Sustainer Ideas

[psw]

Wow, what did happen to that PCB? It's not good to solder under the snow...

I think he just got over excited when he finished it!

Very funny...hmmm....I thought it looked kind of cute...the photo is obviously many times larger than actual size, and I think that much epoxy should stump would be reverse engineers for a bit. Would black have been better?

Actually, what I had envisioned was a perfectly clear coat...would look like glass...but it might defeat some of the purpose and display the "vero"...so not as pro looking. It is a little tricky as obviously that trim pot needs to be kept well clear of any epoxy that could effect it or glue it in place.

Here is a sneak peak of the control hardware I am making...

The tortoise and aluminium knob surround turns...the fake bolt (used elsewhere in this guitar) marks the position and makes it easier and more discrete. The right side shows the control with the knob in the up position...

Here is a pic of what makes it work...

So...two switch functions and a control pot...ideal for the sustainer which will work from this single control without modification or cluttering up the face of the guitar. They work so well that I am installing three matching versions so that a number of tricky wiring and other features can be controlled. In use though, it will look and play like a typical strat with three knobs and a 5 way selector...these switches will be used to change "modes" similar to the fender S-1 switch for instance, so the guitar will have a wide variety of different sounds available.

However I wire it, it is going to be tricky as I am detirmined not to enlarge the control cavity. I did modify the back of the trem cavity to make a secret easy to access battery compartment....

later...I feel some more snow coming on!

Pete

[Franky]

Impressive... Looks a bit like the SH-1 switch system yeah.. But why do you put the 2 switches together? That way you can both activate the sustainer and deactivate the neck pickup?

You should patent it anyway

[Maiden69]

Very nice Pete! I have been following this for a while! I think that the black epoxy will be better suited for your project. This is what I used when making small electronic parts for a gate opener job I had a long time ago. We had a kid that was a wiz on this electronic things, and we develop a timer, like the one in the link below. But with a trim pot as the timer. We had the pot sticking out on 1 corner inside a small plastic box, and had all wiring on the opposite side, then filled the box with black epoxy. I got the idea, once again, from the link below... in the back, all you can see on that timer is a smooth black epoxy face.

http://ecx.images-amazon.com/images/I/51Mq...9eL._SS500_.jpg

Edited November 11, 2007 by Maiden69

[Franky]

Hum did anyone here ever used an impedance-adapter transformer, to use a single-coil pickup as a driver? For example, with a 8R/8k transformer, it could work...

One can notice that there are small transformers on the Fernandes' sustainer curcuits, i'm going to check the impedance of the driver this weekend, to be sure..

Edited November 15, 2007 by Franky

[mrjstudios]

Nice job PSW!!!!

I thought that stuff was epoxy..... definitely use black or something really dark though. You can kind of see through the white, and makes it look like it had an accidental fall into paint

That switch also looks like it will be very usable as a 'controller' -- which is something I was worried about when I found out that you were putting all controls (that used to be 3 pots and 2 switches!!) into one unit. So I am confident that this thing will work in my guitar in the future.

Also, my sustainer guitar is ready to be painted (and this one is going out to a real artist, unlike my last one which was subjected to my spray painting [the red one]), so it will probably be ready to test out your unit by the time you send them out/sell them.

Good work,

-MRJ

[psw]

Sorry for not replying more often, but a lot on my plate and still working on this guitar of mine.

Have just done a bit of complex wiring, but still more to come...looking like a bowl of spaggetti at the moment but sounding good...

psw from guitar nuts 2 forum....

On the general wiring issue...the guitar is working with this kind of thing...

This gives you

PU1

PU1+PU2

PU2

PU2+PU3 Tap

PU3 Tap

Basically your typical parallel strat sounds...flick the "switch below the volume control, and you get...

PU1xPU3 Tap

PU1XPU2

PU2XPU3 Tap

PU1XPU2+PU3 Tap

PU3

(+=parallel, x=series)

So...series heaven, all great usable sounds. The neat thing is that they are pretty complementary so you can go from one to the other with the flick of that one switch. All the sounds are quite different and usable.

