General Tso’s Compressor

[Mike McLane]

Where did you get 100mcd? I contacted Smallbear and what I have puts out around 30mcd. I can get the 100mcd from Mouser, though. I found a data sheet for the Tayda LDR KE-10720 (attached) and it looks pretty darn close. GL5516 calls for light resist: 5-10K, dark resist - 500K, rise time - 30ms, fall time - 30ms. This appears to match except for a fall time of 5ms. Don't know how that would affect performance, but should work until PPCB gets their GL5516's in (which I'm sure will be any day now ?). Thx tons. BTW. . .GUYS LIKE YOU ARE INDISPENSIBLE TO THE SUCCESS OF THIS FORUM AND TO ME IN PARTICULAR. I'M SURE I SPEAK FOR A LOT OF OTHERS AS WELL!!

 

[Chuck D. Bones]

I don't think you need such a high luminosity. I've only built a few optical compressors, but I never had an issue with not enough light landing on the LDR. It was usually the opposite. I've had to move the LDRs back from the LED, or use a lower brightness LED. The General Tso has two LDRs in parallel with the 100K SUSTAIN control. They work in tandem. The gain of that stage is set by the SUSTAIN control, the two LDRs and a 10K resistor to Vref. If the resistance range of each LDR goes from 10K (light) to 200K (dark), the pedal will do everything it was designed to do. If the LDR resistance goes higher or lower than that, it's ok, but it won't make the pedal perform any better.

 

[Pimpoftheyear70]

What LEDs do you use in these locations ?
Water clear or diffused style ?
And why green ??

 

[Chuck D. Bones]

Most of the LDRs we use have their highest sensitive in the yellow & green part of the spectrum. They'll still work with any of the other colors, the LED just needs to be brighter to get the LDR down to the same resistance. I think diffuse is better, for two reasons.
1. Diffuse LEDs will spread the light more evenly across the entire LDR.
2. The LDR position will be less critical since light is going in all directions instead of just straight out the top of the LED.

Here's what I did in my Delegate, it works great. LED is regular brightness diffuse. LDR I bought surplus and cherry-picked. It has a light resistance under 2K and a dark resistance over 1Meg. That compressor needs a much higher dark resistance than the Tso. I actually had to move the LDR away from the LED for better performance. Too much light = slow release.

 

[Mike McLane]

Hmmmm? OK, so riddle me this Mr. Peabody (does anybody remember Sherman & Mr. Peabody?). How 'bout we go back to your suggested "black box" test. I hook up the my looper and "strum roll" an open E chord and run a continuous drone into the pedal. Assume silence gets me >= 200K. . . if full volume gets me =<10K then we're good??? If the schematic indicates that the proper resistance limits are 10K-200K for optimal performance it would only make sense to have to "calibrate" the LED/LDR "relationship" to yield the appropriate resistance ranges. I'm thinking some guidance on the LED spec should be clarified in the buld docs.

 

[Chuck D. Bones]

Sherman & Peabody... ahhh the Classics.

Sometimes I think it's more like Coyote vs. The Roadrunner where I'm Wile E. Coyote and the tone I'm chasing is The Roadrunner. That would make Tayda the ACME Corp.

The 10K-200K resistance limits I provided are one edge of the performance space. Any diffuse green LED should work fine. Just because I have a tendency to overthink things doesn't necessarily make it a good idea.

 

[Dan0h]

Most of the LDRs we use have their highest sensitive in the yellow & green part of the spectrum. They'll still work with any of the other colors, the LED just needs to be brighter to get the LDR down to the same resistance. I think diffuse is better, for two reasons.
1. Diffuse LEDs will spread the light more evenly across the entire LDR.
2. The LDR position will be less critical since light is going in all directions instead of just straight out the top of the LED.

Here's what I did in my Delegate, it works great. LED is regular brightness diffuse. LDR I bought surplus and cherry-picked. It has a light resistance under 2K and a dark resistance over 1Meg. That compressor needs a much higher dark resistance than the Tso. I actually had to move the LDR away from the LED for better performance. Too much light = slow release.

View attachment 7167

Hey Chuck, what is that housing around the LDR? Did it come that way or did you rig that? Your work is epically clean man. Cheers.

 

[Chuck D. Bones]

Thanks.

The LDR came that way. Bought it from Electronic Goldmine. They had 'em on sale, 5 for $1.99. Normally they're $1 ea.

https://www.goldmine-elec-products.com/prodinfo.asp?number=G15176

They have quite the selection of LDRs. Be warned, they are untested surplus. Lotta part-to-part variation and a few are duds. They're inexpensive (when they're on sale) and I don't mind testing them.

