(https://coastaudio.files.wordpress.com/2013/10/schematic-really-cheap-compressor.png)
Anyone built this "Really Cheap Compressor"? Looking very simple.
Just wonder how well is it doing. How to add a threshold/attack/release if necessary?
I built it and I'm very happy with it. I leave it on almost all the time. Lots of squish available at high settings if you want it, very subtle at low settings if you prefer that. I found it a bit dark, stock, so I added a bright mode on a switch. I also increased the coupling caps a bit so I could use it on bass as well.
I got some distortion at high settings at first. I pulled out the ldr I had scavenged from some home appliance and replaced it one from the Radio Shack assortment pack, and the distortion went away.
Quote from: induction on April 29, 2015, 07:20:21 AM
I built it and I'm very happy with it. I leave it on almost all the time. Lots of squish available at high settings if you want it, very subtle at low settings if you prefer that. I found it a bit dark, stock, so I added a bright mode on a switch. I also increased the coupling caps a bit so I could use it on bass as well.
I got some distortion at high settings at first. I pulled out the ldr I had scavenged from some home appliance and replaced it one from the Radio Shack assortment pack, and the distortion went away.
Thanks for your info. How do you add the bright switch and do the mod for bass? I am surprised that it is doing well when it is such simple.
I built one, but found it had to many parts, so build this, which has been on my board ever since: http://www.diystompboxes.com/smfforum/index.php?topic=101118.0
In either case, be prepared to audion different LDRs until you find one that works satisfactorily. If they generally have too high resistance, you will get op-amp clipping, and if they have too low resistance, you will get little compression and output. The LDR response time, which can be quite variable, also sets the attack and decay times.
Quote from: samhay on April 29, 2015, 07:53:24 AM
I built one, but found it had to many parts, so build this, which has been on my board ever since: http://www.diystompboxes.com/smfforum/index.php?topic=101118.0
In either case, be prepared to audion different LDRs until you find one that works satisfactorily. If they generally have too high resistance, you will get op-amp clipping, and if they have too low resistance, you will get little compression and output. The LDR response time, which can be quite variable, also sets the attack and decay times.
WOW This one have even fewer parts. So what will be the major different on the performance between two?
I think the Flatline is the superior design: Less likely to distort (dry is not fixed gain), higher input impedance and uses the non-inverting input, more stable LED driver, and you can easily add the threshold control like in the RCC. Still fits in a 1590A easily.
I built it- love it. 3 circuit mods: 470pF paralleling 220k input R, 10nF across the vol pot, and 2K2 output R on vol-2.
I tried various LED-photoresistor combinations, different colors, single LED etc, and wound up using 3mm diffused green with a GL5537 (Tayda). I filed a flat on each side of the LEDs, superglued together, and used heat shrink tubing to make a nice compact package. Ends are sealed with black RTV.
(http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2200_zpsupmlsidq.jpg) (http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2171_zpsrzatafc1.jpg) (http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2179_zpsa0m0sdyr.jpg)
And it fits in a small box ;D
(http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2192_zps43e4d820.jpg)
Quote from: nguitar12 on April 29, 2015, 07:31:21 AM
How do you add the bright switch and do the mod for bass? I am surprised that it is doing well when it is such simple.
Bass mod:
I used a 2.2uF cap for coupling to the leds (instead of 1uF), so it would work well for bass. Doubling the 15nF input and 100nF coupling caps didn't seem to have any effect, so I left them alone.
Bright mod:
470pF and 10nF caps in parallel with the 220k input resistor and the level pot, respectively. Those are controlled by a switch. I added a 2k2 resistor before the output permanently. This mod was suggested by mictester, the designer of the circuit.
>WOW This one have even fewer parts. So what will be the major different on the performance between two?
The design is somewhere in-between the 'Really Cheap' and 'Flateline' compressors. A major advantage over the former is that if you turn the compression down, you turn the gain down and thus turn the noise-floor down. The Flatline has a more complete envelope detector than the other designs and is a good choice too.
They are all fairly simple circuits, so I suggest you breadboard all 3 and see if you have a preference.
Quote from: Cozybuilder on April 29, 2015, 09:33:09 AM
I built it- love it. 3 circuit mods: 470pF paralleling 220k input R, 10nF across the vol pot, and 2K2 output R on vol-2.
I tried various LED-photoresistor combinations, different colors, single LED etc, and wound up using 3mm diffused green with a GL5537 (Tayda). I filed a flat on each side of the LEDs, superglued together, and used heat shrink tubing to make a nice compact package. Ends are sealed with black RTV.
