Truth_David Posted February 8, 2009 Report Share Posted February 8, 2009 Incidentally, VG-99 users who have tried the Moog Guitar say that it is not as awe inspiring as they would have hoped, because the actual quality of the sustainer seems to be tonally weak, and it is not immediately possible to find a good strong sustain that will pick up and get the strings vibrating straight away.<br /><br />The moog people certainly have put out a lot of hype and there has been some discussion of what the thing is actually capable of...however, other than some small and very selective and non-informative company demo clips...very little is known about it and I remain a little skeptical still. Much of my skeptisisim is based on my experiences of similar experimentation and the apparent limitations on what can be done. However, nothing put out by moog suggests they have overcome these things.<br /><br />Also...I have not heard anything from owners or uses of the moog guitar...i'd be interested in seeing some links perhaps...but even then, the lack of any technical information and a lot of mis-information is remarkable...even if it has caught the imagination of not just a few people.<br /><br />Still...it is not all that much more expensive than the roland systems...all things considered.<br /><br />Sample and hold things are a different effect, but just as valid musically as feedback, which tends to be more organic. However, in the bigger scheme of things, this technology is only a tiny part of the creative arsenal and shouldn't be seen as a magic bullet to creation and invention in music or guitar technique. Guitar synths have been around in various forms for a long time, yet they have never really been taken up or used to any real extent. Even things like sample and hold delay effects and other things have been available without much creative or commercial use. The eBow has been around much longer, yet very little prominent use has been made of this remarkable device...so that says something I think for the expressive nature of the guitar without these devices!<br /><br />pete<br /> Pete, I just Googled for the Moog Guitar and found this blog: http://blog.wired.com/music/2008/06/moog-unveils-ba.html According to this, one would be talking in terms of around $6500 US, so it is something like three times the price of the VG-99 with the FC-300, the stand and the GK-3 p/up, so in fact not only is it priced way beyond the range of most players, but in comparison the VG-99 system offers far greater versatility, and with the sample and hold and sampling features in addition to everything else it has in its arsenal, why not plumb for whatever guitar takes your fancy around the $4000 US mark plus the VG system, then just do what I want to do, which is simply to add the sustainer technology in order to get the icing on the cake? Incidentally second hand VG-99s are now starting to find their way onto e-Bay at somewhat reduced prices, but even brand new there are great deals to be had, way below the m.r.p. I certainly do not regard it as being expensive, given that effectively it is like having two Variaxes with two separate effects processors and synth modeling and D-Beam and Ribbon Controller technology. I really think that Moog went out on a limb and bit off more than they could chew....this is likely to be as big a fiasco financially as the Synthaxe, and for the very same reasons - it was also overpriced and overspecialized, so that in the end only the very very privileged few ever got to own one or even play one, and now there are so few remaining that they cannot be obtained for love nor money, and even if they could be found, there are simply no competent technicians to service them and replace parts where necessary. The Moog Guitar may not be anything like as complex, but my feeling is that they are not going to find many buyers at that price. On the other hand, if they were to listen to what musicians have been clamouring for with the Variax, which is to make the technology possible to retrofit to any instrument, they would gain a far greater chunk of the potential market, as would Line 6 if they decided to the same with the Variax modeling technology. You and I and just about every visitor to this forum are proof of that, because that is precisely what we are aiming for. I have no idea how much success the commercial vendors of sustainer technology who have chosen to go that route have obtained, as opposed to selling instruments pre-fitted with the sustainer, such as Fernadez, but I daresay the kits will have found a pretty good niche in the market..... David L Link to comment Share on other sites More sharing options...
psw Posted February 8, 2009 Author Report Share Posted February 8, 2009 Well...hmmmm.... This project is pretty fickle... firstly...it is the neck pickup that needs to be the driver...it works only with the bridge pickup and on a strat you need to be very careful about bypass switching and such. As always...ALWAYS...test your device and circuit before modifying or installing the thing into a guitar...do not assume it will work... You will need a bridge pickup...it will only work with this design...so be aware of this...perhaps a little more reading required...or more questions to be asked...don't rush into it and get yourself frustrated, or expect it to work with a "this will do" attitude, especially with regards to specs...which brings us to... As for wire...it is not just a matter of working best but working at all in many cases!!!! .312mm wire WILL NOT work with my designs. This is one of the crucial parameters, much has been said about this, many have shown that my formula works and deviations don't in this regard. This is not simply an electromagnet or that a basic understanding of the principle of the device will suffice. As for the ruby...well if it has the preamp stage to the LM386 (a transistor) it might work. With a cheap strat, it will often like a bit more preamp power or rely more ofn the quality of your driver to make it work. Also...there are a few "ruby" like designs...almost all of them lack the additional filtering that I suggest...it might get lucky...the circuit really is an amplifier with buffering, but this device places a fair amount of stress on a circuit and as we don't "hear" the thing when it is working, it is a little hard to tell what is going on. MRJ preented an F/R design modified with some of my suggestions (though not all I see now) and there are other circuits as well...but give the ruby a go and perhaps consider some mods like extra caps. A cheap strat is a great guitar to use...but strats can be fickle. Also...when installed you will need at least a 4pdt switch to turn it on and a complete rewire and you can't hide the battery under the pick guard...this device eats batteries! However...often you can get the battery into the trem cavity with some jigging of the springs...hahaha ok...probably answers some of these questions for now... A pickup and pickup magnets tend to be pretty ideal...neomags not the best. I know it is an internet age, but some of these things need to be sourced from unusual places and sometimes you need to get out and hit the streets...not the net. Magnets are a case in point...often you can find very nice cheap ceramic mags at craft shops...I think that is where my tele ones came from...you may even have some sutable things stuck to your fridge. Similarly core material like scrap steel...or a hardware store. For a while I was collecting street sweeper brushes...these get thrown off and are 3mm spring steel...very magnetic...and lying about everywhere... pete Link to comment Share on other sites More sharing options...
Hank McSpank Posted February 8, 2009 Report Share Posted February 8, 2009 (edited) For anyone reading in the UK....Maplin has a good variety of sizes of enamelled wire, which was surprisingly hard to find on their site (ignore the photo...that's their large diameter stuff!)... http://www.maplin.co.uk/module.aspx?moduleno=44 (you'll see that they have 0.19mm diameter - product code YN90X - which will suit those going for Pete's well R&D'ed design) (it's not necessarily the cheapest source, but nevertheless they have some of the thinner stuff that I couldn't even locate on Ebay, Rapid etc). Handy if you'd be placing an order over £35 as it's free delivery then (don't forget Quidco too!). For magnets, I ordered some small ceramics from Ebay (on the basis they're still quite strong but can be chopped up to suit). There are plenty of ceramic magnets to be had on Ebay, for example.... http://www.tinyurl.com/d6lwvd (if you have kids, they can have some real fun with half of them! Not recommended for reall small kids under 5 though) One thing I've not been able to source, is pure iron rod (I'm pursuing the dark side - hex drivers, so I'm not looking for a long thin driver core like most)...so it'll have to be mild steel for the time being. Edited February 8, 2009 by Hank McSpank Link to comment Share on other sites More sharing options...
