitely a keeper. It sounds excellent.

 

[jamesm]

The main problem for me was sourcing LDRs that were anywhere near the supposed spec. At least you know where you are with a Vactrol!

 

[jubal81]

You can't really use your eyes to judge how well it's working as far as what amount of light you see. LDRs are VERY sensitive to light and just because you're not seeing much light, it doesn't mean the LDRs aren't reacting appropriately.

I shield my couplers very well these days because any tiny bit of ambient light, even from the bypass LED, can limit their max resistance level. Performance issues with LDRs aren't always about getting enough light, but also about getting enough dark.

 

[Chuck D. Bones]

Like Jubal81 said, LDRs are extremely sensitive to stray light. The only way to test a circuit like this in the open is to do it in a darkroom. Optical compressors use the LDRs to lower the gain when light from the LEDs shine on them. If something else is illuminating the LDRs, even a little bit, then the gain is turned down and there is little or no signal to light up the LEDs. I believe Bill's Law is applicable here.

 

[jamesm]

Yeah, I thought that putting blu tack on the LDRs would work well enough to block the light, but apparently not. It's been a lesson form me. Looking forward to getting it boxed up tomorrow.

 

[Boukman]

Hi!!I have built the Tso for a while now and though l love how it sounds i don`t hear the compression so much..If i turn the blend too high i get too much volume (also in he juicy mode).Even with the sustain at full there`s a lot of volume.I used green leds and ordered the ldr listed in the bom and put it facing the leds.I`ll have to check in the dark if they illuminate i guess..Cheers!