So, single coil strat sounds and series humbucker sounds at the flick of a secret switch below the volume control. I need to get something worked out for the other 4pdt and dpdt switches, a piezo with another push pull, and of course the sustainer....

I have done some preliminary work on the sustainer but these connectors are not really suitable so am intending to switch to flying leads from the circuit in future...too much space and not secure enough...

I have to get this wired up to ensure that the "product" has no fatal or otherwise flaws, some testing is still encouraging though, so onwards as time permits.

The wiring though has really opened up the sound of this guitar and this scheme is a terrific combination of sounds, superior in many ways to fenders own S-1 system.

The really neat thing is that in one mode it is pure SSS strat, and the JB split makes a good balance and convincing split sound in the tapped locations. In the other mode (series) you get more depth and power plus some interesting combinations in the middle that at times is tele-ish (neck and bridge tap) and other times gibson. All sound really cool with a fender-ish snap to everything, even a little quack in the series modes. Extremely quiet too. What is great is that the sounds between switch positions in all selections are complementary so switching modes takes you directly to another useful combination with a single switch.

The location and placement of the indicators, behind the knobs also worked out well, easy to turn and convenient, but not so that it will turn by accident...the position of the control is visible to the player easy enough and can be felt, there is a click as it turns too, so you will know that it has worked.

Anyway...will try and reply more frequently and address some of the recent posts in a little while...later... pete

[col]

looking good there...

4pdt and dpdt switches, a piezo with another push pull,

In that diagram what you really have is a 4 pole 4 throw rotary and a triple pole double throw switch. I guess what you are really doing is not quite the same and is 'secret' for now

Interestingly, Maplin in the UK now have available a 4 pole double throw toggle switch - just looks like any other chromed toggle from the outside - will have to get one of those to experiment with

cheers

Col

[Franky]

Hum i think Miyama also made a 4PDT, you can check it out there:

http://www.banzaieffects.com/Miyama-4PDT-o...e-pr-16271.html

[psw]

No...no secret...that is exactly what a super switch is a 4p5t rotary on it's side operated by the conventional lever (SW1)

I am using a pair of ultra miniature toggles that are activated together as in the previous photos above....so effectively a 4pdt switch. SW2 is drawn as a 4pdt but one "bank" is not used so is a 3pdt in effect...may be a use for all these hanging connections and this spare switch when trying for the next combination of 2x5 selections from the second 4pdt switch.

So...the lever switch will have 4 banks of 5 different sonic palettes...20 different tones. Many of them can be seasoned by the dpdt push pull switch on the tone control to give variations. Then add to that a piezo to any position and there is 40+ combinations on a relatively stock strat set up (no visible toggles, no drilling into the face of the guitar). The piezo is intended to be added by the push pull on the volume control. Add to this the sustainer, which works independantly, and you have a madly powerful variable range of tones, many of them unique, yet not confusing or difficult to navigate...the two 4pdt switches allow a quick move sideways across selections while the conventional selector lever allows conventional switching up and down the range of each. Kind of a poor mans version of the S-1 switching scheme but easier to use in many respects and twice the power. Only having wired one by this scheme, it is similar to the S-1 range as standard (kind of parallel and series banks), but some are superior in many regards and with these pickups balance really well with eachother...and extremely humm and buzz free!

4pdt toggles are available, I used one for my original sustainer on off switch. They are quite wide, but seem to hold up ok in use. I have been scouring DGB studio for wiring ideas (fantastic site) and noticed some really clever schemes that use a 4p3t switch (on-on-on) to do amazing combinations when combined with even a conventional selector or even more with a superswitch. I have never come across these myself. I can now see how I could rig up these dual toggles for a three position switch...bit complicated...

In case you missed it col, here is a pic...again...

See how the toggles are activated by a plate that twists the toggles across. The pot is also push pull...you activate the 4pdt by moving that disc via the bolt head handle about 5mm. Basically turning the toggles into a rotary action. The aluminum plates used to make it are shown in some photos...they work surprisingly well but have a fatal flaw for most applications...in order to hold the pots in position they need to be secured to the body from behind. I got the idea from car radio controls...but you can not get what I want suitably, so came up with this.