(http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2200_zpsupmlsidq.jpg) (http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2171_zpsrzatafc1.jpg) (http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2179_zpsa0m0sdyr.jpg)
And it fits in a small box ;D
(http://i1362.photobucket.com/albums/r688/russrutledge/Really%20Cheap%20Compressor/DSCN2192_zps43e4d820.jpg)
Nice job. Very Clean build :icon_wink:
Quote from: induction on April 29, 2015, 09:46:06 AM
Quote from: nguitar12 on April 29, 2015, 07:31:21 AM
How do you add the bright switch and do the mod for bass? I am surprised that it is doing well when it is such simple.
Bass mod:
I used a 2.2uF cap for coupling to the leds (instead of 1uF), so it would work well for bass. Doubling the 15nF input and 100nF coupling caps didn't seem to have any effect, so I left them alone.
Bright mod:
470pF and 10nF caps in parallel with the 220k input resistor and the level pot, respectively. Those are controlled by a switch. I added a 2k2 resistor before the output permanently. This mod was suggested by mictester, the designer of the circuit.
Thanks. I will definitely apply these mods if I built mine.
Quote from: samhay on April 29, 2015, 10:44:39 AM
>WOW This one have even fewer parts. So what will be the major different on the performance between two?
The design is somewhere in-between the 'Really Cheap' and 'Flateline' compressors. A major advantage over the former is that if you turn the compression down, you turn the gain down and thus turn the noise-floor down. The Flatline has a more complete envelope detector than the other designs and is a good choice too.
They are all fairly simple circuits, so I suggest you breadboard all 3 and see if you have a preference.
That's is good idea. So bad that I don't know how to use the bread board. Every times I test out a circuit I have to layout in eagle and make the board. It is a time consuming job for me.
If you can learn how to use Eagle, you can learn how to use a breadboard.
If you need a push, you might find these helpful:
http://diy.smallbearelec.com/HowTos/Breadboarding/BreadboardIntro.htm
http://diy.smallbearelec.com/HowTos/BreadboardSiFF/BreadboardSiFF.htm
Quote from: samhay on April 29, 2015, 02:14:08 PM
If you can learn how to use Eagle, you can learn how to use a breadboard.
+10
Breadboarding is way easier to learn than Eagle.
Built it twice for a multifx. It's ok, it really needs the bright mod though. Stock it sounded muffled.
I wanna build this but I'm confused. There is no mention of a VB or VR or whatever, the junction of the 47uF and 33k even says +9v.
Do I need to make a voltage divider circuit for this? And if so, should I tie the 47u/33k to VB?
Or should I just build Sam Hay "Another Optical" but then with that one, also green LEDs or different color?
Thanks!
Both op-amps have a half-rail voltage applied to the non-inverting inputs. Follow the path from the +9V you mentioned. It goes through two 33k resistors before it reaches ground. The non-inverting inputs are connected to the junction between those resistors: that's Vref. It's not necessary to label it if the schematic shows those connections explicitly.
The color of the led's is chosen to match the optimum frequency that the ldr responds to. You can test your ldr specifically if you like, or just use green or yellow. In my experience the color isn't crucial in this circuit, but let your ears be the judge.
Thanks induction, I see now. So I will build as it looks in the schematic with the exception of adding my standard +9v input like this:
(http://i67.tinypic.com/2rhtzwz.png)
I've built many of the RCC, it's a great simple circuit. It does need the bright mod. I personally use a 5-10p in the feedback loop and 100p in parallel with the 220k. also to avoid any LED consuction noise I strap a 10n across the LEDs. I roll my own LDRs and measure the light and dark parameters and sort accordingly. It's a wonderfully simple design that works very well.
Thanks Groovenut, I was planning on using the aforementioned mods but I will try yours as well. I rolled a diffused green vactrol w/LDR from Tayda and it's on the BB now, sounding really good in the stock form, so now to tinker with the mods and see if it gets even better. Thanks!
I've been playing around for awhile now with the mods and stuff but one quirk I am getting stock or otherwise, I feel like there is not enough compression on a certain area of notes on my guitar. Specifically, between the 3rd and 7th frets of the low E and A strings, so low G up to second E. It creates a strange unevenness when playing single note runs in that zone. Lots of nice squish until that zone and then volume boost on those notes with less comp.
Anybody know why that might be?
