col

Hi diy-comrades,

I've more or less read the thread and just have built my second driver. I took your advice and used 0.2 mm copper wire. It's a humbucker with 2 blades.

Dimensions: 1.4 x 16 x 57 mm.

Coil height: 5 mm

Windings: 2 x 60 turns

Resistance: 8,4 Ohms

My first driver was a single coil design and terribly sensitive to magnetic feedback (pigscreamer). The new driver seems to be able to do the job. I get string feedback without squeal. I've only done some outboard testing with the new one, but I have good hope it will work when it's properly installed.

Col you seem to want a more systematic, scientific approach

Yes, and it's great to see someone willing to join me on that road at last

Electrical properties of the sustainer/driver

There is an easy way to measure the inductance of the driver....

Wow, that seems like an interesting approach, but there are certainly simpler ones that use less equipment - What I'm really looking for is a way that folks can take these measurements without needing access to oscilloscopes, inductance meters or 'real' signal generators. I was hoping to get some method worked out using a DMM, a PC soundcard as generator, and some simple to build DIY circuitry. I have discovered a couple of approaches online, but unfortunately, I don't have an inductance meter yet to verify any results... and I've also discovered that there are LCR meters available on ebay with 'buy it now' prices as low as £10, so maybe the whole discussion is moot ?

L = 1/(4* pi^2 * 6000^2 * 680e-9 ) = 0,001 H = 1 mHenry

hehe that is interesting, a few days ago I worked out a ball park figure for inductance by scaling the specs I found online for fender strat pickups - guess what the resulting estimate was... yep, 1 mHenry !!

(pickup specs)

inductance 2.2H to 2.8H so roughly 2.5H

turns 7650 (up to 8350) so lets say 8000

resistance 5.88k

so if we assume a roughly similar core factor, we can extrapolate some ballpark figure for a sustainer drivers inductance.

awg 42 is 0.0025 in diameter (0.0025 what though? inches) 0.063mm

Rm = reluctance, N is num turns, L is inductance

from pickup Rm = N^2 / L = 8000^2 / 2.5 = 25600000

to driver L = N^2 / Rm = 150^2 / 25600000 = 0.00088 = 880uH

or very roughly, 1 mili Henry (I’m assuming our rail cores will have a little more inductance factor than a strat pickup)

The highest fundamental is about 1100 Hz.

X_C = 2 * pi * f * L = 2 * pi * 1100* 0,001 = 6,9 Ohm

Now we know the impedance at 1100 Hz. Real component R = 8,4 ohm, imaginary component X_L = 6,9 ohm

Z_1100Hz = (8,4 ^2 + 6,9 ^2 ) ^0,5 = 10,9 ohm

There are 2 reasons why at 1100 Hz the driver works less efficiently:

(1) higher resistance (10,9 instead of 8,4), lower current (I = V / Z), lower magnetic flux

(2) phase shift between voltage and current

arctan(X_L/R)= arctan(6,9/8,4) = 39,4 degrees (0 is in phase, 180 is out of phase)

Note that 90 degrees is the maximum phaseshift. At 90 degrees phase shift it's all conductance, the driver works as an ideal conductor. The driver won't dissipate energy, it stores energy and then spits it back at the lm386 (in case of the f/r.

Do you mean that 90º is the largest phase difference that won't cause losses, or that its actually more efficient than no phase difference ?

It would be nice if I could combine (1) and (2) in one single number, a compensation factor for high frequencies. I will use the term specificaly for 1100 hz.

(1) 8,4 / 10,9 = 0,77

That help us compare two drivers for phase distortion, but it doesn't help so much in working out how to tweak the rest of the system to get the optimum performance from any single driver.

It also only takes into account the losses due to phase distortion. What about the fact that we can get a stronger electro-magnet from the same current by using more turns of thicker wire... with the side effect that there is a bigger variation in phase with frequency over the guitars range... where is the 'sweet spot' after we have factored in the phase shifts of the circuit, the pickup, the mechanical aspects of the system and created a zobel network that provides the best possible 'correction' ?

It's possible that the optimum driver might not score very well using your metric?

(2) use cosinus function normaly used for calculation of power: P = I * V * cos(phase shift between I and V)

Surely using that formula, P = 0 when phase shift is 90º... what am I missing?

In order for the sustainer to work the string should be helped. So when the string moves down the driver should pull and when the string goes up the driver should push. Just like a swing the timing when to push and when to pull is important. Phase shift, pushing too late or too soon, reduces efficiency.

Yes.

However, this is true for the system as a whole rather than for the driver in isolation. The driver will have some sort of non-flat phase response(Pete seems to have minimized this through experimentation by making the impedance mostly resistive over the guitars frequency range, but possibly with other negative side effects). The circuit too is unlikely to have a flat phase response, tweaking of output cap, zobel, and other components will change its phase footprint. Theres more, the pickup that our signal comes from will surely a pretty hefty impact on the systems 'phase sum' as well!... And there even more (pheck me this is getting ridiculous)... there is a physical phase issue to factor in along with all that other stuff:

(time to spill some of the beans me thinks - nobody seems interested in a more complete document anyway). Ok, this may or may not be news to you guys - it was to me, kinda anyway. Using standing waves as a model for understanding how the string vibrates is woefully inadequate for us if we want to fully understand this project and build the best DIY sustainer possible. What really happens in a vibrating guitar string is quite different (it's very similar to what is happening in a violin string)

What I am talking about is something called 'Helmholtz motion'. Theres a good animation of how it happens on a bowed string here. What is even more interesting for us is that it also happens on a guitar - when you pluck a string, a kink in the string (Helmholtz corner' moves down the string on one side, reflects off the fret or bridge then returns along the other side (For a plucked string, its not exactly the same as Helmholtz motion - there are two kinks moving in opposite directions from the pluck - but for our purposes its effectively the same).... If you don't believe this nonsense, take a very close look at this amazing video clip of Helmholtz motion actually happening on a guitar!

So why is this so critical for us ?

Well, its to do with the gap between the driver and pickup (something Pete's had a hunch about for some time). If the signal from the driver is exactly in phase with the signal from the pickup, there is a section of the kinks journey around the string where the driver will be 'pushing' the wrong way. As the kink passes the driver moving towards the bridge, the section of string between kink and bridge is still moving 'up' while the section between kink and fret/nut is now moving 'down'. The pickup is collecting the signal from the portion moving in one direction while driver is trying to apply that signal - with force - to a section of string that is now moving in the opposite direction... oops... some hefty damping is occuring at this time!

(Whats more, this damping is happening exactly at the time when it is most important to have optimum drive if you want that full rich sound we've been discussing recently - more about that another time if you're interested...)

Unfortunately, the perfect phase offset required to prevent all damping due to this effect will be different depending on which note you are playing at which fret.

e.g. an E played on the open 1st string will have a kink moving twice the speed of the kink in the same note played on the 6th string 12th fret. The phase offset required to 'fix' the 6th string note would be twice that required for the 1st string!

Of course, the scale length of the guitar also has some impact here.

I think it will be possible using carefully designed circuit and driver along with some basic phase tweaking circuitry to get at least a 'best fit' compromise that gives a more efficient better sounding system and an understanding of what the tolerances are for the sytems components...

I figure that at least to start with, we should be looking at just the high E string and optimizing for that. Then its open to anyone who wants to to try and tweak things to suit the lower strings - hilbert transformer or some such...

I don't think we need to worry about the 'same note lower string higher fret' thing - just optimize for each note being played on the highest string that it is available on.

(I also wonder if we[me anyway] have been confusing ourselves by using the term 'phase' in to related but different ways, but interchanging the two due to ignorance ? is the 'phase' offset of the drive signal the same as the phase offset between voltage and current as described by Fresh Fizz?, or are they closely related but subtly different concepts ?)

