Sustainer Ideas

[JER00N]

Does it matter which size of the magnets is towards the strings? Because I have 4 small magnets next to each other, but then the first is with the N-pole towards the strings, the second has the S-pole towards the strings, the third the N-pole etc... I can image this somehow distorts the field generated by the coil.

[onelastgoodbye]

Does it matter which size of the magnets is towards the strings? Because I have 4 small magnets next to each other, but then the first is with the N-pole towards the strings, the second has the S-pole towards the strings, the third the N-pole etc... I can image this somehow distorts the field generated by the coil.

they need to face the same way. Doesn't matter which side it is, but they either have to be all N up or all S up. Otherwise their electromagnetic fields cancel out (err...I think), which is mabe why you only get good response on the ead strings.

Also, what did you use for potting?

[JER00N]

Also, what did you use for potting?

I used Bison Kit. I think the coil has become a bit loose here and there because I can hear the sustainer makes sound (the same sound as I hear on my amp). I'll have to redo that a bit.

[Radiotrib]

Does it matter which size of the magnets is towards the strings? Because I have 4 small magnets next to each other, but then the first is with the N-pole towards the strings, the second has the S-pole towards the strings, the third the N-pole etc... I can image this somehow distorts the field generated by the coil.

Dag Jeroen, greetings from Den Haag.

I think I've missed something here, because I was under the impression that the sustainer driver didn't have any inbuilt magnets, only a ferro-magnetic core, and it derived its magnetism from the a/c in the coil?? .. Did I miss a page or so ??

I can't find the time to make anything right now - just bought a new car that needs a lot of TLC - so I speculate a lot, and I've been toying with the concept of a tall thin driver instead of a short fat one .. like this ...

Build a short fat one like the current designsd

Fit the top/bottom of the driver with iron plates so the core is extended to cover the whole top/bottom of the driver

turning the driver on its side so the edges of the plates point towards the strings (like a pair of blades)

When excited, that should cause one plate to be positive and the other negative. It would still provide magnetic attraction with a very concentrated field across the small space between the plates, but with almost no peripheral field leakage, since the plates would shield the core, and only radiate/propagate a field from their sharp edges. I dunno, but this might help with the EMF issues.

On the down-side, it would result in a sustainer driver which would need to be installed alongside, or in place of, a standard pickup rather than on top of it.

Any comments? .. am I "barking up the wrong tree" again ???

[JER00N]

Hello Radiotrib,

Yes the sustainer requires a magnet. I don't know exactly why but without one, it won't work. There is a sort of initial magnetic force needed which is provided by the magnet. In a normal pickup, the poles are magnets too.

Yesterday I took some metal sheet (magnetic) out of an old VCR (some parts of it, like the power supply) are built inside a metal cage to absorb electromagnetic fields to prevent distortion of the video signal) and placed it between the sustainer and the bridge pickup, but it didn't really help. I don't know if it's really necessary to also place a shield between the neck pickup and the sustainer, even if I don't use the neck pickup?

[Truth_David]

Hello Radiotrib,

Yes the sustainer requires a magnet. I don't know exactly why but without one, it won't work. There is a sort of initial magnetic force needed which is provided by the magnet. In a normal pickup, the poles are magnets too.

Yesterday I took some metal sheet (magnetic) out of an old VCR (some parts of it, like the power supply) are built inside a metal cage to absorb electromagnetic fields to prevent distortion of the video signal) and placed it between the sustainer and the bridge pickup, but it didn't really help. I don't know if it's really necessary to also place a shield between the neck pickup and the sustainer, even if I don't use the neck pickup?

Hi Jeroon, Zfrittz6 has asked me to translate and pass on this message to you, I hope that it makes sense (my knowledge of electronics is very very limited, so I am unable to judge for myself):

One cannot use aluminium for insulation, nor any non conductive material; one needs to insulate them (the pickup coil and the driver coil, I believe he means) using very fine tin sheets insulated from one another; he used sheets of tin made from painted Pepsi Cola cans in order to finish up with 0.5cm insulation all around the coil together with its magnet and the sides of the pickup that face the middle of the coil, which also has to be insulated in the same way, but it is most important to check that the tin laminates adhere well to whatever magnet is used.

