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

Started by merlinb, April 21, 2012, 10:17:37 AM

Previous topic - Next topic

midwayfair

Quote from: MrStab on July 17, 2013, 10:24:19 AM
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.

Sweet, I might add this to one of my builds. :)

I bet if you stuck a green LED on there, you could have two levels of indication. (Blue might be overkill ...)
My band, Midway Fair: www.midwayfair.org. Myself's music and things I make: www.jonpattonmusic.com. DIY pedal demos: www.youtube.com/jonspatton. PCBs of my Bearhug Compressor and Cardinal Harmonic Tremolo are available from http://www.1776effects.com!

MrStab

i used a white LED as they're all i have lying around, seems to work ok. would be cool to have another light for pre-compression level but i'm just trying to fill holes in a repurposed enclosure lol.
i think i used a BC183 although i might've swapped it out for a 2n5088. id imagine either would work, though

Recovered guitar player.
Electronics manufacturer.

earthtonesaudio

Quote from: MrStab on June 27, 2013, 07:01:22 PM
how bout this?


This circuit is an unusual choice.  Let me explain why I would do it differently:

First, a quirk of the lm386: its output biases to 1/2 of its supply when its inputs are the same.  Another quirk is that its inputs are ground-referenced by internal 50k resistors, so your circuit as drawn would work the same with nothing connected to the lm386 inputs.

Next, your comparator is supplying power (or not) to the lm386, based on what happens at the NPN base.  This means if the NPN base goes below 4.36V, the comparator output will go high (ideally 9V but the device will drop some voltage so probably more like 8V).  This means the lm386 output will go to about 4V and light the LED.

Last, each active device (tl072, lm386) has some internal resistance so this adds in series and provides some (unknown) current limiting to the LED.  You could estimate/calculate this effective resistance by taking the output resistance of the tl072 plus that of the lm386.

I would suggest a series limiting resistor between the tl072 and LED to limit the current to whatever the LED's max is (typically 20ma) and ditch the lm386 for this application.

Ark Angel HFB

I'm jumping into this cold and have not read the thread yet...

Do say it I plan to read after posting if I find my answer I will remove this..

The Keeley comp was based on the Ross... how does this stack up to it?

"..So I hooked up the power and it was the greatest Radio I'd ever heard. Too bad I was trying to make a Tremolo..."

midwayfair

Quote from: Ark Angel HFB on July 17, 2013, 03:37:14 PM
The Keeley comp was based on the Ross... how does this stack up to it?



Pretty sure merlin touts the advantages in the first post. :/

Main things are quieter, feed forward, and more range of compression (both less and more).
My band, Midway Fair: www.midwayfair.org. Myself's music and things I make: www.jonpattonmusic.com. DIY pedal demos: www.youtube.com/jonspatton. PCBs of my Bearhug Compressor and Cardinal Harmonic Tremolo are available from http://www.1776effects.com!

MrStab

hmm, from what i grasp, the LM386 solution was not only at risk of being inaccurate, but borderline-unusable, Alex. thanks for the explanation, hopefully someone in future will persevere with using a comparator and heed that advice instead of just lazily resorting to a bare NPN like i did.

for the record, the NPN-powered dimming compression light i now have is pretty handy for something so simple! even seems to respond to Attack & Release. now i just need another compressor so i can switch between rhythm and lead...

Quote from: Ark Angel HFB on July 17, 2013, 03:37:14 PM
The Keeley comp was based on the Ross... how does this stack up to it?

i know it's not quite what you asked, but i replaced a pot for someone on a Ross clone soon after building this comp, and that just seemed ham-fisted and noisier in comparison to the ET. that's probably subjective.
Recovered guitar player.
Electronics manufacturer.

merlinb

#166
Quote from: MrStab on July 18, 2013, 10:34:18 AM
for the record, the NPN-powered dimming compression light i now have is pretty handy for something so simple!

I'm lost- what did you end up using?

I haven't tried it, but I would have thought this would work. Depends if the 4.5V buffer can supply the LED current or not . The 470R resistor might also need tweaking for the best brightness range. If anyone tries it, do let me know if it works!