(2) can we construct drivers with a higher compensationfactor (closer to 1). Maybe some of you have constructed a much more efficient driver, I only know what I have.

Surely a driver using fewer turns of much thinner wire will have a compensation factor closer to 1?

As the wire gets thinner and we reduce the number of turns, the reactive component of the impedance moves towards zero and the compensation factor moves towards one... unfortunately the drive is also reduced in this case, so efficiency is way lower.

what we need is a compromise between linear phase(of the whole system pickup, circuit, driver & guitar) and maximum drive - over the guitars frequency spectrum... personally I feel that this will involve more turns of thicker wire and some compensation circuitry to deal with the phase response and the 'Helmholtz Gap'.

Maybe its time to have another look at an old idea in the light of new understanding...

I suggested many moons ago having a dual driver with the equivalent of a woofer and a tweater, one section optimized to handle high frequencies, and the other for low frequencies.... this could maybe be in the same format as existing dual core setups, but with parallel coils of different specifications.... we'd probably need annoying extras like a crossover to make the most of it, but its worth consideration.

Sorry for being long-winded, it must be the thread that makes me do it. Also, this contribution of mine seems to be a bit out of sync with what is being discussed at the moment. But it would be nice if every s******er builder has a way way to measure and calculate the inductance of his driver. Even more so if he shares his results with the other builders.

Don't apologize - You sound like exactly the sort of windbag we need around here and definitely not out of sync either, I needed a push to start airing some of my new discoveries... maybe more will follow soon - who knows.

cheers

Col

It is also a little worrying to my concept as obviously there will be enormous variations with different pickups (stacked, etc) below it, again straying from the universal circuit ideal....hmmmm

Yes. And this is a similar variation to that which will occur just due to the inductance factor of the cores of different pickups as I was explaining a few posts ago.

As I see it there are two(maybe 3) options for a ready to install piggyback system like yours.

#1 work out a system that is a compromise between working similarly with different installs on the one hand, and having a less than stellar performance as a result, but needing no edjustment. (pretty much what Fernandez et al have done - and they don't have to worry about piggybacking).

#2 learning and understanding about the impact of inductance and phase on the system, then providing a unit that can be adjusted to match a wide range of pickups - this will need to include some way of measuring the inductance either directly, or indirectly (by indirectly I mean some tuning method where you can for example listen for some characteristic noise from part of the circuit while tuning - or something like that)

#3 (a big old maybe) A digital system that can tune itself in - with a learn mode, it can calculate the required phase offsets and memorize settings for the instrument it is installed on. I'm sure this is possible, unfortunately, the circuitry and software will both be high tech. We're talking about a DSP processing unit (possibly alongside a general purpose cpu) - you would need to do some studying before even considering this approach - it's not something that can work through trial and error. It may be possible to get some sort of specialized DSP oriented PIC thats ready to go, in which case all you would need is some simple support circuitry... but of course it will be hard to avoid having to write your own software.

Personally my favorite is #2 (I love the idea of #3, but I'm not sure it will work any better, and its certainly not such a good DIY option). The thing with #2 is that it can still work with peoples DIY drivers, or with other forms of ready supplied drivers. So someone going for a stand alone driver (say they only use a bridge humbucker anyway) can use the tuning method and therefor the circuit needs no modification to work just as well on that installation...

cheers

Col

I was hoping for some radical changes in harmonic response...there seems to be some...generally even a large load (closer to open) reduces the effect...although you get hints of some notes being more "harmonic" in response...hmmm

Perhaps I should have the cap in series with the load rather across it...anyone got any other ideas to try?

I'm not sure that there will be a huge benefit to using this as a control - might be better to find the best resistor value, then use other techniques to actually control harmonic content.

I imaging that you will enjoy some of the possibilities of the all-pass filter approach .

Did you try the variable output cap thing that I described a few posts back? (in post #3485)

That might give you some control over fundamentals vs harmonics.

cheers

Col

I hooked up a 500k(Audio) pot to the unconnected neck pickup leads. At zero "volume" the same type of response as if open, but with the load, the harmonic resonances are smoothed out and so ring as fundamentals where these harmonics are more prevalent.

There appears to be no change in driver efficiency by doing this, so still a good strong sustain as far as I can tell.

Harmonic mode seems to work a little better...but still not great...certainly not the response that I get with the original sustain-o-strat.

First guesses as to what might be happening.

The parasitic pickup coil is using some of the energy that would otherwise be forming the drivers magnetic field. This is reducing the effective inductance of the driver(thats a big loose guess - real inductance measurements would be nice). If you stick a resistor between the ends of the pickup coil, you reduce the current induced in the coil... not sure if this means that it doesn't take as much energy from the driver or if it does but the energy gets sucked in by the resistor... either way you limit the magnetic field induced by the pickup.

So the guess or really question is: does this change the inductance of the driver!

If so it would also change the phase response, which would change the frequencies most likely to sustain well, and alter which ones will produce harmonics.

Another possibility is that the driver inductance stays the same, but the damping effect of the pickups parasitic field is reduced. The pickup has a very different inductance (many more turns) and therefor phase response to that of the driver (it will damp some frequencies way more than others). So if its influence is reduced, the overall phase response of the two combined will change - causeing a change in which notes/harmonics work well etc.

(that's all assuming that the pickup is damping rather than enhancing, but I think that's a pretty safe bet)

It is important to note that a relatively small change in the phase response is all it would take to go from weak or non-existant 'full harmonic mode' to working OK 'full harmonic mode' !

------------------------

I suppose another way to think about it is that the pickup coil isn't taking energy from the system... it's just part of the driver. the effective driving field is a combination of the pickup coils field and the driver coils field.

Reducing the effect of the pickup will still change the phase response.

We may also lose some energy as heat in the resistor, but maybe not enough to be noticable - particularly in a system with AGC.

One thing that would be really useful is a dual channel scope. That way we could actually look at the phase difference by comparing the drive signal at the circuit with the pickups output signal... and see what difference these adjustments make at different note frequencies.

damn, now I need a scope as well as an inductance meter. LOL

cheers

Col

hmm, I've found that by using the search topic box at the bottom of each page, I can very successfully search for individual posts within this single thread (thats what its for ;p)

well ...golly...I guess I never searched for something while the thread is open or something...geee...well, no more complaints about long posts then, that will help...

hehe, don't worry about it - I only found it through luck when I wasn't searching for anything.

It's a really bad piece of user interface design IMO. It should at least be available at the top of the page. Even worse than that, it has an ambiguous title... 'Search Topic' could mean search for the topic you are interested in.... what it should say is 'Search This Thread'... I'm sure with that title and a more prominent position it wouldn't have been missed.... although, I guess its not top priority as most threads won't get beyond a few pages.

Something else I worked out is that it's possible to create a bookmark that always starts you only the last page of the thread:

http://projectguitar.ibforums.com/index.ph...iew=getlastpost

At least I found that useful

(the URL is truncated by IP.Board, but if you right click it or look in the status bar, you'll see it in all its verbose glory)

That reminds me of an electric motor; if you spin its axle when leads aren't shorted, it rotates more freely when compared to leads shorted-situation.

I will have to try that...no, I have never come across it nor read about that phenomenon...

If you think about it, it does make sense - also the stuff about shorting the ends effecting the driver differently to keeping them open....

We know that moving a wire around near a coil creates a current in the coil, and that a current in a coil creates a magnetic field... with out these effects there would be no sustainer (or motor).

If the coil is disconnected, the circuit is broken, (you might get some build up of charge? enough for sparks?? anyway) no current flows. However, if you connect the ends to each other, you make the circuit and a current will flow when either the string vibrates, or more importantly, when there is another coil nearby with a current flowing in it!