Regards,

David

[JER00N]

Hi Jeroon, Zfrittz6 has asked me to translate and pass on this message to you, I hope that it makes sense (my knowledge of electronics is very very limited, so I am unable to judge for myself):

One cannot use aluminium for insulation, nor any non conductive material; one needs to insulate them (the pickup coil and the driver coil, I believe he means) using very fine tin sheets insulated from one another; he used sheets of tin made from painted Pepsi Cola cans in order to finish up with 0.5cm insulation all around the coil together with its magnet and the sides of the pickup that face the middle of the coil, which also has to be insulated in the same way, but it is most important to check that the tin laminates adhere well to whatever magnet is used.

Regards,

David

Thanks for the translation. So the solution is to use just a lot of tin sheets. But I don't really understand your last sentence, should the tin layers be connected to the magnet or not? Because when they're physically connected to the magnet, the layers become magnetic too thus being not a shield anymore. Furthermore I don't really understand if the layers have to be connected to eachother or not... Maybe some photo's could help?

Thanks,

Jeroen

[zfrittz6]

hola.

No las capas no deben de tocar el iman ni la bobina, deben de estar separadas y aisladas entre si, para eso utiliza cinta aislante y luego las latas con pintura.

Saludos.

hello. Layers do not have to touch to the magnet nor the coil to them, must be separated and isolated to each other, for that it uses insulating tape soon and the tins with painting.

Greetings.

Edited January 18, 2007 by zfrittz6

[Primal]

I think I've missed something here, because I was under the impression that the sustainer driver didn't have any inbuilt magnets, only a ferro-magnetic core, and it derived its magnetism from the a/c in the coil?? .. Did I miss a page or so ??

The driver requires the presence of a magnetic field in order to operate. Think of it like a boat propeller. If you turn on the motor out of water, yes, the propeller will push some air, but you won't go anywhere. Put it in water, and... well, you know. The boat propeller is most efficient when it is in the water.

Now, extend the analogy to the driver coil. Yes, the coil will produce a magnetic field by itself, but it is far more efficient (at the level of power we are utilizing -- this driver ain't no electro-magnet) for the driver to manipulate existing magnetic field (and possibly using resonance to increase in power? Hey, maybe that's one of the keys to getting the driver to work "faster"?).

[JER00N]

I've been thinking about the mid-sustainer again and I wondered if shielding would help anything at all. Magnetic field lines go from north to south along the outside of the magnet, like I have sketched here:

As you can see the lines go through the humbuckers, 'jumping' over the shielding. Because of this I'm a bit skeptical about the whole magnetic shielding thing, even if Zfritzz' shielding works.

I think we should also concentrate on the speaker/piezo system because these systems won't give us the magnetic fields hassle. Has anyone got results from these kind of systems? I tried with a piezo on the tremolo block but I guess it's not powerful enough. A speaker would be more powerful but then you can't move the tremolo anymore...

[zfrittz6]

Hola a todos.

Si, es asi como funciona el magnetismo, pero lo que hace el blindaje mas que blindar es conducir el campo magnetico a traves de las laminas de metal desde el N al S con lo que el campo magnetico se reduce.

Saludos

Hello to all.

If, it is as well as the magnetism works, but what it makes the shield but that to armor is to lead the magnetic field through you laminate them of metal from the N to the S with which the magnetic field it reduces.

Greetings

Edited January 19, 2007 by zfrittz6

[JER00N]

Ah thanks, that makes sense to me

[zfrittz6]

[col]

There is a problem with this 'shielding' (which is not really shielding at all as has been said many times). The more mass of shield you have to bring the flux back around to the bottom pole of the magnet, the less flux gets to the strings. So basically, it will be a difficult balancing act. More shielding closer to the driver and strings means smaller EMI problem, but also means less efficient driver !

