Sustainer Problems!

[shiscabob]

Hey everyone, I'm new to the forum, while I do visit it quite regularly, and I was having some problems with my sustainer I'm working on.

First off I was wondering if when you make the driver, could you possibly use the drivers magnet AS the core, istead of having seperate pole pieces, etc. And if so, would it effect the tone, or overall output of the driver, as I am having problems with not getting any output whatsoever when I try to wind it.(i'm using wire from an extra sc pickup from a cheap strat) Im not sure whether it's the wire itself or if I'm not winding it enough times, and also what to do if the wire breaks( wire occasionally breaks when winding and was wondering if you could simply twist the end to another thread and keep winding)

Next I was wondering what kind of capacitors are in the circuit( I used the ruby circuit as my pickups are fairly high in output) as of now on the pcb I made it has ceramic capacitors I believe(the round orange ones) rated at 47n and 100n.

Lastly, i would appreciate some advice on the subject as I'm too lazy to read through 200+ pages in the sustainer thread.

Thanks, Shiscabob.

[elmo7sharp9]

First off I was wondering if when you make the driver, could you possibly use the drivers magnet AS the core, istead of having seperate pole pieces, etc. And if so, would it effect the tone, or overall output of the driver,

What you're aiming for is an electromagnet (the drive current through the coil),

with some additional permanent magnetism to magnetise the string in the area of drive.

For the rapid changes of magnetic flux required by the driver, a soft-iron core is best.

Magnetised Alnico polepiece slugs are ideal.

Attaching magnets to a soft-iron blade is cheaper.

as I am having problems with not getting any output whatsoever when I try to wind it.(i'm using wire from an extra sc pickup from a cheap strat) Im not sure whether it's the wire itself or if I'm not winding it enough times, and also what to do if the wire breaks( wire occasionally breaks when winding and was wondering if you could simply twist the end to another thread and keep winding)

You'll not get very far using pickup wire to build a driver matched to the LM386 chip,

which is expecting to see an 8-ohm load.

0.2mm to 0.3mm enamelled copper wire will give you the required impedance,

using somewhere between 100 and 220 turns to get an approximate DC resistance reading of around 8 ohms.

Next I was wondering what kind of capacitors are in the circuit( I used the ruby circuit as my pickups are fairly high in output) as of now on the pcb I made it has ceramic capacitors I believe(the round orange ones) rated at 47n and 100n.

So long as your capacitors are healthy and the correct values, I wouldn't worry too much about the material they're made of.

This is not an audiophile-grade application.

If, like me, you discover the LM386 to be underpowered for this application,

there are many alternative power amps you can use.

This is one that works for me,

based on a TDA7052 chip.

Lastly, i would appreciate some advice on the subject as I'm too lazy to read through 200+ pages in the sustainer thread.

I advise you to obtain some wire, wind some coils, build some circuits and come back here with your findings.

Edited March 31, 2010 by elmo7sharp9

[col]

For the rapid changes of magnetic flux required by the driver, a soft-iron core is best.

A soft iron or steel core is by no means the best for rapid changes of flux there is a lot of hysteresis with iron particularly, that said, it is perfectly adequate for a sustainer driver.

Magnetised Alnico polepiece slugs are ideal.

To suggest that iron is best then say alnico is ideal is odd considering there is no iron at all in alnico. As its name suggests, alnico is primarily aluminium, nickel and cobalt.

Attaching magnets to a soft-iron blade is cheaper.

not necessarily cheaper, and definitely more bulky, but it will make it easier to get the inductance of the magnet up to a suitable level.

You'll not get very far using pickup wire to build a driver matched to the LM386 chip,

which is expecting to see an 8-ohm load.

0.2mm to 0.3mm enameled copper wire will give you the required impedance,

using somewhere between 100 and 220 turns to get an approximate DC resistance reading of around 8 ohms.

You are correct about using pickup wire.

Unfortunately the rest is a bit more complicated.

There is quite a big difference between using 0.2 and 0.3 wire when making a sustainer. They can both be used, but the design of the driver will be different in each case.

If you are going to move away from Pete's basic design (0.2mm wire wound up to around 8ohm resistance) then you will need to develop a good understanding of these things:

resistance vs inductive reactance

inductance and how it relates to inductive reactance

how inductance affects phase response.

Pete got around his limited knowledge of these things by using an lengthy iterative experimental approach, and arrived at one of the 'sweet spots'. If you don't want to spent hundreds of hour and lots of money, then you either need to follow his design (or one of the other tested designs) very carefully e.g. 0.2 wire MEANS 0.2 not 0.3 or 0.15, or know (or learn) a lot of complicated stuff about how magnets and electromagnets work. Ideally you need some test equipment as well.

Next I was wondering what kind of capacitors are in the circuit( I used the ruby circuit as my pickups are fairly high in output) as of now on the pcb I made it has ceramic capacitors I believe(the round orange ones) rated at 47n and 100n.

So long as your capacitors are healthy and the correct values, I wouldn't worry too much about the material they're made of.

This is not an audiophile-grade application.

very true. In the original sustainer thread, I vaguely remember some discussion about using tantalum caps. I think this was purely relating to size rather than functionality.

If, like me, you discover the LM386 to be underpowered for this application,

there are many alternative power amps you can use.

This is one that works for me,

based on a TDA7052 chip.

If you find the LM386 to be 'under powered' then your driver and or other parts of you circuit are sub-optimal and you should look there for some more performance.

There are certainly problem with the LM386 - it's distortion characteristics are poor, so the circuit need some careful design to eliminate nasty grunge and fizz, but lacking power is not an issue - it's possible to get some of the strings rattling off the frets with an LM386, and even the 'difficult' B and E strings can be driven well with an LM386.