So if the ends of the pickup coil are connected to each other, the signal in the piggyback driver will cause a current in the pickup coil to flow, and the pickup coil will develop a magnetic field...

Is that field going to oppose the drivers field ?I haven't searched the net or in a book then sat down and worked it out...

My guess is that it would oppose it, because the energy for the pickups parasitic magnetic field needs to come from somewhere... so it must be sucked out of the driver... damn leaches these pickups are...

does that make any sense?

----------------------------------------------

WARNING - you are now entering teaser territory!

As for the other stuff about possible benefits of clipping in the drive signal etc. I have some interesting (very) ideas about that issue and other related topics. These ideas are based on a new more accurate understanding of some fundamental aspects of this project. Basically lots of juicy info that should help give us a much better model for understanding why we get the results we do, what new ideas might be worth trying and what things we will not be able to 'fix'.

It's a big document already, and it's not nearly finished yet, so you'll have to excuse me for not ruining it by giving away all the goodies in the thread

I'm not sure If I will attempt to include sections explaining our existing knowledge:

• the various attributes of different types of circuit
  
• different types of driver
  
• AGC
  
• other ways of approaching the task (like mechanical systain, umbilicals, ebow etc.)
  
• Problems like grunge, squeal, poor performance etc. what the causes might be and what solutions might exist
  

If folks want that stuff, then I'll have a go.

I imagine I'll be looking for somewhere to host it at some point in the near future - so any suggestions would be cool - somewhere thats free and where folks can leave comments and suggestions, so I can fix mistakes and add any suggested topics and ideas.

cheers

Col

first though. I don't think you will find anything by searching as if it is in this thread, which it is, the search function will bring up the whole thread...not the relevant post...a big problem finding things now the thread is soooo looooong...

hmm, I've found that by using the search topic box at the bottom of each page, I can very successfully search for individual posts within this single thread (thats what its for ;p)

I can usually find any particular post or discussion I'm looking for from way back when within a few minutes... assuming of course that I can remember the topic under discussion

e.g. go an put LM13700 into that search box (at the bottom left of this page), you should get about 3 pages of results - certainly a manageable quantity... then when you found the post in the results, click on the 'Post Preview' link under it to jump to the parth of the main thread it came from.

cheers

Col

Some thoughts again..

@col: If I remember right, you are using sustainer, that has an AGC?

If so, does it have a "pick filter" -properties in it?

Depends what you mean by: 'a "pick filter" -properties'

@col and psw:

When you pick a string or hit chord, there is always signal peak, before string(s) settle and vibrate at their fundamental note.

That peak goes to sustainer and is amplified(unnecessarily?) too, causing less efficiency.

hehe, this topic has been worn out in the past lol

What you say is true IMO, assuming that not having drive during the initial loud part of the sound doesn't have a negative impact on the 'alive' quality of the sound.

Personally, as I learn more about the intricacies of how a sustainer works, my view on this keeps changing and right now, I'm not sure either way. My current system doesn't drive the strings until the initial attack is over because of the way the AGC is set up. This works well, but it gives a different feel and a different sound to non AGC systems - one thing it certainly does have going for it though is efficiency.

Pete has recently reported a 'ringing sound' during the attack portion of the note. This goes away when he turns the driver gain down a little. What I think is happening here is that initially, the strings natural momentum due to picking the note and the force applied magnetically by the driver are not 'tuned in' to each other, in some cases with certain notes on certain strings, they can be fighting against each other. If the gain is lower, what you get is a little dip in volume as the driver causes some damping, and then as the energy in the string quickly dissipates, the driver takes over.... This happens on my system as the sustain kicks in - there is a small dip in volume, a bit like you can get with a compressor. If the gain is set high enough so that the driver can nearly match the natural attack energy, there can be oscillations which sound like a ringing noise...

I suppose the ideal would be to have subtle AGC to avoid just the loudest part of the attack phase, and have the gain just low enough so that all other ringing is avoided.

There are a bunch of other related things that I've been studying recently, but I'm writing them up in almost an essay style, because theres just too much to post, and it's not just ideas and questions, there are a lot of answers as well. So I'll see how I get on with that before posting too much of it here. I will most likely try to include lots of other stuff about how the system works, what the options are, what the issues are and what possible compromises and solutions exist.

This won't be a 'how to build a sustainer' thing, but it should be a pretty good information doc for folks who want a better understanding of how a sustainer works (at least the type we are working on)

It came to my mind when testing things and listening some live recordings. Feedback really doesn't seem to start at the time you hit string(s),

unless you have some uranium-core pickups and wall of amps that go to 11.

Very true - feedback, usually takes time to 'bloom'. This is something I was exploring when I started working on AGC based designs. My initial reason for wanting a sustainer was to mimic what happens with a loud amp, but with just headphones or into the computer... its not quite as simple though (what ever is?)

What about the stronger sustain and the warm singing full bodied tone that you get from a loud amp ?

It doesn't have any lag - you get it from when you hit the string! and it is arguably even more important then classic 'feedback'!

If this effect can be produced (at least in part) by a sustainer, then its unlikely to work if the sustainer doesn't kick in until natural feedback would start to occur (this is a limitation of my existing system, so I am talking from experience)

Anyway, I'm sorry if this subject is discussed earlier.

Tried that topic search with word AGC, but it doesn't seem to allow that short acronym.

No problem. Its always good to revisit these ideas and have another look at them - luckily thats exactly what I've been doing over the last few days

If you want to find earlier AGC related discussion, here are a few search words to try:

automatic gain control

lm13700

dynamic range inversion

limiter

compressor

threshold

my circuit ( :D)

cheers

Col

Well the main reason is I have another variax transplant,and two I dont need the switching and all the other stuff of the fernandes.I think the main problems you guys are having has to do with rf and switching the unit in with your other pickups-I dont have that problem.

Pete is busy trying to perfect a massively complex switching arrangement, but not me

I just use a pickup and a driver and that's all. there is an on/off switch and a mode switch.

Personally, I don't see the switching as being problematic - it's no different from designing switching and controls for a guitar or for a stomp box - same kinds of compromises between aesthetics, usability, flexibility and functionality... sure, it will take a bit of time and testing to get it just how you want it, but thats to be expected.

We don't have any problems with rf. The issues we have a related to a bunch of things - as far as we know, they are internal problems caused by the sustainer system itself. Right now, we have no complete solutions to the issues we have, only different systems based on different compromises that each of us have accepted.

If you really do want to try this project, first build a basic fetzer/ruby and a 'standard' driver - e.g. single coil around a bar of magnetic steel, with a magnet on the bottom, the coil should be 0.2mm winding wire and have a resistance of around 8ohms... for me that meant about 140 turns of wire (I used 0.23 wire)... wire should stay within the range 0.18 - 0.25.

When you have that built and working, you will get a better idea of how useful the project is to you. Then would be a good time to start asking about other more esoteric stuff like dual core drivers, AGC, piggybacking etc.

If you want something thats just going to 'work' without experimentation, tweaking, R&D, iteration etc., then come back again in a year and see if we're there yet - otherwise, welcome

BTW, we REALLY would appreciate some demo clips of a Fernandez sustainer doing its thing

If you're up for it, it would be best to avoid fancy playing etc. stick to long notes.... examples of the different modes would be nice, and ideally (for me anyway) some clear demos of how it works on a completely clean (no distortion) sound. Also, if there are any weaknesses, examples of them would be good.

If you can do this, then it will be MUCH easier for us to tell you if this project is up to the standard that you are used to!

