Very simple compressor. Works well IMO

[brett]

Hi
RE: signal loss when maxing the controls
Compress solution: Put a 4.7k resistor between the compress pot and output (pin5) of the 386.  That'll limit compression to a bit less than 2/3 of its current range.  Also, it might be worth trying an LED instead of the 1N4148.  That will "lose" an extra 1.5 V and might improve things by increasing the ratio of signal to the offset voltage. 
Release solution:  A 100k resistor between the release pot and ground.  That'll still allow a 100k to 1100k range of resistance.

I think this project has been a great team effort.
cheers

[mac]

IN summary, the MOSFET helps keep this project simple, even tho they are a pain in the butt.
cheers

That's way I asked. 


mac

[Barcode80]

sorry for the lamer question, but i'm confused. did we settle on a schem for this thing? have we posted it? or are we still testing things...?

EDIT:nevermind. i'm an idiot. found the verified layout.

[Barcode80]

brett is the layout in your gallery updated to the solutions above?

[puzzle87]

This is discrete analog of Aussie compressor.

Enjoy

[puzzle87]

One big minus of these schematics (and similar) – supply voltage 9V must be stabilized.
If not – any little deviation voltage do violence to compressor work.
Below - compressor schematic without this defect (able to work from 6V to 12V supply voltage).

Sorry my English

[Noplasticrobots]

Crap, I don't have that many JFETs so I'll have to try the discrete at a later date.

As for this circuit, I don't recommend building it because it will haunt you. I had a dream about it last night. It had something to do with jumping wires under the LM386 to turn it into a transistor instead of an IC. You know how dreams are...

"It's an IC, but it's a transistor because of the jumped wires."
"Oh, that makes perfect sense!"

I'll second what Brett said, great team effort on this circuit, everyone! 

[puzzle87]

Sorry, previous my schematic not correct. See changes below:

Also added compression control.

[PNG123]

Any one could take some pictures whilst building it for newbies like me please 

[brett]

Hi
The comment about using a stable supply is right.  A fresh battery would also work.  Or a 7809 at the V+input. 
Otherwise, every time the voltage drops a couple of tenths of a volt, the compress control needs to be turned up a bit.
I use a regulated supply, and never need to adjust the control.
cheers

[markm]

Hi guys,
I added a Layout for the EasyComp schem above.
It's not verified, feel free to advise or point out errors.
I haven't even had the chance to double check it against the schem yet so, I myself don't know if it correct at this point. 
http://aronnelson.com/gallery/album15

[markm]

BTW,
There is also a perf version in there too!
If anyone can verify this layout before I get to it, please let me know 

[Bagge]

I took a shot at stripboard layout for easycomp.

[brett]

This discrete compressor might be worth a look.
Works on a similar principle to the AussieMart, but instead of the 386 it uses a gain stage something like the Big Muff output stage. 
In some ways it might be better than the AussieMart.
It isn't sensitive to voltage change in the battery/supply.
I haven't messed with it much, but it works OK.
If you are using hot pickups or a booster, you can reduce the gain and prevent distortion by increasing the 100 ohm resistor to a 220.  Or change it to a 1k pot wired as a variable resistor for variable gain.  That might be handy.
http://www.diystompboxes.com/smfforum/index.php?topic=50248.0
cheers

[psw]

Hi guys...great thread, great project...

Can't make one at the moment to experiment but this circuit has been raised a few times over at the Sustainer Ideas Thread at Project Guitar for possible modification into an improved sustainer circuit...
http://projectguitar.ibforums.com/index.php?showtopic=7512&pid=293444&st=1935&#entry293444

Presently we typically use a preamp and LM386 to drive a "thin" coil under the strings to create infinite sustain. In another thread here, member col has been looking into various compressor like circuits to even out the response of different strings and action in relation to fretting position...
http://www.diystompboxes.com/smfforum/index.php?topic=49595.new;topicseen#new

Anyway...the question is, could this circuit be modified to be used to provide compression but provide enough output to drive a speaker (a la the ruby) or, in our case, a driving coil! Obviously things like the output cap (I use a 100uF) and the gain between pins 1 & 8 could be adjusted and some more components may be required for stability in a high gain mode...

Any thoughts? pete

[brett]

Hi
It'll work "as is", with a couple of provisos.
First, the impedance of what it drives must be 16 ohms or higher.  Otherwise you'll exceed the power rating of the 386.  8 Ohms is too low.  (e.g. 3V out @ 8ohms is a touch more than 1W, which is about double what the 386 is rated for).
The gain and everything else could stay the same.  Your impedance issue is releated to power and current ratings for the 386 can produce, not voltage per se.

If for some reason it is essential to drive a load of less than 16 ohms, you could use a buffer.  Any of the usual transistor buffers are suitable, but use a heavy-duty transistor such as an MJE3055.  It will drive loads as low as a couple of ohms before the smoke comes out.
cheers   

[psw]

First, the impedance of what it drives must be 16 ohms or higher.