Lastly, i would appreciate some advice on the subject as I'm too lazy to read through 200+ pages in the sustainer thread.

I advise you to obtain some wire, wind some coils, build some circuits and come back here with your findings.

heh - good advice

cheers

Col

[elmo7sharp9]

Hello Col.

I followed your progress with interest during 2006.

Long time, no post...

To suggest that iron is best then say alnico is ideal is odd considering there is no iron at all in alnico. As its name suggests, alnico is primarily aluminium, nickel and cobalt.

What I meant by that is:

Soft Iron is best if, like me, you have no budget and have to make do with what you find at the side of the road.

If you can afford them, buy polepieces - because they're a readily available standard part,

it's one less variable to deal with, in a project that's got no shortage of variables.

If you find the LM386 to be 'under powered' then your driver and or other parts of you circuit are sub-optimal and you should look there for some more performance.

That's the route to follow, if you have the time to spare

and you're in pursuit of a guitar-mounted, single-PP3 powered device...

Tweak the coil efficiency to get peak consumption down, gate the power amp on no-signal too.

Apply AGC to even-out the response (and save power when it's not required).

Apply a preset phase-shift to optimise the feedback.

But not on your first attempt!

There are certainly problem with the LM386 - it's distortion characteristics are poor, so the circuit need some careful design to eliminate nasty grunge and fizz, but lacking power is not an issue - it's possible to get some of the strings rattling off the frets with an LM386, and even the 'difficult' B and E strings can be driven well with an LM386.

Alternatively, apply a little more clean power and get something that works first-time with a wide range of coils

(Which is the result most first-timers are after).

[col]

...

What I meant by that is:

Soft Iron is best if, like me, you have no budget and have to make do with what you find at the side of the road.

If you can afford them, buy polepieces - because they're a readily available standard part,

it's one less variable to deal with, in a project that's got no shortage of variables.

The same number of turns of the same wire around a soft steel or iron core will give a different coil inductance when compared with using alnico pole pieces as the core.

For this to work well, you need two thing in your driver:

#1 enough permanent magnetism to align the magnetic moments in the string.

#2 impedance and inductance values that are tailored for your driver circuit and for the frequency range you want the driver to operate effectively over.

the choice of core material is an important variable in setting the inductance. Just arbitrarily choosing alnico and iron because one is 'cheaper' or one is a 'standard part' is completely missing the point.

Those folks that don't understand this or don't have the equipment to measure inductance would be best to clone a 'known to work' driver as closely as possible - not to use arbitrary materials for reasons that are not relevant to the functionality of the device.

If you find the LM386 to be 'under powered' then your driver and or other parts of you circuit are sub-optimal and you should look there for some more performance.

That's the route to follow, if you have the time to spare

That's the route to follow from the get go.

going another way is just postponing the enevitable.

Unless people are going to use phantom power (I already mentioned that), then they will have to at some point build a driver that is as close to optimal as is possible within their means. Or else they will be unhappy with the results they get. The closer they get on their first attempt, the better.

Tweak the coil efficiency to get peak consumption down, gate the power amp on no-signal too.

Apply AGC to even-out the response (and save power when it's not required).

Apply a preset phase-shift to optimise the feedback.

But not on your first attempt!

Fair enough, it would be silly to start playing with AGC before getting a basic design up and running.

However if the driver is designed carefully, or based (very closely) on an existing good design then 'tweaking' (whatever that means) should not be necessary. Similarly, preset Phase-shift (whatever that is) is unnecessary if the driver and circuit are properly designed.

There are certainly problem with the LM386 - it's distortion characteristics are poor, so the circuit need some careful design to eliminate nasty grunge and fizz, but lacking power is not an issue - it's possible to get some of the strings rattling off the frets with an LM386, and even the 'difficult' B and E strings can be driven well with an LM386.

Alternatively, apply a little more clean power and get something that works first-time with a wide range of coils

(Which is the result most first-timers are after).

IMO, if you want battery powered (like me and many others) you go with a PP3 - that's the standard, and it's certainly achievable.

If you're going phantom powered or umbilical, jack up the watts and use a driver with a lower inductance and a higher impedance - that way you can more easily avoid choking on the higher frequencies due to the effect of the inductive reactance.

Personally, I feel that the sustainer effect doesn't add enough 'value' to a guitar to warrant the extra inconvenience of phantom power or umbilical.

Col

[elmo7sharp9]

I am a regular and long-standing e-bow user and like BOTH e-bow and sustainer modes of playing.

I like conventional loudspeaker feedback, too.

I prefer the strong drive and wild nature of my current sustainer to commercial sustainers,

it sounds and reacts more like an amp stack at high volume.

So long as I can get sustain on every note, I'm not concerned about the drive level.

I quite like the varying response in different areas of the neck

(Something that the commercial manufacturers wouldn't issue, for reasons of mass appeal).

I run it at full bore, just below the threshold of squealing,

using damping techniques - in both hands - to get the effects I want.

The 'arbitrary materials' are not so arbitrary - they're what I had to hand at the time!

Within sensible limits, the core material is not so crucial to me.

The only coils I've made that were totally useless had too little iron

to produce sufficient flux (I was pursuing miniaturisation).

For me, external power is a bonus.

It's cheaper than batteries and frees me from the constraints of coil efficiency.

I already have a 13-pin synth lead and a regular guitar cable,

I can bear a power lead as well.

Bring on the Christmas Tree Lights!

Let those who must have a self-contained system

(and possess the equipment and the will to refine the driver coil) pursue that path.

I'm certainly interested in reading about the results produced by this approach.