(www.soundclick.com is a good place to upload mp3 clips)

Col

Hey guys I have a variax that I transplanted to a strat.I bought a fernandes sustainer and I use it with just the single coil driver and I drive it from the analog output of the variax(no bridge humbucker required).It works great.My question is could I use just the ruby circut and a wound driver and drive it the same way as the fernandes?I would like to simplfy things and I only need standard and harmonic modes(no use for mix mode).I would like a 3 way toggle(les paul type) to switch from off to standard to harmonic.Any thoughts?Col, Pete?

I would say - forget it

If you already have a commercial sustainer installed and working, why would you want to go through the difficult and time consuming process of building your own?

(other than for the 'thrill of the chase' in which case, get a cheapo guitar to experiment with)

It is possible (although highly unlikely IMO) that a DIY system could be better than your fernandez, but one thing I can assure you is that it will not be simpler - either to install or to use !

If you want to remove one of the switches or controls form your fernandez unit, the first thing to do is get in touch with the manufacturer and ask them for advice. With any luck they will tell you which wires to connect together so you can ditch the control.

good luck

Col

hehe, One of my harmonic modes IS an all-pass filter

Actually....I was thinking of the circuit as a means to address phase correction and/or harmonic control. It would appear to me (taking the above statement into consideration) that the output cap has an effect on phase angle of the drive signal...thus different responses in harmonics results.

Yes, thats one of the uses of all-pass filters, phase adjustments. You have to remember though, that an all pass filter can only give a specific desired phase adjustment at one frequency - so if you want to 'correct' phase you will need a higher order all-pass filter (like lots of simple ones together), and the circuitry starts getting ridiculous.

Yes, the output cap has a phase response - many components do. It is useful to think of the whole circuit including the output cap (and maybe the driver) as having an overall phase response. This overall response changes when you alter the circuit, changing component values, adding or removing components can alter the phase response.

This can have a big impact on the success (or failure) of a sustainer system - particularly in the case of the 180º phase harmonic mode, I might post some more details about that later.

Your statement in another post that a 1000uF cap would perhaps be advantageous (with high power, etc) was puzzling as, increasing the cap value appears to increase the bass response. As far as I can tell a 220uF is enough to drive the low strings...on my bass experiment...I used a 470uF and I think that that may have been too low even on that.

I said "output cap on circuit causing distortion in low frequency signals (to avoid this completely we'd need a 1000u cap!)"

This is true - with a cap smaller than that there will be some altering of the signal because the cap will restrict the fundamental frequency and cause some phase distortion.

I didn't state that we must use a 1000 cap, just that anything smaller will not allow the pure signal through at low frequencies.

It seems that you like the effect caused by a smaller cap anyway, so there's no need to worry

Additionally, the audibility of this effect depends on the phase response of the rest of the system.

e.g. I have a 220 output cap.

with my dual core driver, I get true fundamental sustain right down to the low E

with my single core driver (everything else the same) I get mixed mode on the lower notes... this will be due to the driver having a different phase response and changing the overall phase response of the system at those low frequencies.

The output cap is just one part of the equation.

Besides less power, what other effects could one expect from a lower coil resistance? By my theories of course, less phase difference/"faster" coil... On this guitar which runs on the lowest possible settings and with a reduced range of effect, perhaps it could do with a little energy reduction.

I think coil resistance is less important than inductance as long as you stay within a reasonable range.

Too high a resistance and you will lose too much energy through heat in the coil.

Too low a resistance and the battery wont be able to supply enough current (and the LM386 would fry anyway)

so as long as you stay within the 4ohm - 16ohm impedance range, you don't have to worry too much.... the difference between 6 ohm and 10 ohm will I think be less than the impact of a similar change in the inductance due to number of coil turns or the core properties (or both)

Hmmm....I am not sure that I can borrow one. Is inductance also measured by frequency also, does not inductance also vary with the frequency of the signal going through it. I know there is some standard (1000Hz or something) but it may well be worth testing the coil at it's operating frequencies too. I am not sure that an inductance meter can do this.

Inductance does not change with frequency, however measuring inductance accurately is very difficult a low frequencies which makes good accurate meters more expensive.

Can you explain or link to an inductance meter somewhere....are we talking about a huge investment....will it offer other features or this enough to make having one worthwhile to myself and others here?

I'm no expert in this area - I haven't got an inductance meter(yet). There are lots of options, from fairly cheap DMMs with an intuctance feature... maybe £30 - £50... to better dedicated LCR meters at £80-£120 and then you have professional stuff at hundreds or even thousands of pounds !!!

I know that the cheap DMM meters use a fairly low frequency and simple method, and that there accuracy is not great, but the biggest question is probably the range.. and at this stage, I'm not sure what range we need... are we talking mili Henries, micro Henries... I dunno. If the cheap option can measure the range that the coils we are making fall within, then they will probably be good enough - I don't thing a +-5% accuracy is going to be a big issue for us

Here is a DMM that has inductance - zoom in on the picture, and look to the upper left portion of the selector switch - that section marked H is the important section (other meters might have L for inductance instead of H for Henries). This meters smallest scale is 2 milli Henries - is that small enough for us? I dunno.

This is a dedicate LCR meter at twice the price of the DMM - it goes right down to less than 10 micro Henries ! and the accuracy is way better.

An option that would certainly make it easier on me is to provide pre-modified pickups with the driver mounted permanently. This would take some of the quality/aesthetic burden from me and give ultimate consistency....plus an extra layer of simplicity to building the system. Of course it also undermines some of the appeal; pickup choice...but hey!

Yes, that would certainly help - but that makes it even less DIY... and besides, you still have the phase response of the pickup (the bridge pickup that is) to consider! that will also have an impact on the overall phase response of the sustainer system... feck me, it never rains but it pours! lol enough to drive you insane innit.

Also, in the kinman designs, and some others it would seem, the lower noise canceling coil can be made smaller and wound with alternative wire with additional mass inserted in the lower coil...or so it seems. Kinmans have additional lower slugs at the coil ends and others like the fender noiseless are wound on two part bobbins attached via screws, again adding mass.

The kinmans are designed to cancel electrical noise - so the humbucking coil section must collect an equal magnitude and range of noise, while at the same time having as small an impact on the magnetic side of things as possible, so fewer turns of thicker wire to give a much lower inductance and a much lower resistance, but still working well as an aerial to collect that electrical noise.

Thats no good for us though - we're not worried about electrical noise, it's magnetic radiation thats killing us, so if we were trying to cancel our problem using coils, we would need to somehow make the strength of the field generated by the EMIbucking coil match that of the main coil.... using fewer turns of thicker wire will not work!

But using enough turns to match the field magnitude will kill efficiency because it will also cancel the drive field...

Thats why utopian isotopes idea is so promising - it potentially solves that quandary! by moving the cancellation coil over to the pickup, the field strength can be equal at the pickup due to proximity, but have very little impact on the driver because it is a much smaller coil.. I think the idea needs a lot of work and refinement, but it has a lot going for it.

It is an area that I have often thought I'd like to explore...and to do so "scientifically"...I guess an inductance meter may also be in order. I have also had a desire to include aluminium into pickup designs...would such materials add to inductance even though not magnetic, they can harbor eddy currents through their conductivity. Similarly neodyminium magnets...I had a design mapped out and someway towards constructed that used ball bearings in the poles with neodyminium magnets attached and some machine bolts that could be screwed to adjust the mass of the internal core.

I am imagining for these kinds of things, that inductance is only a small part of the equation and that frequency response and resonances are more of an issue...

I don't think aluminium or neodymiums will have a big impact on the inductance, but they won't have a big impact on frequency response or resonance either!

Anything that does change the inductance will change frequency response, resonance characteristics and phase response because they are dependent on the Inductance.