I have done some femm plots in the past that demonstrate this well.

I have also built a shield that fits my driver. its made of fairly flexible sheet, so I could experiment with different 'settings'

It is very important to leave enough of a gap between the shield and the driver ( at least a few mm) and also, experiment with how close the top and bottom edges should be.

Personally, I feel that a shield good enough to have a noticable impact on the interference will also have a considerable impact on battery life. You will need a more powerful amp drawing more mA to get the same string reaction.

I'm sure this can be balanced out somewhat by using class-D amps etc... I would suggest however, that before spending loads of time on shields, everyone should try both bi-longitudinal drivers like mine, and also bi-lateral drivers like sustainiac. These provide a much better interference cut without being detrimental to the battery life.

Theres a very good reason why sustainiac don't use a shield (other than a tiny plate that covers a small section of the driver.)

Also why the other driver that has a shield (kramer?) only has a small thin partial shield... any more would suck too much flux away from the strings.

cheers

Col

[JER00N]

Thanks col for the explanation. Can you please post the design of your latest working bi-longitudinal driver? I've been looking a bit through this thread but it's almost impossible to find it in such a huge topic

[col]

Thanks col for the explanation. Can you please post the design of your latest working bi-longitudinal driver? I've been looking a bit through this thread but it's almost impossible to find it in such a huge topic

Its just the same configuration as a standard 1 magnet humbucker...

each of the two coils is ~ 4 ohm, wired in series that gives the magical 8ohms.

for the core I used (iirc) iron(drawn steel?) bar 56mm x 10 mm x 2mm, i think the coils were 3mm thick.

the magnet is at the bottom with its North on one core and its South on the other...

wind both coils the same way (clockwise or anti-clockwise), then connect the end of the first coil to the end of the second. The two remaining wires are your driver wires to attach to the amp circuit.

Here is a pic of my driver in position:

you can see the two coils, and the cores protruding underneath, between the cores you see the hefty slab magnet with N along one edge and S along the other, butted up against the sides of the cores.

At the far end, you can just see the blue insulating tape protecting the connections. I managed to solder multicore wire onto the winding wire before the potting epoxy had set, so the connections are pretty stable.

(If you are going to use epoxy, make sure you get a slow setting one. Araldite do one called 'precision' that gives you an hour or two before its set. This gives you time to work, and time to clean it off your hands before it burns - just in case you get it on your skin. Best to buy a bag of disposable protective gloves.)

So basically, its just two identical 4ohm coils connected in series and out of phase with cores with opposite polarities provided by one magnet.

Any more questions, don't hesitate

Edit: I forgot so mention that i rounded off the ends of the cores with a file so the winding wire would not be stressed or damaged.

cheers,

Col

Edited January 22, 2007 by col

[JER00N]

cheers,

Col

Alright, thanks. Does this kind of driver also avoid feedback even when it's close to another pickup? Would it be possible to fit the coils next to each other instead of after each other (like the Sustainiac seems to be) so it can be mounted in the single coil mid position of my H-S-H guitar ?

[col]

cheers,

Col

Alright, thanks. Does this kind of driver also avoid feedback even when it's close to another pickup?

It doesn't avoid feedback, it reigns in the flux of the magnetic field which means you can move it closer to a pickup - if you go really close, you will still get problems. I think you will discover that although you might not get squealing feedback, you will still get fizz and grunge that is (at least for me) unacceptable.

Would it be possible to fit the coils next to each other instead of after each other (like the Sustainiac seems to be) so it can be mounted in the single coil mid position of my H-S-H guitar ?

yes, the main difference is that if you go for the 'bi-lateral' approach, you will need to use two magnets, one for each coil and have one of them 'upside down with respect to the other', so you won't be able to use a nice salvaged humbucker magnet like i did.

[Samuel McBrian-Brian]

I don't know if anyone has ever mentioned this, but I think someone should.