I've had success with a tall, thin driver,

using the same circuitry that drives the 'orthodox' 3mm high squat driver.

And I know I'm not alone in this.

So long as the coil is mechanically stable and reasonably matched to the drive circuit,

with sufficient loop gain applied to overcome inertia - sustain will occur.

The dimensions of commercial sustainer drivers strongly suggest that PSW's design

is not the only choice, and if we ALL go down his route, we'll never find the alternatives.

The loyalty that the LM386 chip continues to attract is a mystery to me.

In designing a headphone distribution amplifier for a friend's studio,

I found the LM386 to be, by far, the worst choice in terms of clean power.

It's not even a particularly economic choice, requires a Zobel Network, has limited clean headroom...

There are plenty of chips, some with integrated preamps yet, at comparable prices.

And, let us not forget, commercial sustainer designs generally use a discrete push-pull power amp.

For clarification only:

The 'preset phase shift' I referred to earlier is a feature of the Floyd Rose sustainer

(Built around op-amps U1A and U1D on that well-thumbed diagram)

and is an 'on test' optimisation of the the phase response, via a preset pot.

[col]

There's no need for you to defend your sustainer or the philosophy behind it.

The 'arbitrary materials' are not so arbitrary - they're what I had to hand at the time!

If you read my post more carefully, you would realize that I was referring not to your own personal driver (which I'm sure is excellent), but to the advice you were giving to the OP. You were suggesting he selected core materials based on arbitrary factors not directly related to sustainer functionality.

You might also note that I am suggesting that people without the technical knowledge and equipment should base their driver as closely as possible on a 'known to work' existing design. I gave Pete's as an example, but if you were to post evidence of the success of your own design, and post the specifications, then my recommendation is equally applicable to your driver.

As far as the LM386, everyone knows its not great - I listed a bunch of limitations in the thread. No-one is 'loyal' to it. However, there are very few proven existing designs out there with circuit diagrams and evidence of successful functionality that are publicly available. Some of the few that are available use the LM386, so when I see inaccurate vague advice about driver materials, and then in the same post the LM386 is being described as 'underpowered', it make sense to set the record straight.

Btw, isn't it annoying when people spatter their posts with bold type. IT'S EVEN WORSE THAN SHOUTING

cheers

Col

Edited April 1, 2010 by col

[elmo7sharp9]

There's no need for you to defend your sustainer or the philosophy behind it.

There's no need for anyone to defend their sustainer or the philosophy behind it.

We are all in the same boat...

We're attempting to make something that demands knowledge and experience from

multiple esoteric fields, simultaneously.

It's impossible to develop a sustainer without becoming strongly attached to your own ideas

(That's what gives us the impetus to do something new).

But it's very easy to get stuck on a peninsula of your own expertise,

from which you can see everything quite clearly, but only in your own way.

Unfortunately,

that's precisely the mechanism that caused the closure of the Monster Thread.

The 'arbitrary materials' are not so arbitrary - they're what I had to hand at the time!

If you read my post more carefully, you would realize that I was referring not to your own personal driver (which I'm sure is excellent), but to the advice you were giving to the OP. You were suggesting he selected core materials based on arbitrary factors not directly related to sustainer functionality.

Yes, I really mean what I said...

use whatever you've got to hand, it's not crucial.

Many different coils will work...

I've currently got a 1mm high, 90-turn coil with a 5mm wide mild-steel core

just sitting on the neck-most coil of a plastic-encased humbucker - very loosely coupled to the magnetic circuit, and it's sustaining as well as any other.

You might also note that I am suggesting that people without the technical knowledge and equipment should base their driver as closely as possible on a 'known to work' existing design. I gave Pete's as an example, but if you were to post evidence of the success of your own design, and post the specifications, then my recommendation is equally applicable to your driver.

For details of the last one I made (not the current one on my bench),

take a look and listen to the material at My Website.

That's a 200-turn coil of 0.19mm wire, driven by the circuit at the bottom of the page.

I take special care to not recommend any specific drive circuit, core material, gauge of wire or number of turns,

because I don't believe the current orthodoxy on this is helpful..

As far as the LM386, everyone knows its not great - I listed a bunch of limitations in the thread. No-one is 'loyal' to it. However, there are very few proven existing designs out there with circuit diagrams and evidence of successful functionality that are publicly available. Some of the few that are available use the LM386, so when I see inaccurate vague advice about driver materials, and then in the same post the LM386 is being described as 'underpowered', it make sense to set the record straight.

And when I see newcomers being recommended an unnecessarily troublesome path to follow,

I, also, have to set the record straight.

Why use the LM386 when there are solutions that deliver the goods more consistently,

across a broader range of drivers,

with a smaller component count?

Btw, isn't it annoying when people spatter their posts with bold type. IT'S EVEN WORSE THAN SHOUTING

Typographical style objections?

Now you're just being petty!

[col]

Yes, I really mean what I said...

use whatever you've got to hand, it's not crucial.

Many different coils will work...

...

That's a 200-turn coil of 0.19mm wire, driven by the circuit at the bottom of the page.

The irony is painful to my sides...

You're defending your point about many different coils working and attacking me for suggesting that beginners should follow known existing designs.

Then to back up your point, you describe and provide a link to your sustainer which turns out to be an almost exact copy of one of Pete's old driver designs.

In addition, your sound clip is heavily distorted - show me a driver design with significantly different specs from the ones in the old thread, and provide clips demonstrating it's ability to provide strong CLEAN sustain - then you will have my attention

Edited April 1, 2010 by col

[Ripthorn]

I haven't built one yet, but I am going to at some point. I am just curious what other chips one would recommend instead of the 386? I've heard many say there are better chips, which I am sure there are, but I haven't seen a chip number of any of them, really.