psw Posted February 8, 2009 Author Report Share Posted February 8, 2009 Nice finds there... Would you believe for a while there I used to be able to walk into a shop in nearly every shopping mall or locally in a variety of locations and get 0.2mm wire in affordable reels (under $10 and enough to make 3-4 drivers)! Unfortunately, about 2 years ago the supply started to run out, now these guys seem to be getting out of the parts all together. We still do have some places, but they are getting rarer and stock is more limited all the time. My own circuit uses some SMD's and for layout reasons need to be a specific type for some parts...no doubt there will come a time when even this is going to become difficult...such is the way of the world! So...I wonder would I have been able to develop this design now! I was using so much wire though that at one point I bought a commercial reel (about $100) and have being using this ever since. The design was found by trial and error and I still have small supplies of thinner and thicker wire about...originally I used thinner 0.15 wire i think with another design and the single string coils at about page 2 but I do know, and many others have demonstrated that 0.2mm is the magic number for this particular concept. ... Magnets are addictive. So much of our world and our technology is based on these things and forces. The whole world is one big magnet of course...but very little is truly understood. These forces, and the forces of gravity are some of sciences biggest mysteries...yet they seem so self evident...yet invisible...such enigmas. Do be careful with magnets though...there can be many dangers that need not be attempted. In particular, although I have said it before, I always feel wary because people may be encouraged to do things when they are told it is dangerous, especially when it is something that seems so inert as a magnet. All magnets, but particularly neodymium magnets can be DANGEROUS...especially if they are cut, put under pressure (say in a vice) or even filed. Consider the forces inside a magnet...some are incredibly strong. Now, what happens if one breaks...well you get two magnets and two north and south poles. Now, some magnets like neo's are scintillated materials, ceramics are just that...and the result is that they can break unexpectedly if forced. Now, if the bonds that hold these things together are broken you may get hundreds of tiny sharp shards of magnet. Now, consider what happens when you get unlike magnetic poles...they move apart as quick as possible...with a powerful neodyminium for instance...it will explode razorsharp pieces in all directs...includig towards your eys and other soft scar-able places...and those of nearby children others. At the very least....if you survive...they will be no good to use for anything. Lest you think it a good idea to try to do this, it is not a spectacular thing worth seeing like dynamite...but it could be the last thing you see. Generally, ceramics can't be cut, neos are not really suitable. Alnico can but tends to demagnetize with cutting (in fact all magnets suffer damage through cutting and filing). However, it is possible to use multiple magnets and if seriously pursuing something like hex designs, neos may fit the bill if you get clever with them. A neo magnet of about 3x3mm round can have the strength of a single strat pole piece...but the magnetic shape and throw as discussed earlier is completely different due to the high power and compact size... All of this can be explored to some extent in 2D with the FEMM program which is free and makes the strength and lines of magnetic force visible. If nothing else, it allows you to understand the interaction between magnets. This may be important to hex designers, as it may not yet be fully appreciated how magnets and ferrous materials couple and interact. And, it is not really hard...and a little addictive!!! But magnets are cool...be careful around young children, but when my son was around I used to tape a small neo to his hot wheels cars and he could use another magnet to push or pull (depending on the magnet direction) and even turn them on a table top...like a battery-less, wireless remote control! However...close supervision required. Be aware that magnets can also destroy computer discs and tapes...and interfere with other electrical things occasionally. Very strong magnets can distort tv screen pictures for a laugh...but then they can also take off a finger pretty easily if you are not careful. As for cores...there has been various discussions...i even got crazy with some exotic pure iron powder and epoxies to make my own things and magnetic shielding things...but all this gets pretty expensive and mucky. I generally use ordinary steel flat iron, 3mm thick and cut it down to rough size with a hacksaw, then grind it to shape (make sure all sharp corners are taken off and rounded where necessary). You need so little of it, that I am still using the same piece as the original from the hardware store. However, magnetic materials are everywhere...computer cases are often highly magnetic and easy to cut...but steel brackets and such can often yield a useful piece. As spoken about earlier...you can get fancy, but don't get carried away, this will make little difference and with DIY tolerances, you have more important things to worry about...hehehe There are still a lot of things here that are "theories". These are based on observations from real experimentation, but of course are open to be disproved. Time and again they are proven by failures to hold up. Such theories include the "thin core" ideas and of course the "thin coil" both of which I continue to feature in my designs...there are plenty of others and discussion for masochists who wish to look further into these as they came up in the thread or wish to debate them with me...lol. My basic formula seems to adequately fulfill the job...but there are many other areas worth exploring. However, there are limits and wishful thinking (whcih I am as prone to as any, if not more so) is not enough. For instance, there are a lot of merits to the dual coil designs. I tried some, but it is notable that I have tended to abandon them...but others like col, curtis and a few others have had some success...and of course the sustainiac bi-lateral designs have been emulated. However, where these things are used to push the use further, they have largely failed the hopes of the designers. For instance, the idea of a mid pickup driver for a strat (a driver in the middle pickup slot capable of running with the neck or bridge or both pickups and with substantially easier installation (no bypassing required)) has been a dream that has yet to be fulfilled...you can see at least one design of mine and I did try a bi-lateral thing as well without success. ... Anyway...a long and unintended post from me....late night... pete oh...while I have not suffered from shrapnel from magnets as such, there are dangers. I can say with general certainty that I have suffered as a result of some glues. It is subtle and cumulative...but people should be aware of the dangers and take suitable precautions...at least not be too blase about them. Epoxies can be bad but the most common culprit is super or CA glues. Also, be aware that even if you think that you have escaped damage of lungs and brain, and only the occasional fingernail glued to your flesh...there is always cancer to come and bite you later. This is one reason that I have backed off on doing more work in this line. Still, guitar building can be a dangerous business...saw dust can be toxic (with exotics and MDF) and a router can chew you up in no time flat of course!~ None of these risks are really a part of any of the project that I have put out and the use of PVA potting has generally been found to be superior in practice anyway and completely safe! Link to comment Share on other sites More sharing options...
col Posted February 8, 2009 Report Share Posted February 8, 2009 One thing I've not been able to source, is pure iron rod (I'm pursuing the dark side - hex drivers, so I'm not looking for a long thin driver core like most)...so it'll have to be mild steel for the time being. As you're in the UK, you should be near a B&Q store (assuming they don't succumb to the economic situation). Some B&Q stores sell a selection of bars of steel (has a variety of names - flat iron, cold steel, drawn steel - all seems like the same stuff though). I've used this stuff successfully as driver core material. It comes in various dimensions: 6mmx2mmx500mm; 4mmx4mmx500mm; 10mmx3mmx500mm etc. It is easy to cut and shape using a hacksaw and file, and its soft magnetic properties seem to be at least adequate for this project. Another approach that has been somewhat successful is to rip apart old transformers and cut up the laminated core - this stuff should be pretty good for our purposes as long as you can get it and have the tools to cut it up. Sculptors use Iron, useful to remember if you're set on using iron. Although if you are going to mail order stuff, it would be worth checking out some of the more esoteric magnetic alloys that are available. These have far better magnetic properties than plain old mild steel or iron Iron is not actually that great for this project, it is difficult to obtain, difficult to work, and it has problems associated with very low resitance (high eddy current levels), Hysteresis characteristics are poor... Its good points are also not anything like as good as those of some of the steel alloys e.g. magnetic permiability of Iron (99.8%pure) is 5000μr, Silicon GO steel(used in mains transformers) has permiability of 40000μr. check out this page for more details on materials (yes its from that same site I posted a link to earlier - I told you all the info was there;)). cheers Col Link to comment Share on other sites More sharing options...