EG has these LED/LDR isolators, but a lot of them turned out to be unusable, so caveat emptor.

 

[zgrav]

I think the speed of the LDR has more to do with how much snap you do or don't get from the rise in the sound envelope when you hit the strings. I don't think the speed of the LDR matters in most phaser applications, but it can make a difference in how crisp the attack is in a compressor.

 

[Chuck D. Bones]

Absolutely. When I test LDRs, I drive the LED with a square wave and watch how quickly the resistance changes. Here's an example using one of the TO-18 can LDRs. Green traces is LED current, 1mA/V. Blue trace is LDR voltage. LDR has a 100K pull-up to +9.3V. At 4.6V, the LDR resistance is 100K. LED is blue diffuse, LDR is up against the LED, near the top. At this time scale, the LDR resistance drops almost instantaneously when the LED turns on. The LDR takes 60ms to get up to 100K and 120ms to get up to 200K when the LED turns off.

 

[Dan0h]

Thanks.

The LDR came that way. Bought it from Electronic Goldmine. They had 'em on sale, 5 for $1.99. Normally they're $1 ea.

https://www.goldmine-elec-products.com/prodinfo.asp?number=G15176


They have quite the selection of LDRs. Be warned, they are untested surplus. Lotta part-to-part variation and a few are duds. They're inexpensive (when they're on sale) and I don't mind testing them.

EG has these LED/LDR isolators, but a lot of them turned out to be unusable, so caveat emptor.

This Website!

 

[Mike McLane]

Holy crow! I just checked Mouser and I can get 5mm green LED's anywhere from 1.6mcd to 180mcd. In the range mentioned here. . . 30, 38, 80, 100. I've got 30 now so assuming Chuck's right about 100 being too much that leaves me with 38 or 80 as possible alternatives (or a new set of ears). What's a feller to do?

 

[Mike McLane]

Just got to tearing into the bugger to check the range on the LDR. It reads 12K under normal room light. . . 3.5K with my LED reading lamp on, HOWEVER, in complete darkness I only get 28K (spec sheet calls for 500K). I also checked (however subjectively) that the LED's were operating and they do flash when a signal is put to the unit, ever so small a blink when Sustain is at min, significant reaction when max. In fact I notice that the LED on one side is more signal sensitive than the other. Any thoughts?

 

[zgrav]

Are you getting those readings in circuit? Are you getting the same resistance ranges of both of the LDRs? Looks like your 100K sustain pot and the two LDRs are in parallel, so they would all three affect the resistance you measure for either one of the LDRs. if you are only blocking the light on one of the LDRS the light hitting the other LDR will affect your reading.

you need to max out the resistance on your pot and then make sure both LDRs are in the dark when you measure the resistance on each of them. keep in mind that the most resistance you could measure for each LDR when it is in the circuit will be maxed out by the 50K pot.

 

[Mike McLane]

Ah, yes. I didn't have the schematic in my build docs, but dl'd the update. I see what you mean with the SUSTAIN knob. . . I max'd it out and retested and got a reading of 107K (probably the real value of the pot) under total darkness. I guess that means that the LDR's are functioning properly.

 

[zgrav]

Sounds like the LDRs are working. are you testing the effect with all other light shut out from the LDRs?

 

[Chuck D. Bones]

Are you getting those readings in circuit? Are you getting the same resistance ranges of both of the LDRs? Looks like your 100K sustain pot and the two LDRs are in parallel, so they would all three affect the resistance you measure for either one of the LDRs. if you are only blocking the light on one of the LDRS the light hitting the other LDR will affect your reading.

you need to max out the resistance on your pot and then make sure both LDRs are in the dark when you measure the resistance on each of them. keep in mind that the most resistance you could measure for each LDR when it is in the circuit will be maxed out by the 50K pot.

If we're still talking about the General Tso, the SUSTAIN pot is B100K. Pots tend to have a pretty loose tolerance, like ±20%, so 107K is in spec.

 

[zgrav]

if the pot and LDRs are all parallel, wouldn't that drop the effective resistance across all of them?

 

[Chuck D. Bones]

Yes, in accordance with this formula:



Let's assume the LDRs are each 2M and the SUSTAIN pot is 115K, then R = 103K

 

[zgrav]

Or 50K if the resistance on the LDRs is 200K each with a 100K pot, using this calculator.