As for the variety, I am sure there are as many ways to build a sustainer as there are to build a guitar. I think the analogy fits. It's good for a first time builder to start with a guitar that is well known and has been done a gazillion times, then move into the more adventurous territory. Of course, that's all philosophical.

[col]

As for the variety, I am sure there are as many ways to build a sustainer as there are to build a guitar. I think the analogy fits. It's good for a first time builder to start with a guitar that is well known and has been done a gazillion times, then move into the more adventurous territory. Of course, that's all philosophical.

Yes, that would be a good approach. That's certainly what I did, and obviously what elmo7sharp9 also did (hehe).

[elmo7sharp9]

The irony is painful to my sides...

You're defending your point about many different coils working and attacking me for suggesting that beginners should follow known existing designs.

I'm not attacking or defending anything.

Must you persist with this adversarial approach?

Have you learned nothing from being temporarily barred from PG?

Then to back up your point, you describe and provide a link to your sustainer which turns out to be an almost exact copy of one of Pete's old driver designs.

It's not the only coil of mine that works,

just one that had a web page already made,

in response to an enquiry on another thread.

In addition, your sound clip is heavily distorted

Heavily distorted sound clip?

Clean your ears!

I've produced some very dirty sustain in the past, I know the difference...

That's a glass bottleneck on clearly sustaining strings,

direct-injected into my soundcard.

There's no pleasing some folk!

- show me a driver design with significantly different specs from the ones in the old thread, and provide clips demonstrating it's ability to provide strong CLEAN sustain - then you will have my attention

I'm beginning to doubt I'd want your attention.

I can see that you'll not be satisfied till I've published a mud-cored driver that recharges the battery from the string excursions, while sustaining six strings worth of flattened-fifth intervals.

[Fresh Fizz]

I have some clips if anyone is interested.

It's on mediafire so you have to download, no pass, no viruses!

The signal goes from built-in fet treble booster into the sound card. ( No FX, no speaker sim)

The sustainer itself uses 2 AGCs, the Ross compressor mod does the lion share of compression and there is a AGC with a feed forward side-chain to tame the strings. My goal has always been a without hands sustainer. There are no controls & only fundamental mode.

And some very old strings, 2 years I reckon and after many turns of restringing.

Results feedback on open strings and 2nd (12th fret) and 3rd (7th fret) harmonics

open strings

Playing the big trouble spot, only second harmonics feedback, no fundamentals

2nd fret

All across the fretboard, at the end of the clip you can hear how the sustainer performs with open chords.

more sustain

Cheers

FF

[col]

I have some clips if anyone is interested.

It's on mediafire so you have to download, no pass, no viruses!

The signal goes from built-in fet treble booster into the sound card. ( No FX, no speaker sim)

The sustainer itself uses 2 AGCs, the Ross compressor mod does the lion share of compression and there is a AGC with a feed forward side-chain to tame the strings. My goal has always been a without hands sustainer. There are no controls & only fundamental mode.

...

All across the fretboard, at the end of the clip you can hear how the sustainer performs with open chords.

more sustain

Cheers

FF

Great clip

the AGC really does make a huge difference - gives that otherworldly quality to the effect. plus lovely clean tones.

can you refresh my memory about your driver specs and what power source you are using.

cheers

Col

[Fresh Fizz]

Great clip

the AGC really does make a huge difference - gives that otherworldly quality to the effect. plus lovely clean tones.

can you refresh my memory about your driver specs and what power source you are using.

cheers

Col

Thanks

As soon as you start to use AGC you don't need that much power. With 0.5 watt all notes sustain! The problem is that 0.5 watt is way too much when playing higher fretting positions. It's impossible to play a decent scale legato-style, especially on the bass strings. It's more like pick a note, hold it tight, palm mute a bit, go to the next note.

Therefore I can't see why an LM386 is a bad choice per se. There are some versions with less power, so make sure to choose the right one. And without AGC you're going to overdrive the IC anyway. It's that or else you use low gain in your positive feedback loop. Then you wind up with a rather weak sustain that builds very slowly but not a steady-state sustain.

The idea you came up with (AGC with a feedforward sidechain) is indeed a big improvement.

No more string rattling.

Any more eureka moments?

I think it was 2 times 65 turns of 0.2 wire, 8 ohms

Picture 1

Picture 2

TDA7231A poweramp, 12V outboard power supply.

Cheers

FF

[psw]

Good clips there FF and very similar to response to the kind of effect that Col achieved as well with the AGC.

I think that once you are in DIY land and get a feel for things, everyone may want a different response and this has been the way for quite a while I suggest. Personally, I don't like things to be that 'controlled' or mild in response but I do like clean of course.

My tele sounds much like this with the intensity control turned right down, very clean and economical on a battery, right down the 386 is only running on a buffer and 20x amplification so I get a pretty clean drive signal there. The tele can be pushed to 200x amplification without a hint of squeal and pretty much no AGC applied and so sounds like a very loud amp feedback guitar but clean (other than the effect of the strings being driven hard electromagnetically) and extremely touch sensitive...hence my suggestion that such simple amplification strategies can work perfectly adequately for this effect...the sound is very much like and ebow with the ability to swell into notes with light picking or hammering on the strings.

My intention was never to mimic the commercial units but to see what I could make of it and tailor to my tastes, the latter implementations though had a lot of range..others have had different aspirations as it should be.