Akula Posted February 8, 2009 Report Share Posted February 8, 2009 Well...hmmmm.... This project is pretty fickle... firstly...it is the neck pickup that needs to be the driver...it works only with the bridge pickup and on a strat you need to be very careful about bypass switching and such. As always...ALWAYS...test your device and circuit before modifying or installing the thing into a guitar...do not assume it will work... You will need a bridge pickup...it will only work with this design...so be aware of this...perhaps a little more reading required...or more questions to be asked...don't rush into it and get yourself frustrated, or expect it to work with a "this will do" attitude, especially with regards to specs...which brings us to... Yeah I intend to replace the missing bridge pickup with a humbucker. So my config would be H-S-S, with a driver next to the neck pickup. And yes, only the bridge pickup would be used with the sustainer, to avoid all that nasty squealing. As for wire...it is not just a matter of working best but working at all in many cases!!!! .312mm wire WILL NOT work with my designs. This is one of the crucial parameters, much has been said about this, many have shown that my formula works and deviations don't in this regard. This is not simply an electromagnet or that a basic understanding of the principle of the device will suffice. Okay, fair enough. I'll get some more wire. As for the ruby...well if it has the preamp stage to the LM386 (a transistor) it might work. With a cheap strat, it will often like a bit more preamp power or rely more ofn the quality of your driver to make it work. Also...there are a few "ruby" like designs...almost all of them lack the additional filtering that I suggest...it might get lucky...the circuit really is an amplifier with buffering, but this device places a fair amount of stress on a circuit and as we don't "hear" the thing when it is working, it is a little hard to tell what is going on. MRJ preented an F/R design modified with some of my suggestions (though not all I see now) and there are other circuits as well...but give the ruby a go and perhaps consider some mods like extra caps. Yep, my circuit has a transistor. This "extra filtering" you say, whats do you mean by this, just more capacitors to take out certain freqencies? Oh and on the subject of the circuit, I'm a little confused as to how the harmonics effects work. Does this require extra circuitry, or will a plain old Ruby circuit get me those really high harmonic notes? A cheap strat is a great guitar to use...but strats can be fickle. Also...when installed you will need at least a 4pdt switch to turn it on and a complete rewire and you can't hide the battery under the pick guard...this device eats batteries! However...often you can get the battery into the trem cavity with some jigging of the springs...hahaha Haha, yeah, I must have taken off the pickguard to this guitar about five times in the last week, quite frustrating. The trem cavity will probably hold the battery, if not I'm willing to rout out some more wood. Thats the great thing about modding a cheap guitar - if it fails, its not like losing a really expensive guitar. I'm more of a bass player, and this guitar was a bit of an "Oh why not" kind of purchase. It cost me about £30, so it's both a play-around guitar and a testbed for future guitar mods/making. I'm planning my first guitar build, a 5-string bass, but this is going to be far into the future, and off-topic Cheers for all the help - Akula Link to comment Share on other sites More sharing options...
col Posted February 8, 2009 Report Share Posted February 8, 2009 oh...while I have not suffered from shrapnel from magnets as such, there are dangers. People with small children should be very careful with rare earth magnets. Swallowing one is not an issue, but if more than one is swallowed, they will clamp parts of the digestive system together causing serious injury and even death. I can say with general certainty that I have suffered as a result of some glues. It is subtle and cumulative...but people should be aware of the dangers and take suitable precautions...at least not be too blase about them. Epoxies can be bad but the most common culprit is super or CA glues. This seems highly unlikely, where's your supporting information on this? All the searching I've done and all the information I've found from many sources agrees that cyanoacrylate glues are safe (apart from the danger of bonding things together of course). A few people can develop a mild alergic reaction over time to some variants, but other than that they are SAFE. Epoxy based glues on the other hand are thought to have various negative health effects, some serious, if suitable precautions are not taken. ...the use of PVA potting has generally been found to be superior in practice anyway and completely safe! PVA potting generally superior? who has tried both PVA and Epoxy enough times to make a fair comparison? My drivers have all been potted with epoxy, and apart from the very first one (which failed due to not making the bobbin wide enough), they are all mechanically excellent, no microphonics etc. (have you tried making a bobbinless coil using PVA?) My past experiences with PVA suggest that it is not ideal for this project, mostly because of the drying process. It will take a long time for any PVA in the centre of the coil bundles to cure. Epoxy is ideal in this respect. It may be that after it has been left to cure long enough, a PVA potted driver can be as good as an epoxy potted one, but I don't see how it's possible for it to be 'superior'. cheers Colin Link to comment Share on other sites More sharing options...
Hank McSpank Posted February 8, 2009 Report Share Posted February 8, 2009 And what about potting with paraffin wax? (apart from chance of air pockets) Link to comment Share on other sites More sharing options...
psw Posted February 8, 2009 Author Report Share Posted February 8, 2009 Fair enough...I concede...PVA is not the best potting...but for the DIY level and the manner most approach this project, I think epoxies should only be used combined with some experience. I use epoxies now, but to find something suitable resulted in me needing to try many products at a cost of about $100 to find something suitable for those bobbinless designs...none of these were easily obtainable form hardware stores. Although PVA will take a long time to cure, even remain soft, these properties still provide a lot of damping compared to some of the alternatives people have tried. The important thing to remember is to prepare the materials before hand and bind the coil with tape and clamp at least overnight. The fumes from CA are a vaporized form of the cyanoacrylate monomer that irritate sensitive membranes in the eyes, nose and throat. They immediately are polymerized by the moisture in the membranes and become inert. I know I have experienced this, and of course inhaling it can coat the insides of your lungs. Spilled on cotton and wool it can cause burns or even ignite (I have had this reaction) and potentially catch fire...I have had this with epoxies in a different context in boatbuilding. With general use, it is reasonable safe, however, with potting I regard CA glue to be potentially very hazardous and not filling enough to do the job. Potting post winding generally has been shown to be difficult, but it might work. Again, wax can be dangerous and catch fire if not handled properly. You can't really be sure of saturation as you can with PVA. So, my support for this safe and cheap material, all things considered, is still that it is superior for the level of experience and skill for this project and will work, while many have failed attempting it with conventional epoxies and CA attempts. It also has the benefit of allowing very easy removal for rewinding and modification that these other methods don't provide...it are these kinds of aspects that continues my support for PVA. Similarly core materials and such...yes, there are potentially better materials...but the expense and difficulty of working them can be tough and often it can lead to the idea that these things are required. Ordinary steel is adequate. Getting to carried away can really delay practical results. In the end, if it works, it works... ... The harmonic function is created (generally) by a simple phase switch on the driver leads. It is probably a good idea to read the tutorials and stuff a lot more about these aspects...it is important to understand how this thing works in ways unlike building a stopmpbox. ROG reduced many of the suggested specs for running the LM386 at high gains...the caps to keep the chip happy are there to help prevent internal oscillation in the circuit under these loads. ... I'd like to mention that today my part of Australia has and is suffering from a major disaster in the form of fire storms. Some of it has been in areas that I have traveled recently and not that far from me (although I am well clear), 108 are confirmed dead (so far), 3,000 without homes but it is still very early in the day and the fires are still burning. It is looking to be perhaps one of our worst natural disasters in a land that tends to do such things on a big scale. Anyway...spare a thought as they are predicting worse to come. Thanks to those who who have been offering support (100 firefighters from NZ are being flown in...and there seems to be support offered from the UK). I will be seeing if there is anything I can do to support those who have been affected today. Link to comment Share on other sites More sharing options...