30 AWG copper wire is 0.338 ohms/metre (http://www.powerstream.com/Wire_Size.htm) and using maths we can work out that you need approximately 23.669 metres of wire for an 8 ohm coil. This just takes the guess work out of it, I mean, dozens of places I've read that you do 'a hundred or two' turns.

I vaguely recall talk of six coils, one per string. I do wonder where that discussion led, but you'd need approx. 142.012 metres all up (23.669 m per coil). Assuming they're coming off of the one amp and they're in parallel (to be able to turn them on and off), when only one coil is turned on it would have immense power pointed at one string (rather than the big six string drivers where the power is applied to six strings).

Power equals voltage multiplied by current (maybe). The current will be divided between all the coils which are on, thus the more coils you have turned on the less power each one puts out but the total power is the same.

I think three coils would be easier, especially if you use the pole pieces from pickups as the core (winding 24 metres of wire around those would make them big).

Anyway, I hope this hasn't been all discussed before otherwise I'll get yelled at.

Love From Samuel.

[col]

I don't know if anyone has ever mentioned this, but I think someone should.

30 AWG copper wire is 0.338 ohms/metre (http://www.powerstream.com/Wire_Size.htm) and using maths we can work out that you need approximately 23.669 metres of wire for an 8 ohm coil. This just takes the guess work out of it, I mean, dozens of places I've read that you do 'a hundred or two' turns.

This has been taking the guesswork out of it for quite a while !

Unfortunately though, its not quite so simple in practice. When hand winding coils, its quite easy to introduce some variation into the ohms/metre value, the fine wire can easily be stretched a little particularly as it rounds the ends of the core, any stretch - however slight - will increase the resistance of that section of wire. The other posibility is that part of the coil could be damadged and shorted out, removing it from the equation... So its best to use these techniques as a guide rather than just following them blindly.

I vaguely recall talk of six coils, one per string. I do wonder where that discussion led, but you'd need approx. 142.012 metres all up (23.669 m per coil). Assuming they're coming off of the one amp and they're in parallel (to be able to turn them on and off), when only one coil is turned on it would have immense power pointed at one string (rather than the big six string drivers where the power is applied to six strings).

Power equals voltage multiplied by current (maybe). The current will be divided between all the coils which are on, thus the more coils you have turned on the less power each one puts out but the total power is the same.

how do you propose to turn the individual coils on and off?

with such chunky coils, how are you going to prevent dead spots between the poles that would prevent sustain when bending strings ?

The only way to make this approach feasable as far as I can see is to have a hexaphonic pickup and six driver circuits

(you would need some dramatic miniaturization to fit my circuit into any guitar six times )

You would also probably need to stagger the position of the driver coils to help prevent dead spots - some sort of zig-zag pattern.

fwiw, this has been discussed before a few times - search the thread for 'hexaphonic' and you'll probably find some of the posts.

The fact that it is so complex combined with the fact that the simpler monophonic approach works really well means that no-one has yet attempted a full implementation of this concept - maybe you will be the first ?

cheers and good luck

Col

[onelastgoodbye]

@ Samuel McBrian-Brian

Hello and welcome...Col posted a link to an online coil calculator some time ago, can't find the link right now....I measure the length of wire needed before I start winding and a 4 ohm coil (half of a humbucking driver) comes to precisely 6 m with 0.2 mm wire 32 (awg). It seems the winding of the coil adds some resistance to the ohmage of the wire itself (because the overlapping influence eachother I guess). So don't use those calculations as an absolute measurement. The resistance isn't all THAT important anyway, I doubt you'll "notice much difference between say, 7.5 and 8 ohms. What really matters as far as driver efficiency is concerned, is how well the coil is potted, and the wire gauge. So far it's always been either 0.2 or 0.25 mm wire, but you'll find the 0.2 makes for a much smaller coil than the 0.25.