Now...here is a question...on the schematic...

Can anyone see how to make a dpdt switch change the coil tap of the HB from north to south coil when it is selected by the superswitch. I wired it up wrong at first so that the bridge-most coil was activated when tapped and this gives a very biting tele-like sound...switching it round gives a fuller rounder tone, strat like but with a little less edge which is impressive compared to some bridge pickup strat tones. It also gives a better, or at least different "quack" in the combined positions.

I would also be interested if anyone can see a way to add a noise canceling coil from the stacked SC's as these tapped positions is the only ones not fully humbucking on an otherwise extremely quiet guitar...not bad and a very convincing single coil tone (apparently the JB is renown for being one of the best split sounds around) but as this guitar has coils to spare, it would make an interesting feature.

This thing is turning into a real tone monster, the piezo and of course sustainer features will make for one very unique instrument and show how far, yet still useful and discrete, you can go with this kind of thing. To make things a little easier, I have wired everything to a buss strip of vero...everything connects to and from this, otherwise I would never be able to trace all the wires. It ensures a star ground and star hot so, bypassing all of this should be a bit easier when it comes to installing the sustainer in the thing...going to be a tight fit and require quite a bit of fiddling around a wire replacement to make it presentable.

pete

[col]

Can anyone see how to make a dpdt switch change the coil tap of the HB from north to south coil when it is selected by the superswitch.

Hmm - studying that schematic, it seems over the top. Very complicated and very very difficult to debug.

Many years ago, I built a switching box that let me try out every single combination available from two humbuckers. Treated as 4 single coils that could be on/off, in/out of phase, wired in series or parallel. That is a HUGE number of combinations. By the end of it, after many days of auditioning sounds taking notes and subjective and objective scrutiny, I ended up with only 5 sounds that I considered to be both distinct and of a high enough quality to be useful - all the others were either similar to one of these 5 but less good, or just not useful sounds. After many years of using that guitar with those 5 sounds wired in - it turned out that I only really used 3 of them! (neck humbucker(rarely), bridge humbucker and a very nice quacking combo - not like a strat quack, but equally usable)

I know that your setup is different having a true single coil in the neck and middle positions, however I find it hard to believe that all of the combinations that you have in your complex switching setup are truly useful from a musical perspective. I would spend some more time using and considering the sounds until you have some favourites, then be ruthless with the others... then when you have filtered all but the exceptional combos out, your circuit will be simpler, cheaper and easier to debug, and your guitar will be better for it as well. Im sure that with a superswitch, a 4PDT and a DPDT you can have any smaller set of sounds that you want combined with the sustainer... much better having a guitar with only excellent sounds than one that is like a mad scientists experiment

(Sort of the same with FX boxes and amps - you can tell the good ones because they are useful at most of the settings. Its really annoying when there are only a few good sounds, and you have to tweak loads of knobs all day to find them - when they should be hard wired to a switch with a volume control)

I think if I was to go back to messing about with trying to get sounds out of pickup combinations again, I would spend a lot less time playing with phase combos and series/parallel combos and a lot more time testing different caps in parallel with the coils - tweaking the resonant frequency with more precision to give a more useful tonal palette.

I'd probably end up going down the active preamp/filters route in the end anyway , for the short time that I was using my revision1 AGC circuit with the built in active preamp, the tone of my guitar was noticably better - more vibrant, more alive, just better.

anyway, I digress...

sometimes you have to take a step back and try to K.I.S.S.

cheers

Col

[Franky]

Hi, I've just finished to wire my driver (so far it's drying), the bobbin is about 3mm high, on a single-coil core (with polepieces, hope it wouldn't be a problem), with 0.2mm enameled copper wire, for a total impedance of 8.4R (but my multimeter is not very precise, it tends to add 0.4R to real resistor values (checked with a 1% 1R resistor)..

when it's dry, I solder the lugs and test!

hum, doesn't work quite good.. I can hear the sound of the magnetical feedback in the pickups, but I don't have more sustain.. Do you think it can come from the fact that I used a polepieced driver? I know the circuit is not very good, almost sure the problem comes from here..