Donovan Posted February 10, 2009 Report Share Posted February 10, 2009 Hello all- Sorry to hear about the devastation out there, Pete. My sympathies are with you and your people. Let's see what kind of replies this generates. This is regarding the signal sent to the driver. My apologies if this was covered somewhere else in this encylopedia of a thread. Assuming a clean (meaning sinusoidal waveform with no clipping) is being sent to the driver, what is the physics of what is happening to the guitar string? I got to thinking that if the driver were exactly like a speaker (which it is obviously not), then they might exhibit similar behavior... with no signal the cone is at rest, centered between to extremes. Then, with a pure AC signal, there is a pull on the cone during one half (which side is trivial), let's say the positive half of the cycle. Then during the opposite, negative side of thecycle, there is a push, so the cone can move in both positive and negative directions. This bidirectional movement is possible because the cone is centered when at rest. However, our drivers are not centered. I would venture to say they are nowhere near centered for a couple of reasons which I of course am completely, intuitively assuming... 1) The strings are under constant pull force of permanent magnets. 2) The strings, although loosely analagous to the speaker cone... as they are not magnetic, can not ever be pushed, only either pulled or released by the momentary field. 3) The polarity of the permanent magnets employed is never reversed only strengthened and weakened by the driver's induced field. What I mean is that the driver only has the ability to exert different pull values, never a push value. First theory of the day... if number 1 and 2 is correct, then the string need never be pushed. I have a vague feeling though, that one might envision the at rest condition as a pull of say strength 3 (on some made-up scale), then on the half wave peak, pull is up to 5 (strength of +2). Now the wave crosses origin and we're back to strength 3. Once we hit max negative cycle, we can go 2 less than the positive 3 value, bringing it to only a +1, which is essentially less than the at rest state, so both sides of the wave ARE useful. Next theory... if 3 is correct, then there is no sense in trying to push, only pull, and as such half the waveform is useless. Next lame theory... if both 1 and 2 are correct, but lame theory 1 is incorrect, then The only other idea I have... and this one I am not entirely convinced of at this point as it feels far fetched, is that the string is being pulled on both sides of the wave, in effect being pulled at double the audible frequency. To clarify, imagine that the negative sides of the wave were being turned upward, so there were only positive values. What would be the behavior? How would we know if this "driver" was or was not behaving this way? The reason I think it might be is because I was reading about AC electromgnets today and if I understood the material properly, an AC electromagnet with no magnetic core is possible given constanct frequency, meaning at say 60 Hz, the N and S poles change at 60 cycles per second, yet there is a perceived constant pull. There must be more or less a constant pull, as an equal pull then push scenario would lead to net zero pull, no? I would like to hear what you all think is happening on this level. I am making good progress with my units, with more bugs to work out, but at this point, I would really like to grasp this a little better to decide where I am going next. Keep rockin. Link to comment Share on other sites More sharing options...
psw Posted February 10, 2009 Author Report Share Posted February 10, 2009 PSW attempts a short reply... #1 True #2 False #3 maybe...false Therefore... Theory #1 If I am understanding this...Yes Theory #2 Wrong Possibly you are reading from idea theory...like the theory that an opamp has unlimited (ie infinity gain)...very important stuff but not applicable to the real world. I can kind of see where you are heading, but I think you are stretching things a little. One thing to consider is that, the strings under some magnetic pull and at rest (remember, there are pickups exerting pull on them also)...this is their rest state. Therefore, any positive or negative pull on the string by the strengthening or diminishing of the magnetic power is a +/- exertion of the string in either direction. Also...this is an electromagnet that has to work very fast. It needs to change states ideally in constant sync with the vibration of the string. This is not possible as it takes time to charge to a state, discharge and recharge in an opposite state, resulting in phase differences. If, as in a speaker the string was the core running through the coil...you would have something more analogous to a speaker. However, (although I did try to do things like that) we are trying to influence a vibrating string by proxy with the electromagnetic device. As a result...the string needs to be coupled with the core of the EM which is achieved via magnetisim. As a result...the magnetic 'influence' on the string is vital to 'connect' the core and the string. ahhh...i should have stuck to point form...is any of this making any sense...it's only 6am! pete Link to comment Share on other sites More sharing options...
Donovan Posted February 10, 2009 Report Share Posted February 10, 2009 It was a brainstorm for power savings... that was my angle. One thing to consider is that, the strings under some magnetic pull and at rest (remember, there are pickups exerting pull on them also)...this is their rest state. Therefore, any positive or negative pull on the string by the strengthening or diminishing of the magnetic power is a +/- exertion of the string in either direction. This is exactly what I meant when I wrote... First theory of the day... if number 1 and 2 is correct, then the string need never be pushed. I have a vague feeling though, that one might envision the at rest condition as a pull of say strength 3 (on some made-up scale), then on the half wave peak, pull is up to 5 (strength of +2). Now the wave crosses origin and we're back to strength 3. Once we hit max negative cycle, we can go 2 less than the positive 3 value, bringing it to only a +1, which is essentially less than the at rest state, so both sides of the wave ARE useful. Kind of like the speaker with an offset in the pull direction, so /push/-/neutral/-/pull/ becomes /pull least/-/pull more/-/pull most/. Basically, pull and different levels of release, but never an actual push to where the string is under zero magnetic influence, whatever that might be. A shame is can't be more like the speaker and have true neutrality. Also...this is an electromagnet that has to work very fast. It needs to change states ideally in constant sync with the vibration of the string. This is not possible as it takes time to charge to a state, discharge and recharge in an opposite state, resulting in phase differences. I've heard this many times before and maybe I am misunderstanding, but I often get the notion that you underestimate the driver's capabilities to switch polarity quickly. Of course a halfway decent coil can handle switching states at audible frequencies. If it could not, we would not have speakers capable of high pitches. Even a $2.00 set of headphones can switch within the audible range, so I think it's reasonable to count switching frequency out as a weakness unless the craftsmanship is poor.These cheapo headphones don't seem to have a problem, so what's the difference between them and the driver then? The driver switching speed is not the problem, supplying the signal quickly enough is, but then again not even that is! Supplying a properly phase-corrected signal is. If the signal is properly adjusted, it shouldn't matter if you're 5 or 10 wavelengths behind. I think the biggest problem with this project at least in my case is phase relationship. This can be corrected and I would go as far as to say "corrected" may not even be entirely desirable... given recent experiments, with a little tweaking, I will in one month's time have this "correction" in place and user-adjustable not as a set and forget, but as an on-the-fly, user adjustable phase/fundamental/harmonic control. I have discovered how to tweak my circuit at this point to switch out components which raise the fundamental, the octave harmonic, the 5th harmonic, the and the 3rd harmonic. For the rock player, these Xth non-octave harmonics are much more powerful than the octave harmonic, which suits a more subdued emotion. This past few weeks I have been disappointed, thinking I was getting nowhere, getting killer harmonic sustain (of the Zakk Wylde kind) on some frets and not others, but now I am seeing why and this as an opportunity for improvement and yes I dare say innovation. Think 5-6 distinct modes via selector instead of 2 or even infinite incremental adjustment via knob. I'll be posting clips within the next few weeks. Link to comment Share on other sites More sharing options...