Quote from: Ben Lyman on March 14, 2016, 07:34:12 PM
I've been playing around for awhile now with the mods and stuff but one quirk I am getting stock or otherwise, I feel like there is not enough compression on a certain area of notes on my guitar. Specifically, between the 3rd and 7th frets of the low E and A strings, so low G up to second E. It creates a strange unevenness when playing single note runs in that zone. Lots of nice squish until that zone and then volume boost on those notes with less comp.
Anybody know why that might be?
If it were all the notes on the low E up to the 7th, I would say the cap feeding the second opamp needed to be bigger. However, if it's between the 3rd and 7th on the low E, I haven't a clue what might cause that within this compressor circuit. It sounds as if your guitar has a low output range in that area and it's not triggering the compressor the same way the rest of the instrument is. That's a head scratcher.....
Okay, thanks. I might be wrong about the exact notes, I will try to examine it closer to see if those last few low notes are behaving the same way.
It might be my guitar, I put the pickups in myself a long time ago and maybe I tried to lower the bass side to keep a sort of trebley sound, I dunno. They are DiMarzio Area pickups and I seem to recall they had a higher output and more bass than the original Van Zandt and Seymour Duncan single coils.
> between the 3rd and 7th frets of the low E and A strings
And the open strings?
(As Lawrence says) The 1uFd driving the LEDs and 1K looks skimpy. Assuming this network bottoms out at 1K, 1uFd into it is blocking bass below 170Hz. Which is essentially the bottom octave of guitar. I will not be shocked if this droop (failure to compress) does not become obvious for another half-octave, the range you are pointing to.
Tack 10uFd across the 1uFd and take a run to the bottom. Is that different?
Yes, thanks, upon further noodling and wanking, it has come to my attention that the loss of compression is occurring from the 7th fret down to the open low E & A strings. I will see what I can do about it with that 10uF tomorrow at some point.
Other than that, this comp is very nice, sounds great with my Electra distortion, the bright mod is great for clean and I have some other mods of my own going on right now on the BB. Instead of a 470pF across the 2k2 input r, I have a A100k pot in series with a 1nF for a variable cap. I'm not even sure if I've got it hooked up right and it is probably changing a lot of the intended plan but it's sounding really good. I don't really want more knobs but an internal tone trimmer might be useful.
For playing with distortion I think the 470pF mod needs a switch because, while brightness is good for jangly cleans and hot chicken pickin', it doesn't work so much with distortion if you want to shake the walls.
10uF across 1uF did the trick. 100x better, thanks.
So, this makes 11uF total and I can omit the 1uF as long as it sounds fine without it, right?
I'm not sure if you suggested "tack across" for a specific reason. Its still on the BB.
> if you suggested "tack across" for a specific reason
So if I was totally wrong the mod could be UN-done without more surgery.
Now that we have a clue, you can salvage the 1uFd and put 10uFd in right.
> on the BB.
Ah, missed that, no heavy work to make the change.
Okay, thanks again PRR. One more thing I noticed, as the notes decay there is a very quiet fuzzy crackle, almost like static or a wax paper and comb kazoo. I was wondering if it might be the LEDs "consuction noise" that groovenut mentioned, so I tried a lot of different caps across them but the kazoo persists no matter what. It only seems to happen when the sustain control is from 2:00 (60%ish) up to full CCW. I have bypassed the effect to test my own amp/guitar for this noise but it is definitely only from the comp
Power rail bypass as CLOSE to the chip and LED return as you can manage.
The LED driver must distort in heavy action. Large spikes in the power supply. These could be sneaking back via power rail or round-the-BB ground path.
Or may be something else.....
Okay thanks PRR. I've tried to make a new schematic in a way that I can more easily wrap my brain around, including part numbers. I hope it's all accurate, it should be almost the same as the original version, except I've included all cozy builder's mods and PRR's 10uF for my C9.
:icon_question: This is what's on the BB now, is "power rail bypass" my C1?
:icon_question: Is this an okay way to go with the bright switch or should I worry about it popping?
(http://i63.tinypic.com/13z4r45.png)
Edit: correct 220k input r, correct sus and lvl pot values
Quote from: Ben Lyman on March 16, 2016, 01:28:56 PM
Okay thanks PRR. I've tried to make a new schematic in a way that I can more easily wrap my brain around, including part numbers. I hope it's all accurate, it should be almost the same as the original version, except I've included all cozy builder's mods and PRR's 10uF for my C9.
:icon_question: This is what's on the BB now, is "power rail bypass" my C1?
:icon_question: Is this an okay way to go with the bright switch or should I worry about it popping?