About the six coils, only psw has experimented with that and it was quite tricky to control EMI, let alone build those tiny things. You,d also need six separate amps to take full advantage of them, which means lots more power drain and emi to control. So the hex drivers are 'on hold' basically.

@ Radiotrib

your idea about the coils on their side is similar to what fernandes originally used in ther sustainer guitars. The patent is on ... page 100-ish. We call 'em side drivers here. I want to build one myself if i find the time. Basically, they should offer superior EMI resistance because of the way the magnets are positioned, but it has been stated that efficiency might be poor as the coils are on their side. Check out q-tuner for some cool pickups that use the same idea. Nederlands bedrijf trouwens .

@ Zfrittz

Could you post some pics of your new super-thin driver? I'm interested in how you built it. Any test reports?

[zfrittz6]

coil COL

Coil Zfrittz

Hola la bobina en si es casi igual que la que tiene Col solo cambia el nucleo que en lugar de utilizar una sola lamina de metal yo utilizo varias mas finas y un poco el montaje que el lo monta en el lateral del iman y yo en lo alto y el blindaje por lo demas es igual si necesitas alguna explicacion o algo no tienes mas que pedirlo. David si estas por ahi y no se traduce bien lo traduces tu mejor.

Hello the coil in if she is almost just as the one that has single Col changes the nucleus that instead of using a single one laminates of metal I I use several but fine and a little the assembly that mounts it in lateral of the magnet and I in the stop and the shield by others is equal if you need some explanation or something you do not have but that to request it.

Salu2

[Radiotrib]

Hoi Onelast .. Yes - I have seen Q-tuners - sexy pickups, but when I heard their bass models they just sounded way the too "crisp" for me ...

Anyway, that isn't really the idea I was putting forward. My suggestion was based something I vaguely remember, that a ferro-magnetic plate with a large area to thickness ratio, in close proximity to a magnet, confines the magnetic field (i.e. shields) by sort of "sucking" the flux in towards it, but one which is actually in contact with the magnet causes the flux to stay within the plate, and exit only(ish) via its edges. If this was actually the case, my idea would work something like this (cross section) ...

focusing and containing the field, but producing a strong flux between the edges of the plates, and very little stray field except that transmitted by induction in the guitar string. (which is also an iron core inside a coil ...)

I think there's a magnetic field/interaction modeling tool .... I need to look around and see if I still have a link to it ... AHAAA ... Ansoft was the company ... http://www.ansoft.com/ and the tool is called Maxwell SV .. it's a free download for non-commercial use. Maybe some of you electronics gurus might be able to make sense of it .. I can't

[Samuel McBrian-Brian]

I don't believe you'd need six sensors and six amps. If the six coils are in parallel they could easily be turned off by an on/off switch. I realise that they may have to be staggered, perhaps a humbucker kind of thing with three coils in each single coil.

The driver I have made is from an old single coil pickup, and bending was sustained just as well as the single notes. I think this is because the poles are fairly close together and its not like its a beam of magnetic energy shooting out from the pole.

Thats it, I'm going to make a triphonic coil and if thats a success I'll consider making a hexaphonic one.

Didn't you all find that you had strings being magnetically induced into vibrating that you didn't want to when you were playing on some other strings?

[Truth_David]

Hola la bobina en si es casi igual que la que tiene Col solo cambia el nucleo que en lugar de utilizar una sola lamina de metal yo utilizo varias mas finas y un poco el montaje que el lo monta en el lateral del iman y yo en lo alto y el blindaje por lo demas es igual si necesitas alguna explicacion o algo no tienes mas que pedirlo. David si estas por ahi y no se traduce bien lo traduces tu mejor.

Salu2

Hola Juán, si por supuesto amigo......

Here is a possibly more helpful translation :

Hi. The coil in itself is almost the same as Col's one, just that for the nucleus, rather than using one single strip of metal, I used several fine strips and some fixings that mount on the side of the magnet, and above and on the shielding. Aside from that, they are identical. If you need any further explanation, you only have to ask.

David