psw Posted February 11, 2009 Author Report Share Posted February 11, 2009 I will in one month's time have this "correction" in place and user-adjustable not as a set and forget, but as an on-the-fly, user adjustable phase/fundamental/harmonic control. Brave words indeed...i made similar predictions myself (albeit privately) to no avail...remember also, all commercial systems feature phase compensation circuits...my own designs have focused on superior speed to avoid such things. Phase compensation is not a simple thing, phase alters with frequency...so you will need to have a auto adjusting phase circuit that is triggered by frequency. Further... Of course a halfway decent coil can handle switching states at audible frequencies. If it could not, we would not have speakers capable of high pitches. Not true...so I think you are misunderstanding. The coil will respond quickly, but not necessarily in sync/phase with the physical vibration of the string. While a speaker coil will be able to recreate the sounds, it can not do it exactly simultaneously with the physical source. So, while a driver coil may be capable of transmitting the frequencies, there will be some lag...excessive lag will cause great inefficiencies. There have been some discussions about most of this stuff over the years...and I had thought some of these things before...but I was naive. The concept of a push only driver (or pull) driven by a half wave amplifier was discussed, in fact I did experiments with this concept. I was working towards something I called an string driving "engine". Using a Hex driver, I worked out a way to set off each driver separately and in sequence...much like the cylinders in a combustion engine), keeping up momentum of the string, driving it one sixth of the time, requiring only one amplifier, and avoiding cross influences between the individual drivers...oh so many great schemes... Nevertheless...half wave driving does not seem to be a practical theory, nor prove worthwhile in my experiments...perhaps you will have more luck. These drivers are pushing and pulling from the neutral state which is that of being caught in a mild magnetic field. I feel you are misunderstanding this influence. It is kind of like saying, well if we didn't have gravity, we all could fly...also true. The main reason pickups have magnets, and so to these drivers...is so that the strings are coupled with the core within and by the magnetic field. It is kind of like magnetic wiring. Speakers too have a core...it is the substantial magnet in which it vibrates. I did attempt a system wherein the bridge saddles incorporated tiny coils in which the strings passed. Just ahead of the bridge saddles were two magnets (N/S) to either side of the string. The idea was that the string is now the core of the EM and so, would be alternately magnetized, thus being alternately attracted to the opposite pole permanent magnets... These and more are from my undisclosed experiments and secret trials...however, this was many years ago, and if they worked to any practical degree, I would have published all kinds of things about it already...but I was naive. SO...as my posts are generally far to long...I think you may need to look into and consider these ideas a lot more. Phase changes with frequency. Yes, it wouldn't matter if it were 5 wavelengths behind...clearly it can be ahead and true resonance is fantasy, close enough may be possible. The problem is, that it is frequency dependent and building a set or adjustable delay is not sufficient as this will need to be adjusted automatically by different amounts as the frequency changes. Handling polyphonic signals like chords with such a system...well that's a whole other debate... However...I was encouraged to look at it from another angle...instead of aiming for compensation circuitry, why not aim for a driver of an efficiency that would be close enough to being in sync at a wide frequency range. Hence all my designs that have worked towards this aim. The thin coil designs are not like pickups, but have a very low permanency of the core, and a high concentration of forces directly under the string. The result is that it works fast. The same spec coil only deeper will not produce these results. Speculation has been that the "thin coil" idea that I hit on works largely at resistance, not altering impedance too much with frequency. I'm doing it again...too long and loosing people I suspect. Another thing you should consider in a one way electro magnet is that the constant magnetizing in one direction tends to produce a permanent, or semi permanent increased magnetism, or demagnetizing of the magnet and core. This is the opposite of a fast driver. You don't have the alternating reverse forces that act like a kind of magnetic spring to change states. I know...it sounds kind of good or at least intuitive in theory...but there are more factors at work still than you have dealt with, either like me through ignorance or wishful thinking...both of which have caught me up in the past. Please be careful with such boasts...we can all have egg on our faces...phase compensation circuitry is not new...not needing it is what I brought to the party. The Floyd Rose sustainer patent is a good introduction to the phase thing and strategies for it...though old and crude. But, this was required to correct the drivers foibles....better to aim for something that did not need this correction if possible. Remember also that the pickup has phase delays as well, and it is creating the signal...then there are massive differences due to the differences in location between the pickup sensing the strings vibration at one point, and the signal from the driver at another...and the physical vibration of the string at the point of being driven...not to mention other delays. All these delays are dependent upon frequency and those frequencies are detirmined by location and the characteristics of the driver...all of which we have generally failed to adequately describe. Col points out recently that some things are not only difficult to measure, but may have no practical purpose. So...adjustability good, but it will need to be over a wide number of triggered frequency bands...or create a faster driver that will be good enough...well, that's my feeling anyway. pete PS...you seem to have ignored that I said that #2 was false and therefore, theories following that rely on it are also false. I think you misunderstand the nature of a speaker, the core is the huge potted magnet that surrounds the coil...and that we are surrounded by magnetic fields all the time, what ever the field of a permanent magnet...as long as it does not move, it is the zero point. It is possible to create a neutral zone (not that it matters too much) by using equal and opposite forces as I did with a lot of the hex designs...but it does not change the principle. Think more about or experiment with pickups and consider why must a pickup have a magnetic field...then reflect back to this application...some answers are there... Link to comment Share on other sites More sharing options...
psw Posted February 11, 2009 Author Report Share Posted February 11, 2009 hehehe...i hear tell that they are getting closer to discovering the elusive Magnetic Monopole...perhaps we could inject some quantum mechanics into the sustainer... Sorry if I have been sounding dismissive again, as I say, I did think and experiment with these things...but what does that mean...perhaps you will succeed where I failed...or you may learn more in the process...or both. What I learned, which probably isn't anywhere near enough...was that there were limits to what I could tinker with...and...that there were always so many more things to consider. The Rose patent used a basic system of various caps to create delays to account for phase differences...lead and lag times I think they were described as...it used a frequency detecting device that switched in various filters to cause such delays depending on frequency. However, there seems to be a lot more questions raised in doing this...like how it can handle multiple frequencies from chords, or account for the differences in physical locations and driver and pickup characteristics...hmmm But...upon reading the last post more closely...I see that you seem to be describing a delay or filter to adjust these phase problems manually...there may have been some discussion of Q filters and such to make harmonic effects about the thread...the sustainiac had a kind of mix control, at least on the one I briefly had a go on last year. Whatever, a manual control may well be a good starting point towards an automated system down the line if worthwhile...so go for it. Before this thread...in fact way before and I think described in my first post...I did some experiments with sustainers, crude coils and hi powered amplification (the guitar almost caught on fire!)...amoungst these experiments, I did get sustain roughly (intonation problems, destructive heat and impracticality aside) but was able to rig up various effects in the drive chain. Part of this motivation was in an attempt to study phase effects. An interesting one was the use of a flanger to sweep through delay and phased effects...the result was some interesting harmonic effects that sounded like chirping birds or high frequency wale noises as it swept through harmonic nodes in the strings vibration. Here again is another area that had caught my imagination...but didn't pursue greatly...perhaps it is not really worth while...but it is yet another area which is yet to be fully explored...one of many... pete Link to comment Share on other sites More sharing options...