Inductance is probably more important that any other single electro magnetic concept for this project - at least in terms of understanding why things work (or don't) the way they do... understanding what parts of the construction of a driver or pickup have an effect on the inductance, and how the inductance is linked to the phase response, the amplitude response and the field strength...

magnetic field strength

Somewhere in the back of my mind though, there has always been a hope that a signal could be sent into the pickup coils to cancel noise and fizz caused by EMI and perhaps counteract forces that require such elaborate switching.

I have wondered in the past about using some sort of negative feedback(or feedforward?) 'error correction' to limit the fuzz, which is similar to what you are talking about...

A cancellation signal would have to be buffered, and it would most likely only have to be small, but if it doesn't have the same phase shifts that the driver and pickup have imparted on the noise we are trying to cancel, then it just won't work. Additionally, if any of the noise is caused by clipping due to interaction between pickup and driver, this approach will not work. (utopian isotopes loop idea might still be able to help with the driver/pickup transformer clipping noise because it has the potential to create an opposite transformer clipping effect )

cheers

Col

ok...BTW...I found an interesting circuit last night col (off-line) that is an all pass filter that adjusts phase to frequency while keeping a constant level...simple op-amp design, but something I have never seen before. Might very well have application for phase compensation circuitry or enhanced control. Will try and find a way to send or post this soon.

later pete

You need to pay attention in class boy...

hehe, One of my harmonic modes IS an all-pass filter - I explained it in detail ages ago - it is definitely a useful module for us to use, but it's important to also understand it's limitations. Anyway, there's no need to worry about sending it to me.

=======================================

Here's something I discovered a while ago that you should have a think about as it will have an impact on your 'piggy-back' system. I only just realized the connection or I would have posted about it a while back...

As you know I've been trying to come up with a way of measuring the inductance of coils. Part of that has involved making coils and trying out proposed measuring techniques from books and the net. I haven't been able to verify the accuracy of any single measurements yet, so haven't posted anything, but one thing I have been able to do is get a rough idea of the relative strength of different cores.

One significant thing I have discovered is that if you add some core material to the core, it has an effect even if it is not within the coil... say you have a driver with the core protruding a few mm from the bottom - if you stick some more iron to that, even though it is further away from the coil, it can still have a significant impact - as in 30 or 40% more inductance...

So how does this affect your system?

Well, if your driver is using an existing core from an existing pickup, the performance of the driver - drive strength and frequency response - will depend to some extent (more than you may realize) on the specifications of the pickups core!

The effect it has on the frequency response should be somewhat reduced by the fact that you have an impedance that is mostly resistive due to the thin wire with fewer turns. However, the difference between one pickup core and another could be enough to cause the system to function poorly on some guitars....

I don't think this will be a killer problem, its just something to be careful about - test as many types of pickup as possible (I'm sure you're doing that anyway)...

One way to think about it is that having a core with twice the inductance factor (a realistic possibility) would have roughly the same impact as going up to the next wire gauge (assuming you were to stick to 8ohm resistance)!, and you know through experience how that can effect the functionality.

With any luck, there is some sort of rough 'standard' for pickups, in which case you would only have to worry about the extremes like 'The Invader' at one end.. or some weedy retro copy type of things at the other...

If you can borrow an inductance meter from somewhere, you should test the inductance of one of your piggyback coils without a core, then again when mounted on different pickups - this will give you a much better idea if this is something you should be concerned about. If you find that some pickups can give readings 60% or more greater than others, you should take it seriously as a potential problem.

cheers

Col

If you bow E and A strings simultaneously, you will get totally different sound, compared to plucking; almost like a choir of those two strings; no clean tone is available there.

Nah, If I bow E and A strings, I'll just get a horrible scraping fingernails on a blackboard sound... but maybe with some practice...

But seriously, it certainly doesn't sound like a 'clean' tone on an electric guitar (which in turn doesn't sound at all like a sine wave).

It does have a harmonically rich tone as does a distorted guitar but I would describe the sound more as a resonant sound than a distorted[as in clipped] one. A distorted guitar sound is similar to a square or pulse wave, while a bowed string has more in common with a saw wave (most subtractive synth 'string' sounds are based on unison saw waves)

If we can somehow remove the interference noise from the system while still allowing 'distorted' drive signals, it would also be possible to experiment with using various wave patterns and harmonic overtones as drive signals to try and change the timbré of the sustained guitar sound.

If you are really interested in going for a sound that is not a traditional guitar sound, then maybe you should try using a piezo pickup with your sustainer... that way, there will be no trouble with EMI, and you can try all sorts of different drive signals without worrying about interference noise or squeal.

cheers

Col

That's why I came to conclusion, that instead of hunting too clean sound and feedback, more overtones should be added.

Hmm, yes, more overtones for a more realistic loud amp sound would be good, but fuzzy interference noise isn't the way to get that result IMO

"Natural harmonic distortion" seems to be a major part of that desired tone.

----------------------------

As I've read these postings, distortion or fuzz/fizz seems to be the problem on the sustainer-circuit(s).

Maybe it is just pure nature of tone and feedback, that tries to push out, in form of fuzz/fizz?

-----------------------

Nope, the fizz/fuzz we are talking about is coming form distortions in the drive signal that are being 'heard' by the pickup due to EMI (electro magnetic interference).

What you are driving at is a valid point - the fact that the timbré of a guitars sound changes when there is feedback. However, there is no reason why we shouldn't get this effect from an electo-magnetic sustainer without grunge, fuzz or fizz. That singing full sound of a guitar that you get with a valve amp up loud - it doesn't need a lot of crunch, its more about volume and a good axe (and a good player of course).

One of the limitations of my circuit (the last one that was 'released') is that its maximum drive is not enough to get that singing kind of effect... it can produce reasonable loud sustain and provides lots of new creative possibilities, but it is by no means ideal.

I have some other design ideas on the drawing board, so maybe at some point I'll get some new circuit posted with clips... right now, I'm more concerned about the driver side of things.

If possible, get yourself a violin bow(some rosin also) and test things. Compare sounds between bowed and plucked note, so you'll see(hear) what I mean.

I have a pretty good understanding of what happens with a bowed string - how it sounds, and a basic understanding of the physics (one of my other interests is digital signal processing, and I have read some papers on modeling a bowed string digitally. The math is heavy, but there are good explanations of what's happening when a string gets bowed.... and it has very little in common with soft or hard clipped fuzz...)

I have a sneaking suspicion though that there is a little more to the fuzz issue.

My approach (the first one to be proven to work) to eliminating grunge/fizz/fuzz completely has been to ensure that there is minimal distortion in the drive signal, this means that the EMI can still be there, but you won't hear it because it is the same signal (or close enough)... basically it is masked by the desired 'correct' sound from the string vibrations.

Pete has often talked about the AGC driven sound not being loud enough. I have disagreed, because I know it is loud and the strings can be driven hard. So why doesn't it sound that way to Pete?

Maybe because the sound isn't as rich, its loud, but the overtones that make it sound loud are missing!

It's possible or even likely that driving the string with a heavily clipped wave changes not only the level of audible unwanted interference noise (that we don't want), but also the richness of the string vibrations and resulting guitar tone.

So if thats true, How do we throw the bathwater out without losing the baby ?

hmm, I have one or two ideas about a possible improvement to the 'loop' squeal canceling thing that might help... but other than that, I'm stuck for now.

cheers

Col

I think part of the problem is that selling a ready-to-install system doesn't actually solve any of the problems - there are already off-the-shelf systems available... The issue here is how to make a 'DIY sustainer' more accessible, any ready (or part) made system is by definition not a 'DIY sustainer' !

Well...one man's Ikea is another's DIY. The issues of installing and rewiring alone is enough to qualify for DIY, IMHO even with a commercial system. Some would say that anything short of designing and building something without help is not really DIY, but that kind of makes things like this thread a little of a nonsense.