If you use outboard power, regardless of the stuff about chips (and I went through a lot of different chips and circuit designs throughout) you have the potential for a lot of clean headroom for sure, but for me it has to run off a battery and be light on the modifications necessary for the guitar, in fact, I can't recall ever implementing a sustainer that sacrificed the neck pickup and abandoned the mid-driver thing in part as I didn't like the loss of that distinctive sound on a strat. I treat the sustainer as an 'effect' not an end to itself, after the novelty wears off, you still want a great sounding and playing guitar with the full range that it had and not compromised by these things. But others interests may of course vary...as to the desired effect...as it should.

I do wonder whether the dual coil HB style driver has a particular sound though...it would seem to me that the HD types do seem to have a lot less push or something...the very basic single coil designs are not only easier but seem to me to move the strings a little more and well made and designed don't show a lot of EMI by comparison and in my attempts, needed less power to drive them for a similar effect.

Each to their own with the chips, but I think that it is a lot of fuss for no particular gain to pursue it so vigorously. The LM386 is a solid cheap and flexible well proven little amp, if one has to quibble about the need to add a zobel, consider this, those two components are 20c and can be adjusted and there is a lot you can do with the LM386 precisely because it is so basic...you can adjust gain structure and output caps and stabilize it and even use it as a basic AGC control designed to the specs required for the project. In my latter circuit design I did use some SMD and even tantalum caps and such, partly to reduce size and partly for quality...none of this is strictly necessary and although I am not keen on the F/R idea (it was not my suggestion after all) and I find the preamp part a bit clumsy and does need the addition of filters as I have consistently specified (and the 100uF output cap to simulate my results) to keep the 386 happy, it has shown to work time and again.

Really, I still think a lot has to do with the driver design and construction and the implementation of the whole thing and very good results can be had. I may have spent a lot of trial and error to come up with the results I achieved, but that is as valid a way as any to reach a 'sweet spot' and pass that on. My specs were so others could reproduce the kinds of things that I know work and this has been independently proven enough times to verify it. Sure, a bit thicker wire, more power and clean headroom...but will it achieve the other intentions...low mod, economical battery powered, small circuits, musical sound (to my ears and intentions for the effect), etc?

Rewriting the history of why sustainer threads keep getting closed down is just astonishing...really it comes down to personal abuse and a bunch of known lies and assumptions (usually directed at me) all of which are leading people down the wrong path. The LM386 is just an amp chip, it can do the job as any other can, but it is flexible, obtainable, cheap, indestructable, works and is provable. As more details eek out we see some alternatives actually shut down from over heating (something that I observed years back when I tried them) and the necessity for outboard power to get that kind of clean headroom anyway. Well, each to their own. What I don't see in any alternatives except Col's remarkable work (and FF's perhaps by the sound of it) from a fair way back is any cleverness actually specific to this project.

My circuit designs are not that clever, but the circuit is tuned to get a pretty even consistent result and sound on all strings and even sustain chords for a fair while before the bass strings predominate and the later ones some AGC, but these other things look like the most basic amp/preamp possible and as such are not anymore clever than the F/R in many respects...it's an amp without loading, all that was ever suggested that was required regardless of the chip used. Now Col's circuit design and ideas like FF feedback were excellent ideas with several different modes of drive and more than enough for people with the insight to tweak to their hearts content if wanted. I like that, but for me I went more compact, simpler and rethought the 'drive' control to produce different effects, used the 386 itself to send signal back to provide a very mild AGC and little caps to filter that signal so that the bass frequencies would be attenuated earlier...it is all pretty basic stuff really and given the general and possibly accurate analysis of my methods and even skill in such things, not such a bad plan really...works for me, and at least it wasn't just cribbed together ideas from others that come up over and over, as the F/R was for instance and garnered such scorn. The secret to me is still very much in the driver and matching of things, lower power and getting that 'sweet spot' and building from there and that has consistently been my opinion.

Heavy wire, sure but it changes a lot (more than core material) and perhaps you need more power to get the extra force that this path puts one on...my specifications was not because of some magical mysticism about brands of amps or wire size, but a design that with this elements at least fit the various requirements I set for my own work...low mod, battery powered, economical, sound, response, etc... At least with the basic elements for success in that pursuit I have been completely honest and open about what can work, never suggesting that anything else wouldn't or might well be 'better', but I have yet to see or hear it for the DIY crowd on this one!

...

To the OP...there are specific tutorials about wire sizes and such (no need to read 200 pages, that's just nonsense), plenty on magnets as cores to read and you will really need to and certainly you will have to learn a lot more about these tings if you expect to use pickup wire...0.2mm wire for instance is very easy to handle and wind by hand and solder, no you can't just twist together enameled wire when you break it. Unfortunately this project is something that will require either diligently following instructions and examples of what works and/or a lot of study and experimentation...the treads are numerous and long precisely because of this kind of thing, it was never said to be 'easy', though in truth it isn't that hard, but there is a lot more involved than simple guitar rewiring and such and demands a range of skills and a bit of study and knowledge...actually, many people like the project precisely because of this and the amount you can learn...check out all the tutorials and anything else so you get the basics understood I suggest...

[Fresh Fizz]

Good clips there FF and very similar to response to the kind of effect that Col achieved as well with the AGC.

I think that once you are in DIY land and get a feel for things, everyone may want a different response and this has been the way for quite a while I suggest. Personally, I don't like things to be that 'controlled' or mild in response but I do like clean of course.

If you use outboard power, regardless of the stuff about chips (and I went through a lot of different chips and circuit designs throughout) you have the potential for a lot of clean headroom for sure, but for me it has to run off a battery and be light on the modifications necessary for the guitar, in fact, I can't recall ever implementing a sustainer that sacrificed the neck pickup and abandoned the mid-driver thing in part as I didn't like the loss of that distinctive sound on a strat. I treat the sustainer as an 'effect' not an end to itself, after the novelty wears off, you still want a great sounding and playing guitar with the full range that it had and not compromised by these things. But others interests may of course vary...as to the desired effect...as it should.