Donovan Posted February 11, 2009 Report Share Posted February 11, 2009 Brave words indeed...i made similar predictions myself (albeit privately) to no avail... I can already reproduce this mechanism, so it is just a matter of packaging it for ease of access. remember also, all commercial systems feature phase compensation circuits...my own designs have focused on superior speed to avoid such things. You have mentioned inability to measure/quantify anything, so how do you know this? Phase compensation is not a simple thing, phase alters with frequency...so you will need to have a auto adjusting phase circuit that is triggered by frequency. It actually is fairly simple, regardless of the fact that offset varies with frequency. Implementation is no more complex than using a jfet, opamp, or an LM386. Point me to the points in this thread where others or yourself have tried and failed if this has happened, else I think the assumption can be made that the details of the theory were overlooked or others simply were not interested enough in the idea to pursue it. The coil will respond quickly, but not necessarily in sync/phase with the physical vibration of the string. While a speaker coil will be able to recreate the sounds, it can not do it exactly simultaneously with the physical source. So, while a driver coil may be capable of transmitting the frequencies, there will be some lag...excessive lag will cause great inefficiencies. Again, this is not going to be that difficult to overcome. Humor me for just a moment... if what I am saying is true, wouldn't this be the way to tailor the system to each guitar? [\quote] There have been some discussions about most of this stuff over the years...and I had thought some of these things before...but I was naive. So again, please point me to the place in this thread where others have tried these things and failed and figured out why. The concept of a push only driver (or pull) driven by a half wave amplifier was discussed, in fact I did experiments with this concept. I was working towards something I called an string driving "engine". Using a Hex driver, I worked out a way to set off each driver separately and in sequence...much like the cylinders in a combustion engine), keeping up momentum of the string, driving it one sixth of the time, requiring only one amplifier, and avoiding cross influences between the individual drivers...oh so many great schemes... Nevertheless...half wave driving does not seem to be a practical theory, nor prove worthwhile in my experiments...perhaps you will have more luck. I can without experimentation with this as of yet concede what you say makes sense here and thank you for providing a clearer idea of the speaker functionality. Phase changes with frequency. Yes, it wouldn't matter if it were 5 wavelengths behind...clearly it can be ahead and true resonance is fantasy, close enough may be possible. The problem is, that it is frequency dependent and building a set or adjustable delay is not sufficient as this will need to be adjusted automatically by different amounts as the frequency changes. Handling polyphonic signals like chords with such a system...well that's a whole other debate... The phase shift occurs within the circuit. This is fairly simple to verify. It doe not occur at the pickup at more than negligible amount. I have measured it several times. The interactions that cause it, yes even the frequency-dependent offset, which are really just filtering interactions, can be neutralized. However...I was encouraged to look at it from another angle...instead of aiming for compensation circuitry, why not aim for a driver of an efficiency that would be close enough to being in sync at a wide frequency range. Hence all my designs that have worked towards this aim. The thin coil designs are not like pickups, but have a very low permanency of the core, and a high concentration of forces directly under the string. The result is that it works fast. The same spec coil only deeper will not produce these results. Speculation has been that the "thin coil" idea that I hit on works largely at resistance, not altering impedance too much with frequency. You have clearly made this point repeatedly. What you have done so far is excellent, and many have enjoyed its fruits. However, not all of us (myself point case) can get the same results as you for whatever the myriad of reasons are. That leaves us wit... the circuit is grossly... well gross in most ways. It does not take much googling on the web to see what a POS chip the LM386 is, honestly it is older than me and so prone to oscillation it's not even funny and there are more folks out there with bad things to say than good. The J201 preamp may be ideal, but I can not prove this as I have little success getting a better result than I can get with a TL082 op amp preamp and this is a mediocre at best op amp. I know the reasons you say LM386 ... it's good "enough" for our purposes, but we don't all have YOUR coils, so the circuit is the ticket and I am surprised you are as complacent with this. There is already the path for simple DIY. Let there be more as well, no need to limit. PS...you seem to have ignored that I said that #2 was false and therefore, theories following that rely on it are also false. I think you misunderstand the nature of a speaker, the core is the huge potted magnet that surrounds the coil...and that we are surrounded by magnetic fields all the time, what ever the field of a permanent magnet...as long as it does not move, it is the zero point. It is possible to create a neutral zone (not that it matters too much) by using equal and opposite forces as I did with a lot of the hex designs...but it does not change the principle. Think more about or experiment with pickups and consider why must a pickup have a magnetic field...then reflect back to this application...some answers are there... I did not ignore. Thank you for that. I do feel I understand that better at this point and you are corredt on where I missed this. I was under the assumption the speaker cone was magnetically at rest. So, please confirm so I can be very clear... a speaker cone is not magnetically neutral when at rest? Why did the equal forces stuff you did fail? It seems to me it would be easier to vibrate the strings under tension if they were truly at rest, not being pulled more to one side. Link to comment Share on other sites More sharing options...