As I see it, many are capable of installing and making valuable contributions but lack skills and in some areas, even the materials to successfully complete the project.

I don't know anyone who would consider Ikea to be DIY... maybe DIY interior design... but not DIY furniture making

The issues of wiring and installation exist with all the commercial units - unless you buy a full sustainer guitar.

IMO people who can't solder a simple circuit together won't be able to do this project - they should buy a ready made 'install it yourself' system - mind you, if they can't build a fetzer/ruby, they really need to get someone else to do their installation as well!

For all the others who found this thread because they were looking for information about how to build a sustainer, the problems that I listed still exist, and providing a ready made circuit and driver doesn't address any of them.

As I stated, I think it is possible to solve many of these problems, but it isn't easy

Here's a little dialog to highlight my point:

Jim:

Pete:

Jim:

Pete:

Jim:

Btw, that was a very interesting gotcha about the earthed pickup poles. I suppose it would be OK as long as they are insulated effectively from the strings, although that adds some extra issues about aesthetics and installation complexity.

About the lack of response in 'full' harmonic mode. I've had some problems with that as well. I wonder if the problem is that the circuits and drivers are getting good enough so that they can drive the pitch right up beyond the working bandwitch of the guitar amps... or above our hearing range... ? or maybe just beyond the frequency that the strings can effectively(for our purposes) vibrate at ?... maybe some sort of low pass filter might help to limit the top frequency of the drive signal ?....hmmm I guess it will need more thought. It is certainly an important thing to understand.

As far as a variable output cap, one thing to try (it might not work because of phase response... not sure) is to have a big (200uf)cap wired in series with a pot, then have a small cap(10uf) wired parallel to them... so when the pot is at 0 resistance, the capacitance is 200 + 10, while when the pot is at max - depending on the pot value - the capacitance will be much closer to 10uf... certainly would be fairly simple to test. although I feel that the cutoff is way too gentle for a really useful change in effect - it would probably make more sense from the users point of view to have a 3 way switch with small medium and large caps... maybe even just small and large...

Edit: yep, a 200ohm pot (or as close to that value as possible, set as a variable resistor) in series with a 220uf cap, and put a 10uf cap in parallel with those Then swap that for your existing output cap. That will get you roughly what you want - it will range from what you would get with a 230u cap to what you would get with roughly a 10u cap

cheers

Col

It is certainly possible to apply strong sustain while using gain control... I'm also sure that it is possible to remove audible fuzz and still use a 'high infinite sustain mode' (whatever that is).

This means...no AGC...power applied only to the limit of the ability of the strings to be driven.

For the strings to be the limiting factor, the power would needs to be huge - we would need a mains driven unit and a significantly larger inductor !

There is a point after the signal starts to clip where adding more gain doesn't have much impact on the effect other than to increase background fuzz and waste the battery... It is possible to use AGC to bring the quieter signals closer to this level, and prevent the strong signals from going way over it... so what you get is a more responsive more alive system when playing difficult string/fret combos.. and at the same time slightly better efficiency with no major difference in response in other areas...

AGC can help get closer to the natural limit of the system without exceeding it (exceeding the limit is not going to make the thing wilder and more fun, just noisier, less efficient and more trouble)

this is why sustainiac use the word grunge

Do they...where? I am of course at a disadvantage in that I do not, nor never had a commercial sustainer to compare what I am doing to what is already available.

this is taken from the sustainiac stealth plus installation guide (available on their website)

"11. GRUNGE/SQUEAL TEST AND ADJUSTMENT OF MAGNETIC FIELD CANCELLATION TAB:

If you have squeal and/or grunge in your guitar signal, you might need to add a tab...."

If it weren't obvious before, and it surely isn't so much anymore...my "commercail" intentions were purely to rectify these problems. Of late, I am thinking I should step back a little and provide more for the development and the DIY than to try and compete with a complete system at this stage.

I think part of the problem is that selling a ready-to-install system doesn't actually solve any of the problems - there are already off-the-shelf systems available... The issue here is how to make a 'DIY sustainer' more accessible, any ready (or part) made system is by definition not a 'DIY sustainer' !

cheers

Col

I also have a feeling that it may not be able to apply strong sustain (non-gain control) without this effect occurring at high, infinite sustain modes.

It is certainly possible to apply strong sustain while using gain control... I'm also sure that it is possible to remove audible fuzz and still use a 'high infinite sustain mode' (whatever that is).

If one tries to create feedback naturally, either via high volume or by a particularly resonate guitar or by things such as pressing the guitar head onto the amps speaker baffle (a poor man's model C sustainiac and old studio trick)...some distortion seems to be evident.

In my experience, that depends on guitar and amp settings - of course, the tone and 'quality' of the sound will be different, but it is possible to get sustain at high volumes without fuzzy distortion.

However, a clean sound is not only the "holy grail" of this project...it is necessary for efficient operation to aim as far as possible towards that end. To my mind, I have always sought to retain the characteristic tone of an instrument and the device used to enhance technique.

Certainly it would seem that, besides all that is said above, the shear physical vibration of the strings in such an unnaturally powerful way will cause some distortion, if only in the physical way in which the string vibrates.

Now the definition of 'distortion' is getting very blurred indeed - to the extent that it is losing its meaning!

I would suggest that we stick to a definition like 'clipping of the signal resulting in a fuzzy or crunchy quality to the sound' or 'a noticeable fuzzy or rough quality to the sound'... this is why sustainiac use the word grunge.

AGC can restrain the effect a little too much, loosing the dynamics and touch sensitivity even though providing infinite sustain. As one of the possible characteristics of the device is the potential to push the strings hard and for notes to bloom and evolve, it is not therefore something I yearn for to the nth degree. Compromise and individual choice should be the key. If a players style is distorted already and they want this effect...then go for it...but there is a beauty in subtlety and, while distrotion can be added to a clean signal...it is near impossible to remove.

Any technique can have negative effects if not used sensibly - AGC is a technique that can be used in many ways, and certainly doesn't prevent blooming harmonics or 'pushing' of the strings.

One thing I'm currently looking at is the use of a less extreme AGC that will just prevent the drive signal from causing clipping in the circuit allowing weaker signals to be amplified more without getting fuzz.

This is a depature from my other circuits which effectively switched off the drive when the signal went over the threshold.

This type of AGC - basically a brickwall limiter - allows you to have your extreme gain if you want it (distortion and all), but it also lets you boost the weaker strings/frets, so its a win/win.

In my later circuits, I have aimed to achieve both of these effects by linking the AGC into the gain and the zobel network in order to derive a range of effect.

I've noticed you talking about your new AGC interacting with the zobel... how does this work?

My understanding of what a zobel network is and how it functions suggests that if there is an interaction, then its not actually zobel network! so whats happening ?

Especially with the DIY version with very simple preamp/amps like the F/R solution will most likely necessarily suffer some distortion effects...but still, with good technique, this can be limited to some degree. Col has been able to get far cleaner sustain because of a far more sophisticated automatic gain control (AGC) circuit and using a dual coil, EMI cancelling driver design.

hehe, I've recently taken an old single coil driver and put a 'cage'... not a shield, but as I explained a while back, an external continuation of the core that increases the inductance and focuses the feild.

I swapped this for my dual core driver, and it works well. It has similar low distortion characteristics to the dual core. unfortunately, the drive is not quite as good, but I believe that is due to not having a good enough magnet - certainly not as good as the one on my dual core driver.. still, promising results.

One other thing to consider is that with a fixed position driver exciting the strings in the way that it does, with phase differences, low amplifier headroom and other things coming into play...these strings, even if not driven hard and restrained by AGC will most likely not be vibrating in a wholly natural way and so could be considered to be distorted. Mi9nimisation and compriomise, especially with DIY versions, but even with commercial units...is key.