Edited November 22, 2007 by Franky

[heggis]

Hi, I've just finished to wire my driver (so far it's drying), the bobbin is about 3mm high, on a single-coil core (with polepieces, hope it wouldn't be a problem), with 0.2mm enameled copper wire, for a total impedance of 8.4R (but my multimeter is not very precise, it tends to add 0.4R to real resistor values (checked with a 1% 1R resistor)..

when it's dry, I solder the lugs and test!

hum, doesn't work quite good.. I can hear the sound of the magnetical feedback in the pickups, but I don't have more sustain.. Do you think it can come from the fact that I used a polepieced driver? I know the circuit is not very good, almost sure the problem comes from here..

I first made a blade driver, then I made a polepiece driver, because I hadn't potted the first one well enough. They both drive the strings equally well, but I'm getting less feedback from the second one. I still can't put it anywhere near the pickups, though, so it will still need to be improved. Have you tried changing the phase to see if that gives you any sustain? You can do that by switching the ground and hot wires going to the driver.

Good luck

Heggis

[Franky]

yes I did, no feedback anyway..

I've got to say that my test configuration is a bit messy: I use a strat with all the pickups mounted and wired, and I pick the sound from a PAF floating above the strings, aboveem the bridge pickup, and the driver is floating above the neck PU.. The magnetical field created by the single coils may disturb my driver..

But I think my problem is about the circuit: I test it with a speaker, and it's really chaotic, I know that the preamp is working, I use it as a buffer sometimes, sounds great, but the 386 section is not really operative.. Where can I find a working schematic? With all the improvement, I'm a little lost...

[newfuturevintage]

Hi everybody...first post here....

thanks pete et al for making this available. So useful.

I've spent the past week pouring over the bulk of this enormo-thread and yesterday made a successful single coil driver (couldn't believe it, first attempt)!

I already had a 386 amp in a popular, curiously strong mint-tin made up (no preamp, just the 200x gain circuit from Nat. Semi's data sheet) and used that to test. I'm in the US, so the most commonly available magnet wire is 30 AWG from radio shack, which I used. Need to source 32 AWG for the next trial.

Hope to get post pix eventually, but my observations so far:

I used welding steel for the base plate, and tapped 6 holes for steel screws to use as pole pieces. I used steel washers around these to space the 2-3mm height (& protect the wire against the sharp screw threads), and used thin brass for the top plate. The screws/poles are exposed through the brass. I wrapped to around 5 ohms (ran out of space before getting to 8, hence the need for 32AWG), then potted the shooting match in wax. No problems with microphonics. I used craft-store ceramic bar magnets, and fiddled with their orientation until I got the best response.

Here's where it gets fun. I tested with a squire '51, with the bridge HB. A little fizz, but I expected that. Much worse with the PU tapped to single-coil. To see if this could be helped with a higher powered amp, I plugged the driver coil into half a Crate powerblock amp (2 x 75 watt amp), and the other half of the crate into a 1 x 12" cabinet. No fizz!

With this much power, strange things happen. My understanding is the driver is essentially a choke, and can store energy in its magnetic field. If you get the pole pieces too close to the strings (which are grounded--you can see where this is going, no?), the driver will arc across to the strings! Scared the cr@p out of me. Not the most practical driver circuit available I guess I'm saying don't try this at home. Back to the 386.

Next, and even more interesting, I tried it on my acoustic/electric guitar. It's got a piezo, a magnetic SC soundhole PU, and a condenser mic. With the added bonus of a phase switch built in. With the guitar only plugged into the driver circuit (listening acoustically), it works amazingly well. Almost all the way down to the bridge. I suspect it's got to do with the heavy gauge strings. The phase switch on the guitar effectively works as a harmonic/fundamental switch too, and individual string drive can be tailored a little bit via the on-board 3-band EQ. So strange to hear electric feedback tones emanating from my acoustic guitar!

Also cool was plugging in the acoustic to an amp, when using either the mic or piezo, no fizz or RF/EMI side-effects at all, even with the driver an inch or two away from the piezo. Excellent!

thanks all, and off to do some more toying around.