psw Posted February 11, 2009 Author Report Share Posted February 11, 2009 However, not all of us (myself point case) can get the same results as you for whatever the myriad of reasons are. My apologies...I am only trying to explain my understanding...and much of this comes from years of discussion here and some of thread about many of these issues. I have never been the biggest fan of the LM386 except that it is easily obtainable, understandable and works. It is old, but there is little else outside of SMDs that can match it still. As for the J201 (unavailable here) I have never used it nor the fetzer ruby or similar designs...my suggestions largely come from experience (like the 100uF output cap) or the data sheet for stability that others have chosen to delete...mainly from ROG designs. As for half wave experiments...I can't know exactly where they are, but I did try using diodes in various ways to kill the half wave, also transistor based circuits. I found that these tended to cause distortions. At very early stages I was running a single string driver with computer generated signals...so I was able to test things like sine, square, triangle and other synthetic pure wave forms as drive signals.... As for speed, this has been tested by making duplicate spec coils of different depths...but speed is only a part of this and it may be assumption in large part...but it was only a part. The veracity of the design and the principles (for whatever reasons) has been shown again and again by others reproducing the design...often after failing by altering key aspects. Also, these are not one off things...I have been doing these since 2003...that's a lot of different and very many circuits, coils and different guitars...and in that time many others have joined to make suggestions, influence the work or verify much of it themselves. My impression was that you had come very close to getting the results I had and was well on the way...my coils are not that "magical", I couldn't be more forthcoming on the things...the latest tele driver is literally transparent. This, like so many was wound effectively by hand...the pictorial clearly shows the original being created in step by step photos. But, I don't want to argue about these things...I am attempting to have a discussion is all. It is difficult when I am not entirely clear of the question to provide the answer that best fits. I also can't say that I agree, though I do understand where you are coming from because I did feel and assume many of these things too...if I don't! However, please...prove me wrong! As for the speaker thing...well, I think you have taken my analogy a little too far...the magnet is largely to provide coupling with the string that is being moved. The "pickup like" configuration is only one way in which things can be arranged...magnets can be arranged to provide some 'neutrality' if you feel this to be advantageous...you can attempt to move the strings without a magnet at all if you like, same with pickups though no one has succeeded really. So...do what you have to do...let us see the results in a month or so. My direction changed in the early pages of this thread because I could not emulate the phase compensation that I worried excessively about as being crucial on the advice of LK. The suggestion was that if I concentrated on driver design, I may well hit upon strategies that would make the phase compensation redundant to a great extent...enough to create the effect that was being sought. Obviously, the commercial systems went a different way with extensive phase controls as outlined in the patents. I can't go through 300 pages of posts, nor do I have any motivation to. This thread is a very long discussion with many people over the years exploring everything and anything that comes along. However, much of what I had worked on were not really itemized, because the experiments yielded no positive results...some were mentioned in passing, others may not at all. Some things may only have come up recently, specifically, because recent people such as yourself have raised them, and I chose to share my experience with practical work in the area. I really don't know what to say...can I prove these things, does my design produce sustain...is the whole thing an ego trip on my part to show how great I am...perhaps, that is for you to decide. I thought I was involved in a unique thread populated by a rolling cast of almost obsessed tinkerers trying to better ideas towards a common goal...to that extent, all I can do is offer my experience and thoughts...and encouragement... so go for it... pete Link to comment Share on other sites More sharing options...
col Posted February 11, 2009 Report Share Posted February 11, 2009 (edited) It actually is fairly simple, regardless of the fact that offset varies with frequency. Implementation is no more complex than using a jfet, opamp, or an LM386. Point me to the points in this thread where others or yourself have tried and failed if this has happened, else I think the assumption can be made that the details of the theory were overlooked or others simply were not interested enough in the idea to pursue it. I used an all pass filter to alter the phase yonks ago, as you have explained it is simple to implement, as long as you only want the shift to occur over a small frequency band. This fits in with your switchable harmonics idea. My favorite version of this was set up so that the low end sustained the fundamental while higher up, it produces nice high harmonics - good fun to play with that one (or are you using some sort of gubbins in the feedback loop of an amp to tweak the overall phase response rather than just a simple all-pass ?) btw, have you checked out my (ancient) demo samples ? EDIT: the first song is not sustainer related - it's a demo of a fault in an update of the VST Host I use - the bend is broken hence the duff notes. Phase changes with frequency. Yes, it wouldn't matter if it were 5 wavelengths behind...clearly it can be ahead and true resonance is fantasy, close enough may be possible. The problem is, that it is frequency dependent and building a set or adjustable delay is not sufficient as this will need to be adjusted automatically by different amounts as the frequency changes. Handling polyphonic signals like chords with such a system...well that's a whole other debate... The phase shift occurs within the circuit. This is fairly simple to verify. It doe not occur at the pickup at more than negligible amount. I have measured it several times. The interactions that cause it, yes even the frequency-dependent offset, which are really just filtering interactions, can be neutralized. The most significant phase shift occurs due to the space between the driver and pickup. this changes depending on how far up the neck you are playing, so if you are playing open strings, it is roughly 36º whereas if you are playing at say the 14th fret its more like 100º. Unfortunately, due to the way a guitar string vibrates when you pluck it, there are some major hurdles when trying to 'fix' this. The charateristic timbré of a guitar is due in part to the collection of harmonic overtones that are generated when a note is plucked. As the note decays and these die off, the sound becomes more like a sine wave (more boring). what actually happens is that when you release the string at the 'pluck' two 'kinks are created. They fly on opposite directions, one towards the bridge and one towards the nut. when they hit the bridge(or nut) they are reflected back with opposite polarity (down the 'other side' of the string if you like). If we adjust the phase of our circuit to adjust for the phase difference (between pickup and driver) as the kinks are travelling from bridge to nut, the phase offset as the kinks travel in the reverse direction is increased. What we gain on the swings, we lose on the roundabouts. So with or with out phase 'correction' there will be a certain percentage of the power of the sustainer actively damping the strings, this damping has more of an effect at higher frequencies. Some harmonics for some notes will be more 'fortunate' and some less so, although there will not be enough 'fortunate' harmonics to sustain a natural guitar timbre. To test this out, try switching your sustainer on in normal (fundamental) mode. Now try playing the natural harmonics at the 12th, 7th and 5th frets. They will (depending on how well your sustainer pushes the fundamental), quickly return to the fundamental as the system actively damps those harmonic due to the process described above. If there was no distance between driver and pickup (say with some fancy laser optical pickup, barcode scanner type of thing in the same position as the driver), and the circuit had a reasonably flat phase response, these natural harmonics would be sustained correctly by the sustainer in fundamental mode. As a result of this effect, what the sustainer does as a side effect of the gap between pickup and driver is to cause the natural 'guitarness' of the sound to be killed off more quickly than if there was no sustainer. This is fine if you are using ****-loads of distortion, but if you want a clean or mildly crunchy sound, it takes the life out of your tone. At the end of the day it's one of those compromises. There are other issues that are equally important and similarly have no simple solution, but even with these problems, the sustainer is a great fun piece of kit. We've just got to get the things we can fix fixed to the max . I argee with Pete that by far the most important thing is to improve the efficiency of the whole system. The driver to reduce EMI and the circuit to reduce power consumption. Where I don't agree is all the voodoo hoodoo about speed and quickness . Mental models like that can be useful in understanding some concepts, but you have to be very careful about generalizing them, they are major simplifications, and will hinder further understanding of how the system works. Col points out recently that some things are not only difficult to measure, but may have no practical purpose. So...adjustability good, but it will need to be over a wide number of triggered frequency bands...or create a faster driver that will be good enough...well, that's my feeling anyway. I said that some things are difficult to measure and to standardize and that that made the idea of 'ballpark figures' for power requirements of little use. I certainly have never said that the things we can't measure are of no practical purpose, that would be daft. It would be extremely useful to be able to measure these things (permiability, core factor, field strength, etc. etc. etc) in a reliable, repeatable and cheap way. If we could do that, we could adjust for the differences in the materials we have and then a little math could make stuff like 'ballpark figures' practical and useful. We could then provide people with information that would enable them to build a good working sustainer first time rather than floundering around using trial and error methods. So, please don't misrepresent me like that . cheers Col Edited February 11, 2009 by col Link to comment Share on other sites More sharing options...