(http://i64.tinypic.com/2hxssau.png)
Power rail bypass will be C2 actually and it's physical location needs to be as close to the power pin of the opamp as possible for best results. you could do 100n as well there.
The inverting input of the opamp is a virtual ground. I dont think the bright switch will pop, but all thinkgs being what they are, it may ;)
For distortion/fizz in this circuit it's usually one of two things. LED conduction noise, in which case placing a 10n cap in parallel with the diodes usually fixes this, or rail clipping the first opamp because of too much gain/high LDR dark spec, in which case you could try lowering the 2M2 resistor to 1M and see if it results in the changes you're after.
Have you tried a bigger C7 Ben ?
While its on the breadboard I'd try anything upto 470pf just to see if it improves the crackle.
It might work !!!!!!!
Quote from: Ben Lyman on March 16, 2016, 01:28:56 PM
(http://i64.tinypic.com/2hxssau.png)
Just curious if you meant to replace R6 with 2K2 rather than 220K? I didn't see anything written on this (maybe I just missed it?)
@kipper4, okay, will do, I haven't had a chance to really delve deep into the mystery of the kazoo player. It persisted when I moved one leg of C2 right onto IC+ (pin8) but I *THINK* it went away when I moved the other leg of C2 right onto IC- (pin5). I'll report back later after I've had some more time with it.
@cozybuilder, good catch, my bad :icon_redface: thanks. I will update the schematic
> C2 right onto IC-
Yeah, like that.
hellow and good day sirs
ive built this but i notice my led barely lit up. i cupped my hand over it and it did lit up. im thinking maybe r8 1k might be too high a value. and considering 470ohms. i havenot actually tried it yet with a guitar. Cause i populated it at the office and did not have a guitar here. but signal seems to be getting through.
Also i notice alot omiting c1 will try that if its too dark.
swapped the 1k for a 470 works great for me.
I played with it quite a bit. with the circuit on my bench i noticed that bypass was louder than with the effect.
as i strum a chord or hit a note and then i tried to cup my hand over the ldr and led. it would get louder if i put totaly coverd it it would raise the volume so much it would break. This as sustain on my guitar was dying out. So i know its work and had fun with it ;D ;D ;D ;D
i used aluminum tape. the kind used for refrigeration. to block out ambient light between the ldr and led and wow what a great sounding compressor.
i think 1k resistor needs to be a 1k trimmer. so you can adjust it depending on the led your using. Initially i had typical red opaque led there then green and finally settle on yellow as it was brighter to me.
Really happy with the compressor.
I've been playing with a moded version of this, anyone know how an attack control can be added? Appears to be tied into the "sustain" amp and the led+ldr type (threshold?), but I'm trying to figure out how to impliment a seperate controlOr is it not possible with an optical?
> Or is it not possible with an optical?
It is incredibly difficult. To be simple, change the LDR, some are faster than others. This won't be an on-stage change because every LDR is different in other ways. To be complicated, add rectifier and DC filter, plus NFB around LDR...... and now you exceed the complexity of simple non-LDR limiters, and should go with one of them.
> To be complicated, add rectifier and DC filter, plus NFB around LDR
Is that what the Flatline does? I was hoping for a simple (cap-resistor) solution. Well, I like the simplicity of this one and it works well enough. I'll find a led/ldr pair that works and go for it.
> Is that what the Flatline does?
Uh, almost. Yes it rectifies and filters, but all at POWER levels (LED power), so your options are very limited.
JFET limiters tend to have tweakable sidechains. (JFET gate power is nearly zero, so more tricks are practical.)
Biggest problem I find with Really Cheap Comp is that the attenuated gain is around unity. The LDR will fall below 220k and then the side chain will adjust. I added a series resistor to the LDR to give some boost. Then it needed more gain to adjust the ratio so I changed to smaller resistors. At that point I changed to a noninverting stage to raise the input impedance which gave it better clarity. No need for the treble mod. At that point it looks more like this one http://www.diystompboxes.com/smfforum/index.php?topic=89110.msg754675#msg754675 (http://www.diystompboxes.com/smfforum/index.php?topic=89110.msg754675#msg754675).
I've been playing with the comp ratio with various ranges of resistors parallel and in series with the LDR. With some combinations for a lower ratio I get a clicking too long attack and with very high ratio I get a short blunt (nosey) attack. I plan to have ratio switching between a light comp to a heavy limit. As someone (PRR) said in another thread that the LDR will drop fast at first and then slower near the light end, so I can use that to my advantage to find some compromise attack.