MrStab

#167
Quote from: merlinb on July 18, 2013, 11:16:01 AM

I'm lost- what did you end up using?


if i understand the original schematic right (which i probably don't), i'm doing pretty much what you've just described. not sure if this can be considered an accurate indicator of the output itself, as there's still some stuff taking place after, but it does respond to the settings used in what seems to be a surprisingly functional manner. only difference in my setup is that i'm using 9V+ separately instead of vbias, because i'm lazy

off-topic kinda, but as i'm pressed for real-estate in this enclosure, i'm thinking of just making the pedal always-on and using the former bypass DPDT to switch a resistor in & out of the Release pot wire to alleviate a lead/rhythm disparity i noticed at band practice last night. i'm thinking of putting something between 220k-470k connecting to lug 1 of the Release pot. not sure if i should replace the pot with a 500k or not, else the total resistance could exceed 1M. would that be bad?
Recovered guitar player.
Electronics manufacturer.

merlinb

Quote from: MrStab on July 18, 2013, 11:54:55 AM
i'm thinking of putting something between 220k-470k connecting to lug 1 of the Release pot. not sure if i should replace the pot with a 500k or not, else the total resistance could exceed 1M. would that be bad?
Should be fine.

MrStab

works! i put 320K on one path for handy lead switching. might need a little bit more, but i was hesitant to jump straight to 470k. got an LED ground connected to the spare pins of the DPDT. thanks for the reassurance, Merlin.
Recovered guitar player.
Electronics manufacturer.

Ark Angel HFB

Quote from: MrStab on July 18, 2013, 10:34:18 AM
i know it's not quite what you asked, but i replaced a pot for someone on a Ross clone soon after building this comp, and that just seemed ham-fisted and noisier in comparison to the ET. that's probably subjective.

Naa man that is exactly what I needed to hear. The first post talks about the dynacomp not the Keeley and well... yeah.

I guess I'll give this one a shot... >_> too bad I ordered a bunch of chips for teh Keeley already... looks like I'll take a day and churn out some clones for buddies at cost so I can try this one.
"..So I hooked up the power and it was the greatest Radio I'd ever heard. Too bad I was trying to make a Tremolo..."

MrStab

Quote from: Ark Angel HFB on July 18, 2013, 11:49:09 PM

I guess I'll give this one a shot... >_> too bad I ordered a bunch of chips for teh Keeley already...

if you mean the CA3080, then you're in luck. this layout uses that instead of the LM13700: http://tagboardeffects.blogspot.co.uk/2012/11/engineers-thumb.html, but you could probably use a daughterboard to use it on another/your own layout. Engineer's Thumb only uses one half of the 13700 anyway.
Recovered guitar player.
Electronics manufacturer.

Ark Angel HFB

#172
Quote from: MrStab on July 19, 2013, 01:23:00 AM
Quote from: Ark Angel HFB on July 18, 2013, 11:49:09 PM

I guess I'll give this one a shot... >_> too bad I ordered a bunch of chips for teh Keeley already...

if you mean the CA3080, then you're in luck. this layout uses that instead of the LM13700: http://tagboardeffects.blogspot.co.uk/2012/11/engineers-thumb.html, but you could probably use a daughterboard to use it on another/your own layout. Engineer's Thumb only uses one half of the 13700 anyway.

Yes I'd seen that... To be honest I thought that the ET was a Keeley clone at first with a few tweaks... a bad assumption I'm learning... Still since I already have the circuit for one Keeley done... I think I'll box it up and Box up the ET and do an honest vid showing off both side by side as a fair comp for anyone else. ^_^

Also I'll redo the Vero and try and make it a bit more space saving and add in the extra control options... for those of us that find etch to be weird scary voodoo magic of the Gods...