[mrjstudios]

Hello newfuturevintage; It's good to see some other Americans getting into this! Actually, I can't believe this great thread isn't much more popular in the US..... it should be....

so the most commonly available magnet wire is 30 AWG from radio shack, which I used.

Yes.... it is... I wish I would have known that when I started; my first trip to radio shack did not produce any magnet wire (I think they were out of stock) and so I sourced some 30 AWG (by the way, since this is an international forum, we have to convert everything to mm.)

30 AWG = .25mm

32 AWG = .20mm

Anyway, since then I have also sourced some 32 AWG / .20mm wire, and that is the best stuff so far. However, my first driver (single blade style) used .25mm wire, and it worked VERY well, so don't be discouraged at all by the .25mm wire from radio shack. If you build the circuit/driver combo well enough with .25mm, it WILL work. For what you are doing, the difference between those 2 wire sizes will not make your driver work or not; it will only affect performance, etc.

You really have made quick progress compared to most of the rest of us. Good work.

-MRJ

[newfuturevintage]

You really have made quick progress compared to most of the rest of us. Good work.

-MRJ

Thanks, MRJ...the only bummer about the 30awg/.25mm wire from Rat Shack is it comes packaged with two reels of larger sizes...kinda useless for our nefarious purposes, but still only $5 in total. Maybe useful as picture hanging wire?

Yesterday was spent destroying a pickup from a super cheap '70s Japanese mustang copy for rewinding. I took pix during the process and hope to post them early next week. This driver is working outside the guitar, but the original pickup remaining in the bridge position is too microphonic/weak/cruddy/quiet in general to be useful, so I'm hitting the used music store junk bins today to replace it. (It's going into parasitic oscillation in harmonic mode, and won't drive in fundamental mode. It's also very close to the driver, so the intent is to find a HB and move it closer to the bridge)

One thought I had yesterday about driver placement: in between frets on the neck of the guitar.

If someone were to use the ultra thin drivers Pete's made, and route between some frets, running the wiring through the trussrod channel it might work well and allow you to still use the neck pickup. It'd be a difficult install, doubtless, but might be a useful location. Heck, if you routed around a fret, you could use it as a magnetic core. But now I'm talking about actually creating the driver and installing it during the manufacture of the neck itself, and that's just crazy talk.

Rock on,

ron

[newfuturevintage]

OK, a bit further in this adventure now...

I was getting some serious RF / self oscillation / ground looping going on with a second coil (single), I wound. The driver amp is just a copy of the original Smokey amp, and I was using a stereo jack wired to serve as a master power switch (Tip: guitar signal, Ring: battery -, Sleeve: guitar / circuit ground) . It turns out this was a bad thing as it created a wicked ground loop.

Moving battery (-) to the driver circuit board directly solved this issue. I now have battery (+) wired to a Switchcraft #13 jack. The #13 is a mono jack with a normally-open, insulated switch so battery (+) does not connect to the guitar's signal path.

so far, so good.

Am getting a lot of fizz. Not a big deal with distorted amplifier settings, but not so good for clean tones. It's the typical 386 distortion that's been mentioned before. I tried isolating the driver amp with a 1:1 transformer which didn't eliminate the problem, but did seem to at least smooth out the fizz so it sounded more like an intentional effect, like a fuzzbox. Am thinking a higher power amp is the ticket. I've picked up a LM380n chip which is supposed to be (up to) 2.5 watts. Just need to hack a buffer circuit for it. Will report back if it works or not. In the meantime, what non-386 amp circuits are recommended? I saw something about using a TDA 2003 chip...anybody try this?

ron

[col]

Am getting a lot of fizz. Not a big deal with distorted amplifier settings, but not so good for clean tones. It's the typical 386 distortion that's been mentioned before. .... In the meantime, what non-386 amp circuits are recommended? I saw something about using a TDA 2003 chip...anybody try this?

ron

The fizz you are talking about is probably due (as you suspect) to clipping somewhere in the signal chain.