Thanks,

Yes I get inspired to write some killer lullabies

The neck and middle pickup were sacrificed long before I even thought of making a sustainer. I guess I'm more influenced by Edward Van Halen and Malcolm Young than by Mark Knopfler and Richard Thompson.

For me the same thing, I see it as a gadget

I do wonder whether the dual coil HB style driver has a particular sound though...it would seem to me that the HD types do seem to have a lot less push or something...the very basic single coil designs are not only easier but seem to me to move the strings a little more and well made and designed don't show a lot of EMI by comparison and in my attempts, needed less power to drive them for a similar effect.

I think my HB driver is more height sensitive or maybe I should put it the other way round, further from the driver it loses efficiency compared to a strat-type driver. But it has its charms, when playing 'camp-fire' chords you're not disturbed by aggressive sustain while on the other hand you get a quick response when playing solos.

The LM386 is just an amp chip, it can do the job as any other can, but it is flexible, obtainable, cheap, indestructable, works and is provable. As more details eek out we see some alternatives actually shut down from over heating (something that I observed years back when I tried them) and the necessity for outboard power to get that kind of clean headroom anyway. Well, each to their own. What I don't see in any alternatives except Col's remarkable work (and FF's perhaps by the sound of it) from a fair way back is any cleverness actually specific to this project.

Yeah, the LM386 is all right. I could make it work in my sustainer. With AGC with a long release time. There is no need for more headroom. In my case the Ross compressor limits to 1 W. So my TDA7231A never distorts. But when is that much power applied? Most of the time the other FF AGC limits below 0.5 W. To use that 1 Watt I have to dampen all strings but the high E play open E and wait for the sustain to build.

Very unnatural way of playing!

Cheers

FF

[psw]

To use that 1 Watt I have to dampen all strings but the high E play open E and wait for the sustain to build.

Very unnatural way of playing!

Well, again it depends on the effect or range of effect that you are after. On a lower drive setting mine is similar to your responses and this is great for a lot of things (pads behind chords for instance, or a bit of a harmonic lift as notes fade out if set very low (mine will allow harmonic effects without infinite sustain, people will recall I never liked the word sustainer as this tech is capable of far more expressive things than holding a note forever!))...don't need to alter that much of the technique.

Mine however also goes to very high drive settings, still clean but you do feel the strings being driven...the effect is much more like a very loud guitar that feeds back consistently. Like playing and extremely loud feedback guitar though, you are going to need to change technique to cope with that, damping is of course key...it's not really a plug and play 'effect' but does change the instrument and so requires technique adjustment.

The commercial units tend to my ear in comparison not to create that kind of dynamic sound, far too controlled to my liking, but that's my choice of 'effect'. Part of what I wanted to achieve is a controlled consistent musical effect that could be used at any volume or through headphones for recording to get that big guitar sound and the techniques that go with it on occasion. I also like the subtle details and potential for different sounds and atmospheric effects that the device offers in spades. With a slide the guitar sounds can sound a lot like that david lindley/jackson browne slide such as on running on empty for instance.

HB drivers and single coils are not the only options, sustainiac and some others have done the 'bi-lateral' options which are more like a fender z-coil or p-bass half and half style thing...this might make a good compromise between EMI reduction and single coil throw...but really, if the single coil style works and isn't causing EMI problems, then I didn't see the reason to pursue it too hard for myself. I have every admiration for those that do get it working to their liking, but I certainly didn't find it 'necessary'.

I think also that a lot can be done with the tone shaping of the circuit signal being sent to the driver to even things out...that is why I use things like the 100uF output cap as a rule, with a speaker it is probably very midrange-y to treble (yes, I can't say what the actual response of my driver is numerically, but I know the effect it has and adjust it to sound musical and even). With an amp you routinely use tone shaping to suite a guitar and pickups, even a room or whatever, why not for driver response...it is not that a chip is suspect, but the signal it is sending. Again, a chip like an LM386 allows you to easily do all kinds of tinkering in that regard, but anything will work, maybe in the preamp.

With different drivers I suspect there is going to be radically different profiles as well as the signal from the source pickup, I doubt there will be a single 'perfect' circuit. I did the best I could with my latter designs to make if flexible but was surprised to find that it did work on a range of guitars and pickup types so well, all have been a subtly different effect, but all musical and consistent.

For thin coils balanced on an HB and working, see the 'sustain box' that was exactly that from years ago. Basically it is the same thing, just using one magnetic polarity of the HB to create a SC thin driver...didn't work as well as any other IMHO and the neck pickup sound was severely compromised on my LP, but of course it works, see Primal's HB adaption or others which were pretty much this kind of thing.

Using a known working design is the more obvious approach to these things, deviating by the naive use of pickup wire as 'any coil will do' approach will not work. However, people like Col and FF here desired a different kind of sound and both used HB style drivers to very good effect, and AGC to great effect.

Long time followers of the sustainer thread would know that I jumped right into radically different and complex hex designs early and put that aside to develop something that could be made DIY to keep others happy and silence criticism and suspicions then. I now realize that the simpler forms of device have potential and more than satisfy my needs for such a device, for me! Certainly a chip like the LM386 has enough power to do the job and my battery will run for months with the little use I put it through, or hours with use.

No one need read that whole thread to get the idea about building one of the formula of know working devices, there are tutorials on both driver making and the project generally in the reference section. It is a project that many have found enthralling enough to build upon, but generally after getting initial positive results with know things and taking it further. It is only in recent years that a loud contingent has questioned and spread misleading information about to undermine that starting point that has served well for years.