A fire ant just bit my foot... @#%$ @#$% @#%$.
"..So I hooked up the power and it was the greatest Radio I'd ever heard. Too bad I was trying to make a Tremolo..."

merlinb

Quote from: Ark Angel HFB on July 18, 2013, 11:49:09 PM
Naa man that is exactly what I needed to hear. The first post talks about the dynacomp not the Keeley and well... yeah.
The Keeley and Ross are identical to the original Dynacomp. They're all the same pedal. The Keeley and Ross are just knock-offs with a couple of resistors changed, which apparently justifies the higher prices and general hype. (OK, the Keeley also has an attack control added, big wow... ::))

When I designed the ET, the whole point was to make sure it could do everything the Dyna/ROss/Keeley can do, and more and better and cheaper! :)

Quote
I guess I'll give this one a shot... >_> too bad I ordered a bunch of chips for the Keeley already... looks like I'll take a day and churn out some clones for buddies at cost so I can try this one.
JMK does a PCB for the ET that uses a 3080.
http://www.jmkpcbs.com/JMK_PCBs/Store

psychedelicfish

I've made a PCB layout for the ET, using a bipolar supply. I will use this in a multi-effects box that I will eventually build. I've checked it for errors myself, but if you can spot any, please point them out.
Thanks,
Edward
If at first you don't succeed... use bigger transistors!

MrStab

Quote from: merlinb on July 19, 2013, 03:50:40 AM
When I designed the ET, the whole point was to make sure it could do everything the Dyna/ROss/Keeley can do, and more and better and cheaper! :)


word. my omission of any kind of on/off switch says it all. lol
Recovered guitar player.
Electronics manufacturer.

MrStab

#176
getting that weird ticking this time when the effect is on, the compression level light flickers along with it. seems variable, as i lightly kicked it a coupla times and it disappeared lol. can't see how it'd be a short, though. there's a 470uF to ground after the V+ splits off to the NPN from past experimentation (covered the damn thing in hot glue so can't easily get a 100nF from ground), maybe i should try decoupling the main circuit itself. using an adaptor with at least 1A current available.

Edit: decoupling the board itself did nothing.
Recovered guitar player.
Electronics manufacturer.

Ark Angel HFB

Quote from: merlinb on July 19, 2013, 03:50:40 AM
Which apparently justifies the higher prices and general hype.
Quote

I'm Kinda a one man pedal company making stuff for all my buddies and selling the extra on ebay to make rent... and how the big companies mark up there stuff kinda ticks me off... When you know they buy in bulk... and you know they make them fast... The total pedal cost in parts can't be more than $20 and yet they are selling them for $260... I mean it makes the bank account happy but damn... I'd rather work a little harder price my stuff at like $70 or $80, and have dudes with three of my pedals on their boards than just one...

Maybe I'm crazy it is just that I'm sick of the stupid notion you have to spend a mounts wages to buy two pedals... great tone is cheap... Companies are just holding it hostage...

Quote from: merlinb on July 19, 2013, 03:50:40 AM
When I designed the ET, the whole point was to make sure it could do everything the Dyna/ROss/Keeley can do, and more and better and cheaper! :)
Quote

Thanks for setting me straight man... I'm still learning this stuff. I can design veros from circuits easy, but I'm not on your guys level of knowledge yet about which pedals are clones of others. Still... I'm starting to get the same hype feeling from this as I did with the Valvecaster... a really great pedal that everyone on here can get behind and make the new standard.

I think this project will be the one that inspires me to take the step into etching... what bold new world lays ahead. XD
"..So I hooked up the power and it was the greatest Radio I'd ever heard. Too bad I was trying to make a Tremolo..."

ketterman

I understand doing things like this for the love of the game, but Merlin, you should be getting paid for this. I bought both of your books. Perhaps that counts for something.
I breadboarded the project last night and it definitely delivers on the 'better compressor' description!
Now to decide which options go into the final enclosure...

tpoolan

Hi,
I've bread boarded the original circuit and I think this is a great compressor. So I decided to build your second version with a VU meter exactly as you describe in the GTI documentation, Merlin.

The compressor itself works great in all areas but I can't get any kind of response from the 10 segment LED. It flickers when I plug in the power, which is encouraging.

Any ideas where I should start looking? I know the VU signal is taken from Q1 but I'm pretty lost after that.