Hehe, you might as well say that plucking a string distorts it's shape and forces it away from its natural equilibrium, so for a guitar to make a sound you need distortion - quite pointless though :-p

Web site creation is very new to me, but the potential is there and I have taken the plunge to be able to provide the resource and to utilize it here and elsewhere.

Good luck with the website - it's definitely overdue, particularly for your commercial sideline. If you need any advice or suggestions about web dev, PM me, I've done some web development - right now I'm working on a database driven Library catalogue!

This thread has come a long way (since May, 2004 is it?) with so many contributions, but I still feel there is something missing. While further development is necessary, engaging and entertaining, I still feel that a definitive guide to creating a successful DIY version is lacking. It would seem that this is something important for the building and improvement of the device.

Exactly what needs to be explained (should it go back to explaining the fundamentals of how it works and definitions of things like EMI and why it is a problem?) would be interesting to know. I have tried to switch my way of working (typically penciling on the back of envelopes) so that layouts and such are far easier to read (I hope you have noticed this in recent submissions)...now all I need to do is reduce the word count!!!

Yes, it seems silly that there isn't yet a definitive guide, but there are very good reasons why.

Unfortunately, we're not yet in a position to provide anything definitive yet!

As we know, this project depends on many variables. Some of these variables are 'external' things like the guitar, the pickup(s), the available materials for the driver magnet, core material, wire etc.

Unfortunately, in order for us to provide a single definitive guide, we need to provide a method for people to measure and assess the resources they have and tell them how to use their results to adjust parts of the system in order to ensure a successful outcome.

If we cannot provide this kind of instruction, then the guide might as well say - "read the thread, then use trial and error building many divers and circuits until you get one that kinda works for you!".

I am some way towards a method for understanding the components of the driver in a way that would allow for this kind of methodical approach, but I'm unfortunately limited by lack of equipment and money

The current problem is how to measure the inductance of small coils accurately enough to be useful. I'm stuck because to develop a method, I need some sort of trusted accurate meter to compare the diy approach results with - otherwise, I can't know if they are close enough or completely useless.

IF we can provide a plug'n'play method for measuring inductance and mathematically deriving the other variables - particularly inductance factor -) from the results, we can tell folks:

• how to choose the best materials from what they have available
  
  
• if these are going to be good enough
  
  
• what dimensions to make the core
  
  
• how to calculate the correct number of turns of wire for their core
  
  

When they have built their driver, they can use the same techniques to measure it's actual inductance and calculate its impedance

then we can explain how to tweak their circuit for optimum performance with that driver.

If we can't provide step by step plug'n'play instructions in this way, then folks will need skill, knowledge, patience and huge resources of time and perseverance, it wouldn't be a Do It Yourself sustainer, it would be a Develop It Yourself sustainer.

For a while now, I've been at the stage where I've got all the equations and just need a method for calculating inductance with reasonable accuracy.

Then I would have to build a few drivers to test some different theories and do some testing and work out what the best basic setup is as a compromise between simplicity and functionality.

This is the main reason why I've not done any further practical driver research - I'm not willing to do anymore trial and error development unless it is at least partly driven by analysis. I have lots of different ideas to test - some will be valid, and some not, but without reasonable accurate measurements and calculations, it is impossible to know if something failed because the idea is bad or because the implementation specs were outside of the 'success envelope' for that approach..

enough, cheers

Col

Seems that some of that annoying fuzz-element came from lower strings rattling against frets.

...

Clean tones sounds much better now, although there is still subtle amount of fuzz (clipping/distortion).

Hehe, there are so many different possible causes for background fuzz in this project: here are a few

1. EMI: in these cases, distortion in the unwanted interference signal will be audible, so it can be at least partially avoided by making sure there is no(or very low) distortion!.. (of course, there are as many possible causes of the distortion as there are ways for EMI to get from driver to pickup )
   
   • EMI from driver reaching pickup through the air
     
     
   • EMI from cables reaching pickup
     
     
   • EMI from circuit reaching driver
     
     
   • guitar strings acting as an EMI bridge between driver and pickup
     
     

[*]pickup and driver = parasitic transformer causing clipping in interference signal

[*]Shared earth of pickup and driver allowing one to load the other

[*]output cap on circuit causing distortion in low frequency signals (to avoid this completely we'd need a 1000u cap!)

[*]physical vibrations from driver reaching pickup

[*]strings hitting frets

[*]faulty components or bad construction or driver and/or circuit

[*]other faulty gear like stomp boxes or amps causing fizz that we didn't notice before because we didn't listen hard enough

[*]The ghost of Jimi diggin' our groovy sustainer antics

I'm sure there are others that I've missed, but you can see the trouble - fuzz removal is like peeling an onion.

Col

How do you post an audio clip here? Since Photobucket doesn't allow mp3 or any other audio-formats? Do I really have to have my own website, where I can upload the files, so that I can link them here ?

There are a number of sites like photobucket but for audio - the one I use is Soundclick

check out my sustainer demos on soundclick

hmm, I need to get some new demos done... so I need to do some experiments for that to be worthwhile...

cheers

Col

To the point: My present sustainer-circuit gives too much juice to the driver (but no squealing, because of the loop), so that it puts out pretty raw Tony Iommish feedback, even at minimum setting.

Sounds interesting - here is a crucial question for you:

When you set your amp and guitar for a crystal clear clean tone, does the sustainer add some (subtle or otherwise) fuzz to the sound ?

You see, the thing I and many others are concerned about is not squeal - that's easy enough to deal with - the really difficult thing to deal with is background fuzz. So do you get a nice clear clean sound with loads of feedback ?

can you post an audio clip ?

cheers

Col

With this in mind, it may be possible to setup the loop with a low resistance trimpot so that 'tuning' can be done with a screwdriver!

Yes it would be an ideal solution, but since that loop is more capacitive/inductive element, additional resistance would add unnecessary stress to sustainer circuit.

Of course, that is a very good point. Maybe the only way to get it working with a trimpot would be to use additional electronics to split and buffer the signal - much more work would be involved.

Is there any way to wire the loop in parallel with the driver coil, keep the overall impedance at a sensible level throughout the range of the trimpot using a resistor, and still get enough current through the loop without impacting efficiency? probably not...

I certainly like the down home DIY simplicity of the basic wire loop

-------------------------------------------

btw, utop iso, are you using a basic mechanical switch to flip the driver connections as your harmonic switch?

No switches used, just plain point-to point wiring, as I am experimenting things.

...No inverter is in use; just an eq, that acts as a buffer with some gain on certain frequencies.

If you have a spare op-amp section available, you can try a variation on this kind of thing

a switchable op-amp inverter - just need an SPST switch.

IIRC, I used a different variation on this as part of my multi-harmonic mode switching setup

just google switchable op-amp inverter.

cheers

Col

Hehe, I really need to get some testing done on this loop thing - I'm having too many ideas that shouldn't be even considered before getting a better understanding of the device.

Here's a good one:

Assuming the loop CAN cancel interference produced by the driver, that would free us from worrying about distortion in the driver signal. This means that we could happily use a square wave to drive the strings - this could give us an huge efficiency boost. Think that the whole concept of class-d amps is that the power amp can be amplifying a square wave and you get the picture - think class-d efficiency but without the headscratching and complexity and without worrying about SMD components.

It also makes AGC much easier... for example my first circuit using LM13700 would have been fine if is didn't introduce so much distortion, but if we can cancel that distortion, then that circuit becomes viable again....

I wonder how much cleaning the 'loop' can do for us? enough to limit squeal... or enough to remove fuzz and fizz as well ?