If your driver/pickup combo is very efficient, and you tweak the gain settings on pre-amp and power amp just right, you should be able to get minimal fizz and still have a range or response. You will struggle with the 1st and 2nd strings.

My AGC (automatic gain control) circuit was developed partly with this problem in mind - the one area where it does succeed completely is removing fizz. Basically, if the input signal is low, the gain is higher, if the input signal is high, the gain is lower. The circuit is setup so that its maximum gain is just a fraction below the clipping threshold, so close in fact that a new battery with > 9v will push the low frequencies into gentle clipping !

Using a similar circuit but with a bigger power supply would certainly improve the response time of the sustain. However, I think that more work on the driver combined with some more modern AGC processing could also improve the response time while still keeping the 'no fizz at all even with pure clean tone' functionality.

As far as trying out different power-amp solutions, I wouldn't expect much of an improvement unless you either ditch the 9v battery power and go umbilical, or switch over to class-d amplification. Unfortunately all the viable class-d solutions ore on tiny SMD formats - forget about soldering tiny pins, most of these have no pins at all, anf the whole chip is maybe 3mm long!

Still, I have proved that it is possible to get a good working sustainer up and running with NO FIZZ ! that uses off the shelf pin-through-hole components, LM386, 9v battery, and also a very sub-optimal (I feel sure about this) driver.

I don't think you can do it with 9v battery, lm386 (or equivalent) and no fizz without using some sort of AGC though.

EDIT: go here for my sustainer demos, try clean demo and atmospheric noodle for clean tones. Also not that these demos are old, they were done before i fixed a bug in my speaker simulator build, so you will here some crackle here and there - it's not caused by the sustainer. fwiw, it's all sustainer guitar => homemade amp sim => homemade speaker sim => soundcard... there's a little reverb and delay added after on some tracks... (one of them is a full on hack'n'slash montage )

cheers

Col

Edited November 27, 2007 by col

[newfuturevintage]

Hi Col--

Thanks for the reply. Agreed about the source of the fizz. It's exactly the same as a smokey amplifier, both in threshold & decay characteristics. My guess is it's due to increasing EMI hash being chucked out of the 386 as it's output wave squares off.

I've been giving some thought to the design of a simple AGC. Am thinking the easiest design would be an optical compressor either in the buffer stage or directly on the 386. Really just an LED/photoresistor or opto-isolator driven from output and controlling the feecback to the inverting input would work. Of course, it doesn't have to sound good as it's just to smooth out the driver's response and keep the amp from clipping.

Is your AGC schematic public? I've read so much on this monster thread I lost track. If it is, can you point me to which page it's on?

Failing good performance at 9vdc, I'm OK with going to 18vdc for power (my test-bed guitar has a lot of open space inside the body). I found when driving my test coil with the 75 watt amp it was possible for the coil to be somewhat far from the strings and still drive the high E string. Perhaps higher power & an AGC would be an effective combination...

As for your demo files, they're one of the reasons I was so excited to build this in the first place, especially the long, clean atmospheric demo. Very cool. Probably listened to it a half dozen times while researching this.

back to the soldering iron for me

[jamforthelamb]

Hey Everyone,

New to the board.

I looked for the answer to my question, but gave up after looking through about 50 pages

I think I have a very light grasp of the way it works. I'm looking to build my own using an old single coil for the driver, and the Fetzerruby circuit. Where is everyone getting their 32awg or .2mm wire in the States (specifically AZ or the Southwest)? I'd like to find it locally, but wouldn't mind ordering some if need be.

I'm planning on putting this in my very first guitar. It's a cheap strat style body with only one pickup (humbucker in the bridge), and a stop tail bridge. The guitar is pretty thin in comparison to most. Will that have an effect on the sustainer ?

Thanks and God bless!

Kevin

[Captainstrat]

Hi dudes & dudettes, newbie in the house! What I'm about to ask may be a stupid question (I'm not an electronics expert) but I've been wondering: could a resistor be wired to the output of the mini amp circuit to divide the neck pickup's impedance to 8 Ohms?

This way with a DPDT switch one could select between regular guitar operation (and not sacrifice the neck pickup position) and sustainer operation where the existing neck pickup becomes the ad-hoc driver?