Really! I thought that the LM386 was designed to drive an 8 ohm load and it seems to do that fine...so does this circuit some how change that impedance...

So, how high can you go, how about 14,000 ohms?

Also...do you know the power consumption of the circuit as is, perhaps it could be used successfully as a compressor/preamp running another LM386 or similar for the poweroutput?

Anyway, thanks for your prompt reply, with such a simple circuit, it wouldn't hurt to make it even if I did blow it up. Also, 16 ohms would not be impossible to make with some creative design changes to the driver. My driver coil is only 3mm thick and it is this thinness that provides the speed and EMI reduction necessary. Currently people are developing dual coil drivers so the heart of the system is still being developed.

For sure we will be playing with this circuit at some point to see if it does have an application to the kind of things we are doing...thanks for your reply and ideas, and thanks to everyone for contributing and developing this idea further...good stuff... pete

[brett]

Hi

Really! I thought that the LM386 was designed to drive an 8 ohm load and it seems to do that fine...so does this circuit some how change that impedance...

Oops!    The 386 is rated at 1.25 W.  That's about double what I was thinking it could handle, and plenty for the compressor to drive an 8 ohm load.  But it would be sailing close to the wind with 4 ohms with the voltages involved.
cheers

[psw]

Ok...thanks for that Brett...not to hijack the thread or anything, but you guys obviously have been playing around with compressor concepts...

Over at sustainerland...http://projectguitar.ibforums.com/index.php?showtopic=7512&pid=293498&st=1935&#entry293498 the 2 c's (clol and curtis have been to-ing and fro-ing over the question of different types of compressor in the sustainer application lately...

Now, I seem to get adequate results without compression and I can't say that conventional compression techniques have made an appreciable improvement for me in the past...however...

Col...
The feed forward topology on the other hand doesn't worry about the level at its own output, instead it uses the level of the input signal as its reference.
for many applications this won't make any difference to the final outcome, but for us it does because for us the input level depends on the output level. The feed-forward compressor will keep trying to get the input level to be consistently at its threshold level - even if this means that its output is below or above the threshold....

Curtis...
Ahhh, yes, of course! I keep on forgetting that we're not compressing blindly to set an output - the string and driver form the signal source that gets fed back into the system via the pickup. The whole thing relies on a lot of nested feedback loops to function properly.

What the?!!?!!!

Anyway I kind of get it...yet....

Basically, this is what they are trying to achieve...

The idea is that we want the sustain level and response to be even no matter which string is played and which fret it is played on. There is an enourmous difference in response depending on the magnetic mass of the string and it's tension. Playing higher on the neck makes any string more responsive because it will be closer to the driver - this effect is magnified by a higher action.

And what they are looking for is a feed forward compressor design...

col...
So assuming that we both mean the same thing by 'feed-forward compressor' it has the potential to give us a much better string/fret balance than a common or garden feedback compressor circuit.

(Another way to think about it is that we need a feedback compressor, but the driver/strings/pickup combo is part of the compressors feedback line.)

got any good feed forward compressor schematics?

So...maybe there is a direct use for this type of circuit, or maybe they are on to something with this particular application...the sustainer is a strange thing to work with and there are a few variables. This variation on the typical 386 circuits (with suitable modifications to drive the load) may well make an improvement and is certainly a terrific idea...maybe not quite what they had in mind, but quite likely superior to the fetzer/ruby combinations currentsly in use.

If anyone was to visit the sustainer thread...please don't try and read it all...with approaching 2,000 posts and over 65K visits it would only confuse...mention of this circuit is on the page linked above and a few pages back details of col's present design currently installed in his guitar (using a LM13700) but this circuit is still under review.

Thanks for your input and I'll leave you guys to continue this terrific thread...maybe there are even more applications for this circuitry in FX design...keep thinking... pete

[psw]

g'DAY...

Have been experimenting with variations on this circuit for the sustainer application...won't work as is, but certainly got me thinking in new ways about the problem and getting some headway...so thanks for all those contributing to this very interesting thread...

Anyway...I think Brett mentioned it somewhere, this and compressors in general seem to be sensitive to power supply....given the higher gains I am using this is probably a lot worse (I use the LM386 as a Power Amp to driver a load with the sustainer), this seems to be particularly apparent.

Are there any really simple (ie small) ways of stabilising the supply so that fluctuations in voltage or current do not effect function...from a 9 volt battery...that doesn't draw much of that power away or further drain it? I don't want to loose power in the process if possible or shorten battery life!

Otherwise...have been inspired and helped by this to understand J and M FETS much better and transistors in general. Till now I have avoided discrete components favouring opamps for their lego like simplicity of constructing amp stages...also tricky to get some types. Speaking of which...is there a good site that gives substitutions?

Any advice welcome... pete