---------------------------------------------------------------

Pete's suggestion about ditching the loop and directly applying some driver signal to the pickup is also very interesting... I guess we would need to buffer the 'side-chain' with an op-amp(maybe inverting?) and have a gain control, or a passive volume control for 'tuning'...

I wonder if this approach might need some Phase adjustment? while the loop doesn't?

If we could get side-chain approach to work, it would certainly simplify the installation and make 'manufacture' cheaper, although the op-amp would probably use more power than the loop does, and if it needed phase correction, the circuitry suddenly becomes non-trivial and control cavity size gets even more critical... still, its definitely worth some experimentation

cheers

Col

Polarity seems to be one of the key elements. You can change loop polarity by switching its wires, or just flip the loop upside down. If polarity isn't right, it fails to work, and you don't get much audible results.

Ah, that is good news - if that wasn't the case, then my idea as to how it might be working would have been wrong

Note on loop construction: Wire loop isn't too stable, unless it is glued on something that preserves its shape. Nevertheless it is acceptable, but I still encourage you to make the loop out of an aluminum, brass, steel or copper plate; so that it keeps its shape and is more solid. Doesn't matter how thin plate is, as long as it doesn't vibrate along the (electro)magnetic flux. Solid loop should take care of that squeal-phenomenon.

One thing to watch out for is that steel has a fairly high magnetic permiability, so putting a lump of steel near your pickup could have some negative effect on the sound of the pickup !

I would stick to aluminium or copper for this reason.

My guess about the need for 'tuning' is that if the wire loop is creating more interference current than the driver it will cause squeal/noise and if the driver is creating more interference current it in turn will be the culprit, 'tuning' is basically the process of tweaking the level of interference that the loop is causing so that it matches the driver and they cancel

With this in mind, it may be possible to setup the loop with a low resistance trimpot so that 'tuning' can be done with a screwdriver! if so, the loop could be fixed shape and size and fixed position!

If this is possible, it would make for a much easier and potentially better looking install.

btw, utop iso, are you using a basic mechanical switch to flip the driver connections as your harmonic switch? if so, that would explain why one mode creates more noise than the other - it may be that for others, tuning should be in normal mode - basically whichever is noisier ...

Depending on the pickups polarity, there is only one 'correct' polarity for the driver. In the past, some people have had more noise in normal mode and some in harmonic - me and another poster (can't remember who but he explained it better than I did?) came to the same conclusion as to why - something to do with shared earth and polarity of the coils, but anyway - my understanding is that either normal or harmonic mode will be noisy unless you use an op-amp inverter in the preamp for mode switching

cheers

Col

...It does the trick same way as wire loop when positioned between driver an pickup.

Strange thing is that it works also behind the pickup!!

Since I have LP Junior-type bridge, under it is enough room to place loop there.

The more I experiment with it, less I understand...

It does actually still make sense putting the loop behind the pickup...

Consider my original explanation...

Now ignore the part where I said the loop has to be close to the both the driver and pickup!

The reason it doesn't have to be near the driver is because it already has the driver signal flowing through it!

So the important thing is that it must be near the pickup.

Have you tried positioning it around the pickup ?

If that works, it would be ideal in terms of minimizing space requirements and making it look better.

Positioning the loop is quite critical anyway, so it takes some "tuning" to find that sweet spot, where squealing goes away.

What kind of adjustments do you make to 'tune' it in ?

What happens when you switch the polarity of the loop, does is then fail to work ?

As soon as I have a little time, I'm going to play around with this a little

cheers

Col

@ psw: I meant that the loop "binds" or limits driver's magnetic flux, and prevents its affection to PU. Because that loop is between driver an PU, it allows the driver to concentrate magnetic force on strings, instead of PU.

@col: You just might be on the right track about that loop thing.. somehow your text makes sense to me. In practice it is simple, but in theory it is far beyond my knowledge.

Maybe the pickup "sees" drivers magnetic field as a parasitic element? And the loop works as a sink or drain that absorbs most of the parasitic forces?

I'm pretty sure that the loop doesn't work as a 'sink' or 'drain'

The only way to stop the magnetic field from the driver reaching the pickup is to use some sort of magnetic material to guide the field from its north pole to its south pole. Your loop doesn't do that - its non magnetic copper, and it is too thin even if it was some sort of magnetic material.

What I'm suggesting is that your loop is introducing MORE interference !

If I'm correct, this new interference is (at least partly) the opposite of the interference that we want to remove, and therfor the two are canceling each other.

This sounds a lot like the ideas I presented way back of active shielding. If people have stuck around long enough, this is where I proposed a second reverse wound smaller coil around the inner sustainer coil. Experiments were never done into this and as I recall, or perhaps it was my own reservations, it was felt that sacrificing some coil to provide an opposite effect would be detrimental to performance in that format. Perhaps there was more potential in the idea after all. Maybe an outer coil of very few turns and a thicker wire would be of use to produce the effect.

This new loop idea is IMO very different from the 'active shielding' ideas that have been posted previously. I don't think any of them would have worked well due to efficiency issues caused by signal cancellation... Assuming this new idea works (still un-verified) it would be because of the layout - making a large loop that goes very close to both driver and pickup.

And the small detail that (At least IMO) it isn't doing any controlling, restricting or 'shielding'!

I guess we'll have to wait and see how the verification testing goes

cheers

Col

What leads me to that loop? Hard to say. First I tried to eliminate that squeal with magnets. Magnet tends to excite strings too much: no result.

Resistor in series with driver: no result.

Piece of steel between driver an pickup: no result.

Iron does just the same: no result.

So nothing magnetic fits between the driver and pickup.

Because the driver uses strong electromagnetic force, that is the same force we can use to create physical "bridge" between pickup and driver, so that magnetic flux doesn't float too much anymore. Instead it flows from the driver---->strings--->pickup.

Interesting idea this loop of wire!

(assuming it works and isn't just an unfortunate coincidence) I would like to suggest a different reason as to why/how it might work - fun of course, I'm just brainstorming

Once you've removed the fizz caused by distortion in the driver signal and problems with earth wiring, One of the most likely suggestions as to the cause of the remaining interference noise between the driver and pickup is the idea of the pair acting as a parasitic transformer.

The pickup has many orders of magnitude more turns than the driver, so when the two are close together, any voltage induced in the pickup by the parasitic transformer will be much larger than the system can cope with and will be heavily clipped.

I wonder if this loop of wire, if twisted in the opposite direction to the driver can somehow cancel out some of this by creating an opposite polarity parasitic transformer ?

It is only a single turn (or small number of turns), so its inductance will be much lower than the driver, but it is closer to the pickup and furthermore, it's turns ratio is significantly higher... its never going to be powerful enough to completely cancel out the driver, but maybe there is enough to have a noticeable impact on the quality ?

.....................

hmm, looking at the picture again, the loop goes very close to the driver and the pickup... maybe the current induced in the loop by the driver is in turn inducing some cancellation current in the pickup ?

This seems like the most obvious answer... if the polarity is opposite to the polarity of the current induced in the pickup by the magnetic field of the driver, then there should be some noticable cleaning up of the sound.

So rather than being a transformer as described above, its really a way of passing a current from the driver to the pickup that is roughly equal and opposite to that produced by the interaction of their magnetic fields...

The big difference in induction due to the very low number of turns and lack of a core is made up for by having the coil/loop physically very close to both driver and pickup (iirc, magnetic field strength drops off at the square of the distance)

Of course, because it is only one (or very few) turns of wire, it will not have any significant impact on efficiency

This seems like it could be a great breakthrough (assuming of course that it really works)

Either way, it's great to have some new ideas injected into the discussion - thanks Mr Isotope

cheers

Col