Just a thought...and my apologies if this question has already been asked, this is quite a lenghty thread!

[newfuturevintage]

Hi Kevin--

I got the 30 AWG/.25mm I built my first two coils out of from Radio Shack. $5. Takes substantially more room & more winds than the smaller wire.

I was able to source the 32awg/.2mm wire from a surplus electronics shop in CA called "Halted Electronics". They do mail order at www.halted.com . I paid $8us for 1/4LB (114 grams), which appears enough to make a driver for each of the surviving members of Menudo. Most any good electronics shop that has a hobby section should have it though.

[jamforthelamb]

Hi Kevin--

I got the 30 AWG/.25mm I built my first two coils out of from Radio Shack. $5. Takes substantially more room & more winds than the smaller wire.

I was able to source the 32awg/.2mm wire from a surplus electronics shop in CA called "Halted Electronics". They do mail order at www.halted.com . I paid $8us for 1/4LB (114 grams), which appears enough to make a driver for each of the surviving members of Menudo. Most any good electronics shop that has a hobby section should have it though.

Thanks man. I'll likely try the Rat Shack wire first just for the prototype. I'll check a few place as well to see if they have it.

God bless,

Kevin

[col]

Thanks for the reply. Agreed about the source of the fizz. It's exactly the same as a smokey amplifier, both in threshold & decay characteristics. My guess is it's due to increasing EMI hash being chucked out of the 386 as it's output wave squares off.

My personal take on this is that there is always enough EMI from the driver reaching the pickup to be audible. The reason the AGC seems to work is that because there is no (or at least very low) signal distortion, the EMI signal is very similar to the desired guitar signal - therefor the EMI is not really gone, it is 'masked'.

I've been giving some thought to the design of a simple AGC. Am thinking the easiest design would be an optical compressor either in the buffer stage or directly on the 386. Really just an LED/photoresistor or opto-isolator driven from output and controlling the feecback to the inverting input would work. Of course, it doesn't have to sound good as it's just to smooth out the driver's response and keep the amp from clipping.

I have though about an optical type circuit - just went down the jfet route because I could get the parts. Be careful about how the circuit is designed, most 'simple' compressor/limiter circuits - particularly those for guitar fx - do impart quite a bit of distortion to the signal, this distortion will bring back the fizz if its strong enough!. Thats one of the reasons I gave up on my first AGC circuit based on the LM13700 IC.

Is your AGC schematic public? I've read so much on this monster thread I lost track. If it is, can you point me to which page it's on?

yes, it is public - it's certainly not an optimal circuit, and its not simple either - There are plenty of things I would try and change if I were to start working on it again, but you're welcome to use it as long as you post any improvements here.

schematic

power stage layout

main circuit layout

there's a post somewhere in this thread explaining where at least some of the connections should be made for the rotary mode switch.

After you've had some time to study the diagrams, if you're still serious about building the thing, I can try and find some of my notes and check if that is the most recent revision of the circuit.

Failing good performance at 9vdc, I'm OK with going to 18vdc for power (my test-bed guitar has a lot of open space inside the body). I found when driving my test coil with the 75 watt amp it was possible for the coil to be somewhat far from the strings and still drive the high E string. Perhaps higher power & an AGC would be an effective combination...

Personally, I think 2 batteries is one too many. I think this should be as accessible as possible, and the circuit and one battery is already pretty large.

I think is should be possible with some more refinement of the driver and circuit, to improve the aspects of the system that I'm not so happy with. As it is, it's very usable, has an acceptable level of fizz (ie you can't hear it) and reasonable battery life - certainly a good piece of kit.