This has never been a project for just anyone, it's not hard really, especially with a known working design, but not for everyone...just the way it is, this is not a rewiring job!~

[Tim37]

not to throw this off topic but i just wanted to sayh

Pete its nice to see you back in the sadle.

[therizky]

hi...whats up?

this is my first post,and straight to sustainer thing...ive been working on this project and i dont seem to get it work properly,

here is my setup,due to fet transistor existence in my country,i dont use ruby or fetzer valve, i use little gem as the power amp,i think its ok,coz the fet part on the ruby just act as input buffer,and for the driver, i wound 300 turns of 0,10 wire,about 38 AWG in america.

the thing works,but only produce WEAK sustain and harmonic,on sustain mode i still need to turn my amp power a little louder,and on harmonic mode,it should be a squeal to produce harmonic,meaning that i should get the driver and pickups close to each other.

my first thought was,the little gem doesnt enought preamp in order to drive the string,but how could it doesnt strong enought? since lm386 could produce 200 times of gain?,i think the error is on driver coz it use thin wire and more turns...but i dunno

please help.

[col]

here is my setup,due to fet transistor existence in my country,i dont use ruby or fetzer valve, i use little gem as the power amp,i think its ok,coz the fet part on the ruby just act as input buffer,and for the driver, i wound 300 turns of 0,10 wire,about 38 AWG in america.

The buffer is there for a reason.

If you don't have a buffer, the low input impedance of a LM386 based amp iwl 'suck tone' from the output.

The fetzer/ruby is not ideal for a sustainer as it was designed as an amp to give a warm valvy tone, not to provide optimum buffering and drive for a sustainer,

the thing works,but only produce WEAK sustain and harmonic,on sustain mode i still need to turn my amp power a little louder,and on harmonic mode,it should be a squeal to produce harmonic,meaning that i should get the driver and pickups close to each other.

there should be no squeal. the harmonic mode should work without moving the driver closer to the pickup.

my first thought was,the little gem doesnt enought preamp in order to drive the string,but how could it doesnt strong enought? since lm386 could produce 200 times of gain?,i think the error is on driver coz it use thin wire and more turns...but i dunno

The gain of the amp is not really relevant. what matters is how much current it can supply before distortion sets in (over the frequencies we are interested in).

The LM386 is designed to feed into an 8ohm load. that's roughly 150 turns of 2.0mm - 2.3mm wire.

if you get much lower than 8ohms (e.g. 6 or less) you will get distortion and overheating.

if you get much more than 8ohms (e.g. 12 or more) you will get a lot of distortion if you are pushing the amp near it's limits.

unfortunately - assuming you are using a core about 55mm long and a few mm wide - 300 turns of 0.1mm wire gives a dc resistance of ove 70ohms. Thats WAY to high.

It's important that you ignore the folks here who are suggesting that lots of coils will work, and that it doesn't really matter what wire or core you use - that is ignorant misinformation from people who don't understand what's going on.

Different wire will work, but only in a configuration that is carefully designed and in which that particular wire gauge is optimal.

For a 55mm long core in a single core driver being driven by an LM386, a good starting point is 0.2mm or 0.23mm wire, and wind it until its DC resistance is about 8 ohms. you NEED a DMM (digital multi-meter) for this. It should work out at about 150 turns.

If you decide to use a bi-lateral or a bi-longitudinal multiple coil conficuration, then you might need to use a thinner wire guage - I think for bi-longitudinal 0.18mm is about optimum. 0.15 may be too thin (I'm going to find out soon though)

If you have a more powerful camp/circuit that can handle a wider range of output loads without distortion, then you could go with more turns of a heavier guage - say 0.28mm or 0.31mm.

do remember that there is a significant difference between even 0.2 and 0.25 wire. so going from 0.2 to 0.1 is an enormous difference.

cheers

Col

[col]

A little addition explaining why 300 turns of 0.1 wire is over 70ohms

to start with lets state that 125 turns of 0.2mm wire around a 55mm core is 8ohm (this is roughly correct)

Firstly the wire:

0.1mm wire is not (as may seem intuitive to some) half the size of 0.2mm wire. This important thing is how much copper there is per unit length of wire - this is proportional to the area of the cross section of the wire.

Assuming the cross section is a circle. The formula for area of circle is PI * radius squared

So half the size of 0.2mm wire would be sqrt( (0.2^2)/2 ) = ~0.14. So in reality, the nearest guage to 'half' of 0.2mm is 0.15mm not 0.1mm

( ^ means 'to the power of', so 0.2^2 is the same as 0.2squared)

0.2mm wire is actually 4 times the size of 0.1mm wire !

this means that 0.1mm wire has 4 times the resistance of 0.2mm wire (for the same length of wire)

( 0.1^2 / 0.2^2 = 4)

Secondly the turns:

300 is 2.4 times the 125 turns from the 0.2mm driver.

so lets take our original 8ohms and multiply it in turn by the turns ratio and by the wire guage ratio.

8 * 2.4 = 19.2 (so even using 0.2 wire, 300 turns bumps the resistance to 19.2 which is already too much for an LM386)

19.2 * 4 (wire gauge ratio) = 76.8

These results closely match the results you get inputting these coil specs into the Ziegler pickup calculator

So although it seems initially that there shouldn't be much difference between '125 turns of 0.2mm' and '300 turns of 0.1mm', the difference is actually very significant.

cheers

Col

[psw]

Col pretty much summed it all up there...

