THE ENGINEER'S THUMB... At last, a better compressor!

[Giglawyer]

Are there any more PCBs available for this? I would love to build a five knob one.

[MrStab]

i'm kinda bored but low on parts - anyone have an idea if you could use a simple comparator to make a compression indicator? nothing fancy, not actual compression amount or anything. currently probing pins and hitting stings to see what voltages i could compare.

[merlinb]

i'm kinda bored but low on parts - anyone have an idea if you could use a simple comparator to make a compression indicator? nothing fancy, not actual compression amount or anything. currently probing pins and hitting stings to see what voltages i could compare.

One way would be to compare the voltage on the base of the PNP transistor with a fixed voltage (about 4V). When it starts compressing, the base voltage will drop below 4V.

[MrStab]

One way would be to compare the voltage on the base of the PNP transistor with a fixed voltage (about 4V). When it starts compressing, the base voltage will drop below 4V.

sweet, cheers! i'll look into that. would there be any way to make the bias variable by input level or is it just not feasible in this case? and would the circuit be loaded down at all by tapping the PNP?

[merlinb]

sweet, cheers! i'll look into that. would there be any way to make the bias variable by input level

Not sure what you mean...

would the circuit be loaded down at all by tapping the PNP?

Its driven by an opamp so it should be OK with loads down to a couple of k-ohms.

[MrStab]

i mean would it be possible to only light up when played (and compressed, ofc)? i tried what you suggested (with a voltage divider to get as close to the PNP base voltage with threshold/ratio both fully-off) and the light would go on and stay on when the threshold pot was turned past a certain point. so i'm probing around for a source to take the pre-compressed signal from to use that as Vref.
maybe i could somehow use an NPN or an LM386 or something (either with or without a comparator) to accomplish this & maybe even get ratio-dependent brightness too?

[MrStab]

how bout this?



Edit: i actually built this and it kinda works! it's imperfect for now, i probably screwed something(s) up, but at this intoxicated, late hour it seems that the intensity and/or sensitivity of the light is affected by the Threshold pot. it seems to just stay on dimly when it's at 0, which ofc isn't ideal because it doesn't really represent 0 compression, but it doesn't respond at all to playing when in that state which is good i guess. i appear to have an issue where the reference voltage goes up the same as the comparison voltage, prolly symptomatic of a solder bridge although i can't see one (yet).

[Roger Martin]

Merlin B,
How is it doing with 5 knobs so far ?
Is it a worth-it-to-try mod or maybe a 3 knobs is sufficient enough ?
Would u be so kind to share the 2 layer 5 knobs PCB layout here ? Thank you in advance. 

[merlinb]

Would u be so kind to share the 2 layer 5 knobs PCB layout here ? Thank you in advance.  

Here you go, a single-sided version:
https://dl.dropboxusercontent.com/u/57831278/115x60mm.bmp
https://dl.dropboxusercontent.com/u/57831278/TheEngineersThumbGTI.pdf

i mean would it be possible to only light up when played (and compressed, ofc)? i tried what you suggested (with a voltage divider to get as close to the PNP base voltage with threshold/ratio both fully-off) and the light would go on and stay on when the threshold pot was turned past a certain point.

The base voltage is proportional to the compression, so if the circuit is working properly you only have to compare this voltage with a fixed reference and the LED will switch on only during compression. If you have it on breadboard or something, then you may simply be picking up hum, which is being compressed, causing the LED to be on permanently...

[MrStab]

The base voltage is proportional to the compression, so if the circuit is working properly you only have to compare this voltage with a fixed reference and the LED will switch on only during compression. If you have it on breadboard or something, then you may simply be picking up hum, which is being compressed, causing the LED to be on permanently...

that's good news, less hassle lol. it's set up in the enclosure itself (cover off), but definitely full of imperfections as yet. i spent 13 hours straight trying an ADA Flanger the other week and it made me think there must be a plottable curve for time spent:available brain power ratio, as i made a few errors despite what i thought was triple-checking lol. maybe i could put in diodes to get the right forward voltage so the light comes on all staccato-like and pseudo-advanced.

while we're at it, any idea why my build squeals at 0 Release time? different problem probably, but coincidentally also related to a pot at min. resistance. thanks for all the help so far!

[merlinb]

while we're at it, any idea why my build squeals at 0 Release time? different problem probably, but coincidentally also related to a pot at min. resistance. thanks for all the help so far!

What component values are you using for the release pot and series resistor? Maybe they're too small.

[MrStab]

i'm pretty sure i followed the BOM here to a tee http://www.sabrotone.com/wp-content/uploads/2012/05/EngineersThumb.gif,
 so a 1M linear for release pot.

i'll have to fix an issue i've created from too much tinkering - now when i turn the Ratio up just a little bit it clicks into just a fuzzy kinda distortion. could be the BC327 as i was hastily soldering around that, but i'll fix that before trying anything else.

i rebuilt the compressor light circuit to make it less "experimental", the comparator stage worked (trimpot to match bias to transistor base), and everything on the LM386 worked bar the output for some reason. even with the 10uF 200x gain setting the output was too low to light up. worked with the first attempt though so i've probably screwed something up

[Roger Martin]

Would u be so kind to share the 2 layer 5 knobs PCB layout here ? Thank you in advance.  

Here you go, a single-sided version:
https://dl.dropboxusercontent.com/u/57831278/115x60mm.bmp
https://dl.dropboxusercontent.com/u/57831278/TheEngineersThumbGTI.pdf

What a superb sharing, Merlin B !
Awesome job, man. Thank you. 
I'll work on this asap.

[merlinb]

the LM386 worked bar the output for some reason. even with the 10uF 200x gain setting the output was too low to light up.

I would have thought the LM386 is overkill for this application; an ordinary opamp could do the job.

[Roger Martin]

I have a few questions...

1. So this PCB has measurement 115 x 60 mm from edge to edge of the width and height precisely ?
https://dl.dropboxusercontent.com/u/57831278/115x60mm.bmp

2. How do i eliminate the LED bar graph, coz it's hard to find/buy one here ?

Thank you

[merlinb]

1. So this PCB has measurement 115 x 60 mm from edge to edge of the width and height precisely ?

Yes. I always try to use round numbers when designing a board!

2. How do i eliminate the LED bar graph, coz it's hard to find/buy one here ?

The bar graph is just ten LEDs packaged in a solid plastic block. You can use ordinary LEDs instead, wired up to the same pads, like this:

[Roger Martin]

Woo hooo super awesome ! 
Let me try, let me try ......hahaha....
Thank you so much, Merlin B

[sugonidamaso]

Thank you so much for this! Is there a pcb for a 1590b series?without led block? too much to ask but thank u in advance.

[sugonidamaso]

Thank you so much for this! Is there a pcb for a 1590b series?without led block? too much to ask but thank u in advance!

double post sorry.

[MrStab]

take a feed from Pin 7 of IC2 (the opamp connecting to the PNP), hook it up to the Base of a NPN, and wire V+ through the Collector and LED from the Emitter for a simple-yet-handy compression indicator. score!

just re-read my last posts, a power amp IC does seem like overkill in hindsight.

my aforementioned distortion was caused by the board shorting against the treble switch. really is the most obvious stuff sometimes.
now to fix that damn ticking when it's bypassed.