I would like to be able to control the attack and decay of the AGC. Currently, the attack is ok for getting a 'realistic' loud amp type of sustain or feedback sustain, but its not fast enough for the cool kinds of 'swell' effects and other e-bow style tricks that are possible with my LM13700 circuit... I reckon that this would improve with a better driver, but the ideal would be a 1 chip limiter / class-d amp. Basically, I think the slow response of the system is mostly down to the gain being limited by clipping level. The driver is where I'll be looking for improvements. I'm getting more scientific about the approach, using modeling and analysis rather than guesswork and a trial and error iterative dev process. I want to try using a bigger driver core and heavier guage wire, but this will be more difficult to get working because if you don't do all the measurements and calculations correctly, it won't work (or you'll fry your LM386)

Problem is no cash for the tools and parts, so its going slow hehe. I'm going to test a bunch of different core sizes and materials my measuring/calculating their 'core factor' that way I can design the driver to get a better compromise between efficiency, maximum inductance and frequency response... it also means I can use the 'correct' values for the zobel network on the output of the power-amp stage.

I will probably reduce the number of 'harmonic' modes, or at least spend some more R&D on that part of the system.

As for your demo files, they're one of the reasons I was so excited to build this in the first place, especially the long, clean atmospheric demo. Very cool. Probably listened to it a half dozen times while researching this.

Thanks

I really should do a few more, thing is that I pretty much exclusively play acoustic blues stuff these days... (blind blake, Rev Gary Davis, Bert Jansch, Mississippi John Hurt... that sort of thing), so I would have to commit myself to some practicing and probably a few recording sessions... thats assuming I don't have to break out the soldering kit just to get the damn thing working lol..

I'll get around to it soon I'm sure. I have the parts for another www.runoffgroove.com amp sim box that I want to build, that usually drives me to a week or two of electric before I get back to some real guitar

back to the soldering iron for me

get a plugboard for testing stuff out

like I used for my first working AGC based sustainer system - check it out !

cheers

Col

Edited November 28, 2007 by col

[newfuturevintage]

My personal take on this is that there is always enough EMI from the driver reaching the pickup to be audible. The reason the AGC seems to work is that because there is no (or at least very low) signal distortion, the EMI signal is very similar to the desired guitar signal - therefor the EMI is not really gone, it is 'masked'.

This makes a lot of sense. On my guitar if I'm in fundamental mode and not overdriving the 386, the tone is more or less the same as with the circuit powered off. If I'm in harmonic mode, the tone thins out to closer to a single coil pickup (Am using a humbucker).

I have though about an optical type circuit - just went down the jfet route because I could get the parts. Be careful about how the circuit is designed, most 'simple' compressor/limiter circuits - particularly those for guitar fx - do impart quite a bit of distortion to the signal, this distortion will bring back the fizz if its strong enough!. Thats one of the reasons I gave up on my first AGC circuit based on the LM13700 IC.

.

.

.

yes, it is public - it's certainly not an optimal circuit, and its not simple either - There are plenty of things I would try and change if I were to start working on it again, but you're welcome to use it as long as you post any improvements here.

That's a crazy circuit! Thanks for posting the links to it. I'm going to try the optical route for the time being. I've ordered a few Vactrols from allied electronics and will start playing with them. The idea is to start with a variation of craig anderton's optical compressor circuit in Electronic Projects for Musicians book as it's simple enough for me to understand, and will also take care of the buffering in one fell swoop. Will post details if it's successful.

Personally, I think 2 batteries is one too many. I think this should be as accessible as possible, and the circuit and one battery is already pretty large.

I think is should be possible with some more refinement of the driver and circuit, to improve the aspects of the system that I'm not so happy with. As it is, it's very usable, has an acceptable level of fizz (ie you can't hear it) and reasonable battery life - certainly a good piece of kit.

I would prefer to stick to 9v as well. That said, a lot of active basses run at 18v (EMG in particular recommends 18v for better headroom), so it's not uncharted territory, but yes, 9v is target. I think the main benefit of 18 would possibly be better sensitivity on thinner strings. I could be way off base though!

...the ideal would be a 1 chip limiter / class-d amp.

this makes a lot of sense to me. Given the unavailability of DIY - appropriate class-D chips, I wonder about finding small off the shelf amps and hacking them.

get a plugboard for testing stuff out

like I used for my first working AGC based sustainer system - check it out !

Makes me a bit sad I pitched mine when I moved a few years ago...never did replace it. And it even had a power supply built in. Time to pick one up again. Oh well!

Thanks for the input Col!

ron