You need the transistor or similar to provide a signal conditioned for the input of the LM386 for it to work effectively and to match with the high impedance of guitar pickups...it's not really there to provide gain but to avoid this "loading" and tone suck. It does not need to be the fetzer design but it will work and although I can't get the specified part down here either, there are substitute equivalents that work in its place (just have to watch the legs are the same function).

Same with the wire gauge, although some have suggested otherwise, there are optimal ranges of wire and I have found for my design that 0.2mm is important to get the expected results...thinner wire is definitely not the way to go.

That you are getting a little sustain is something, the problem is that the coil is totally mismatched an incapable of providing much force without a lot of power into it to even get a weak response and in harmonic mode it's just squealing from leaking EMI and not EMF.

A more efficient driver helps considerably fix these problems. Although a thicker wire might be successful, I think it will take more power than the 0.2mm that was tested extensively to get the same response.

As Col is indicating, that was for this particular design, but such things may alter with significantly different approaches like dual coils...but I wouldn't be attempting that kind of thing yet!

It's not clear exactly the set up, but there are a few components like magnets and potting that are also important to the success of this project, but if you follow the know working specs and don't cut corners (like leaving out the buffer) it will work...good luck

[Fresh Fizz]

hi...whats up?

this is my first post,and straight to sustainer thing...ive been working on this project and i dont seem to get it work properly,

here is my setup,due to fet transistor existence in my country,i dont use ruby or fetzer valve, i use little gem as the power amp,i think its ok,coz the fet part on the ruby just act as input buffer,and for the driver, i wound 300 turns of 0,10 wire,about 38 AWG in america.

the thing works,but only produce WEAK sustain and harmonic,on sustain mode i still need to turn my amp power a little louder,and on harmonic mode,it should be a squeal to produce harmonic,meaning that i should get the driver and pickups close to each other.

my first thought was,the little gem doesnt enought preamp in order to drive the string,but how could it doesnt strong enought? since lm386 could produce 200 times of gain?,i think the error is on driver coz it use thin wire and more turns...but i dunno

please help.

Hi therizky,

What you could do if you have plenty of 0.1 wire is the following;

measure the resistance of your sustainer driver coil.

Calculate how many turns of windings you need to get 16 ohms.

Let's say you measure 64 ohms then you need 300 * 16 / 64 = 75 turns.

What you have to do now is to wind 2 coils with 75 turns on top of each other and connect them in parallel. Then you obtain an 8 ohms driver.

Some solutions for a buffer.

The bipolar transistor with the 2 resistor voltage divider seems to be the most appropriate solution. (below Dr. Quack)

Good luck

FF

[therizky]

cool...thanks for the feedback...now im thinking that im one step closer to never ending notes..,haha

1."The buffer is there for a reason.

If you don't have a buffer, the low input impedance of a LM386 based amp iwl 'suck tone' from the output.

The fetzer/ruby is not ideal for a sustainer as it was designed as an amp to give a warm valvy tone, not to provide optimum buffering and drive for a sustainer,"

-have tried that input buffer provided by Fresh fizz last night...unfortunately the little gem doesnt interact very good with buffer on its input...

maybe ill find another way...

but is theres any otehr guys around here have experienced working with this circuit (with or without buffer on it) ? iam afraid that this circuitr wont

work for this

2."0.1mm wire is not (as may seem intuitive to some) half the size of 0.2mm wire. This important thing is how much copper there is per unit length of wire - this is proportional to the area of the cross section of the wire.

Assuming the cross section is a circle. The formula for area of circle is PI * radius squared

So half the size of 0.2mm wire would be sqrt( (0.2^2)/2 ) = ~0.14. So in reality, the nearest guage to 'half' of 0.2mm is 0.15mm not 0.1mm

( ^ means 'to the power of', so 0.2^2 is the same as 0.2squared)

0.2mm wire is actually 4 times the size of 0.1mm wire !

this means that 0.1mm wire has 4 times the resistance of 0.2mm wire (for the same length of wire)

( 0.1^2 / 0.2^2 = 4)"

-ok,if 0,10 wire has 4 times resistances than 0.2 wires,then the whole thing should be like this.

if 125 turns of 0.2 mm is 8 ohms.

then 125 turns of 0.1 mm is 32 ohms

then all i have to do is just divide it by 4,the overall turns for 0.1 mm wires to achieve 8ohms is just 31,25 turns.........is this correct?

and also i have tried that pickup calculator,it shows 34 turns...not so different

3.,the starngest thing about my driver is,no matter how many turns of wires and the sizes of it (i have treied 100,200,300,150,64 turns of 0,1 mm and a thicker wire,which i dunno the dia. but still on the same numbers of turns mentioned above) the driver itself generates sound! hum,squeal,and even the sound of the string itself....meaning that the driver acts as speaker even without a CONE.....is this normal?...if it is normAl,then thres nothing to worry about...and if someone ever asked,i didnt pot or glue the driver...i only use alitlle "playdoh" to make the driver stay still...is this the answer that my sustainer become very weak? that the energy come out as sound and not electromagnetic pulse?

4."measure the resistance of your sustainer driver coil.

Calculate how many turns of windings you need to get 16 ohms.

Let's say you measure 64 ohms then you need 300 * 16 / 64 = 75 turns.

What you have to do now is to wind 2 coils with 75 turns on top of each other and connect them in parallel. Then you obtain an 8 ohms driver."

- i actually dont raelly understand this formula...please explain. my driver measures 76.8(as col has posted)

so i multiply the number of turns with 16 ohm(why 16 ohm?) then divide it with the impedances

300*16/76,8 = 62,5 the results is different again......then why should i wind two coils? and connect them in parallel? are they some kind of sharing currents or something?sustainer driver actually only need one coil isnt it?