Ben Lyman - seeing as me and kipper have been chasing non-existant threads of yours, have you done a demo of this compressor? and, what LDR (as usual) and what IC type did you use, please? I have one on the bb at the mo, finally found a suitable ldr.
[I also got a noise, maybe your kazoo (wahzoo?), but only on the open G string, and it sounds like something loose, or someone holding a bit of paper against the string. but only for a section of the decay. odd.]
Ya Duck, that's the kazoo noise alright. It went away when I boxed it up so probably just breadboard noise, nothing to worry about. When planning your layout, place C2 (my schematic) right next to IC pin4 if possible.
LDR is Tayda's Waitrony 1M dark and IC is a TL072
I drew it up like this so I could understand it better. PRR suggested a 10uF for my C9 instead of 1uF, so you might try that on the BB, see what happens on your low E string.
(http://i63.tinypic.com/13z4r45.png)
hrmm, I tracked the kazoo down, I thimk. it correxponds, if that's a real word, to the U2 (lyman dia) clipping. the BB I thort sounded good yester, sounds like carp today, just the loose tissue kazoo, all everywhere. I tacked a 470pF across R10//R11, and it seemed to help, but it's now off the BB.
I also thought I noticed that the compression indicator led (not shown lyman dia) was flashing brighter when the R12 (lyman, again) was increased from 1k to 1k5.
[Coming to this topic a bit late!] I was thinking of building this compressor but was first trying to understand how it works. The op-amps are connected in a standard "virtual ground" configuration so my understanding is that, with no input the outputs from them should float at half the supply voltage. When an AC audio signal is applied to the input, their outputs fluctuate about that mean 4.5 volts. If so, then the output of the second op-amp is always positive with respect to ground and so one of the leds should always be on and the other should never light at all. I'm obviously missing something....can anyone put me right please?
C9 blocks the DC standing at U2 outpin [Lyman's dia]. so the leds get no volts until some AC comes along, positive swing turns one led one, negatives turn the other on.
Many thanks for the reply. Yes that makes sense. It just seems very counter-intuitive that, with a single ended supply, you end up with the leds being driven by a voltage that swings above and below zero volts! I'm going to have to build the thing and convince myself.
> seems very counter-intuitive that, with a single ended supply, you end up with the leds being driven by a voltage that swings above and below zero volts!
You want the audio output of a box to swing both ways around zero volts.
In 9V-world we typically swing 1V to 8V, then cap-couple the output referenced to zero volts.
The LEDs (two for both-ways action) are driven the same way.
Hello,
I successfully built the Flatline using the (and a ready-made vactrol, as advised on the Madbean Afterline instructions).
I added a sensitivity (which is really great imho), and an attack knob.
Overall, it is a nice compressor, quite transparent and versatile, great for invisible compression.
I would have 3 questions :
1- Is there any component/value to change to get a HPF sidechain in the compressor path ?
I would like to remove the bass (say around 100-150Hz) from triggering the compressor circuit, without removing them from the "audible sound". I am thinking about a switch with the stock value (which is great) on one side, and a "HPF sidechain" on the other side of the switch... but i don't know which component value i should change to reach this effect :icon_confused:
2- I have also tried adding an indicator led by plugging it into A3 and C3 (from the Guitar FX Layout veroboard) : the led blinks when the compression is stronger, so it gives a rough estimation of the compression occuring, but i noticed that it also some very nasty crackles as soon as i don't have the sustain set up to the maximum position.
Is there any way to keep an indicator led, but not having it modifying the processed sound ?
3- Also, given that i used 25V capacitors, can it run @ 18V or should i better stick to 9V?
looking at the Lyman dia again [where is Ben these days?], C8 tailors the response of the led drivers. reduce that to push the response f higher.
add a superbright led and a CLR, say 10k, between C9 (-) and ground, it should follow at least half the envelope signal.
18V on 25V caps? knock yourself out, what could possibly go wrong?
Hey ! I build one of this RCC.... i have de problem that when the guitar signal fade out, and de gain in the compressor is higher, I start to liste a noise.. like a fizz, hiss, i don't know how to describe it :sssgsggsggGGGSGGSGSG
Of course I don't here that when I play. But when I stop and the compressor makes it magic and increase de gain for tje lower signal amplitud, the noise app.
Is this normal?
could be, its expected behavior, you could add a noise gate, that takes out everything below a certain threshold, and hopefully the hiss as well.
cheers, Iain
> describe it :sssgsggsggGGGSGGSGSG
Turn it down. The idle gain should NOT be so high it screams in the quiet parts.