psw Posted February 12, 2009 Author Report Share Posted February 12, 2009 My apologies col...i did not mean to misrepresent you... however, in the above post you point out correctly how much the space between pickup and driver and the fretting of notes theoretically effects phase...so, any figures are in part on relative to a particular driver, circuit guitar combo...or at best a general design. As for the "fast" driver...I am not trying to propagate voodoo. All I mean by this is that the driver designs I have aimed for have sought to avoid much of the "lag or lead" inherent in the AC electromagnetic devices...ie phase differences. But even if this were possible, there would still be factors related to the circuit...but more so to the distance factors and individual installations perhaps... My suppositions have been based on numerous practical experiments...but hardly with the true rigors of science. I have not been aiming to prove a point, but make a sustainer! The results of the thin driver are assumed on things like direct comparisons with the same magnet, core material and guitar and circuit but with various deep and thin versions of the same spec. Also, different specs (like wire gauges) and different specs of the same designs. Many of these assumptions have been based on suggestions by others of why the thin driver works where others have failed...so perhaps I am wrong to suppose that these were reasonable assumptions to adopt. However, something is going on that makes the drivers of this design sensitive to specifications and quality of construction. There is a fairly wide margin though...the newer ultra-thin drivers also work well at three times less the thickness. I have and have put forward other "theories" on why the "thin coil" may work as well as it does not related to the fast driver things. Condensing the coil into as small a practical space (to limit EMI spread, etc), having the power concentrated directly under the strings, the number of overlapping winds within the coil that the thin designs maximise...none of which may be true, nor relate to any supposed voodoo. But the veracity of the design, no matter what the reasons are, has been shown time and again by those who have made it, and those that have altered it. But it is only my contribution...the dual coil designs that you followed are equally worthy, and using both this strategy and the thin coil ideas a reasonable approach, why not! There are also different aims and expectations at work...this is a good thing. You are right about the 'unnatural' sounding sustain that can result...to achieve more what you are seeking, yes we will need to get a bit more clever and I think things could be improved. However, at the present time, my work has stopped...I have achieved endless sustain and harmonics, and while I would like "better" I also acknowledge that feedback and endless sustain is in itself non-guitar-like and for now content that it works as well as it does. After all, feedback by traditional means also has this same characteristic. And the return to hex systems...why not...I certainly put in an effort in this pursuit. In the least, if dual coils are an improvement, and bilateral designs seen to be advantageous...then six individual coils (or even more) may well be even more advantageous too for EMI reduction alone! SO...I seem to be digging more of a hole for myself...again apologies for any perception or inadvertent misrepresentation...or dismissive...that is not really my intention. There is plenty of approaches, and much territory to explore. It is a simple fact that other than comparison tests, I have not been able to collect much in the way of concise data, nor the ability to effectively interpret it if I did. Nor is it my responsibility to do such things, all I can be responsible is the truth of my experiences. Plenty of people better equipped, able and crying out for such information have come and gone and failed to supply any more elucidation either...but, many have passed through successfully building working devices or taking the project a little further...often the same people! Link to comment Share on other sites More sharing options...
Donovan Posted February 13, 2009 Report Share Posted February 13, 2009 Pete- I am not thin skinned, so keep it up. I am not trying to be argumentative either, just challenging what may not make sense to me. This is part of the way I learn. I do tend to conced once the understanding arrives. or are you using some sort of gubbins in the feedback loop of an amp to tweak the overall phase response rather than just a simple all-pass ?) btw, have you checked out my (ancient) demo samples ? EDIT: the first song is not sustainer related - it's a demo of a fault in an update of the VST Host I use - the bend is broken hence the duff notes. Col- Currently, I am experimenting with cascading 1st and 2nd order all-pass filters, passive and active, after a closely controlled band-pass stage. I am cutting out in my 1st stage anything from below 80 Hz and anything above 2.4 kHz. This is in hopes of reducing the overtones that likely are contributing nothing "good" to the sustain. Results are good so far and I am theorizing this is the key to getting an overall phase correction. I have listened to your clips a few times and they are very nice, my favorite being the atmospheric noodle. Do you still have some fizz ? Listening closely, I can hear some, unless it is my crappy PC speakers. I have a strong fizz now to where distorted playing is my only option, but I have plans on fixing this with novel techniques which redirect EMI to a safer path,rather than attempt to "shield", but it is not my main concern at this point anyway as I am mostly a distortion aficionado. I do understand it affects this channel as well, but I can live with it at this point. I didn't comprehend your statament about broken bends, VST? What is that? The most significant phase shift occurs due to the space between the driver and pickup. this changes depending on how far up the neck you are playing, so if you are playing open strings, it is roughly 36º whereas if you are playing at say the 14th fret its more like 100º. Are you sure this is not just the fact that phase shift varies with frequency? What are you basing this assumption on? I can see how something like that might be easily/intuitively supposed, so can you provide a source for the theoretical to educate me? My intuition (though it is just that) would suppose that the macro distance is not the issue as there is a whole number of wavelengths within that distance and as they are whole waves, their effect on shift is inexistent. I would think it is the last, leftover, fractional, tiny, almost quantum portion of the wavelength remainder that causes the shift. Just conjecture at this point, but I want to know more. The charateristic timbré of a guitar is due in part to the collection of harmonic overtones that are generated when a note is plucked. As the note decays and these die off, the sound becomes more like a sine wave (more boring). This is exactly why I am feeling increasingly strong that a high-accuracy reproduction of sound is a necessity of at least equal importance as the driver with regard to the whole system... hence I am growing increasingly anti-LM386, though my latest circuit has finally after hours of tinkering, stabilized it enought to avoid oscillation at full gain. what actually happens is that when you release the string at the 'pluck' two 'kinks are created. They fly on opposite directions, one towards the bridge and one towards the nut. when they hit the bridge(or nut) they are reflected back with opposite polarity (down the 'other side' of the string if you like). If we adjust the phase of our circuit to adjust for the phase difference (between pickup and driver) as the kinks are travelling from bridge to nut, the phase offset as the kinks travel in the reverse direction is increased. What we gain on the swings, we lose on the roundabouts. If the latter portion of that is the case, how can anyone (with the currently proclaimed DIY configuration) be getting fundamental or harmonic mode reliably all the way up the neck on all strings? Or are they actually not? How can there be anyone NOT encountering phase problems? To test this out, try switching your sustainer on in normal (fundamental) mode. Now try playing the natural harmonics at the 12th, 7th and 5th frets. They will (depending on how well your sustainer pushes the fundamental), quickly return to the fundamental as the system actively damps those harmonic due to the process described above. Agreed, and I am not sure how I feel about this. I LOVE harmonics. A lot of my lead techniqure revolves around false harmonics. My mixed mode (as I do not yet have a 100% fundamental mode) kills artificial harmonics and brings them to either the fundamental or the harmonic that is dominant at that particular fret/string location. I argee with Pete that by far the most important thing is to improve the efficiency of the whole system. I too share this sentiment, but I do have confidence that much of the work to be done in figuring out the optimal driver has been completed by Pete and I feel confident in believing this and/or at least I feel that the evolution of the circuit is WAY behind that of the driver. I also know my driver is capable of reproducing all required frequenices. My challenge so far has been getting them to all be accesible at the same time. I know little of your driver, but would like to know more, sans trying to find it in this encyclopedia. What I do not feel confident in at all is the whole fetzer ruby idea. It is just not worthy of inclusion in this project IMO and I am continually surprised at complacency with this circuit and that it is not viewed as a huge hindrance to progress. I have looked at your circuit, but do not want to implement anything I don't have a complete understanding of. Being a circuit design noob, I am learning quickly, and the curve is quite steep, which I enjoy, but... I do not have the expertise at this point to look at your schematic and say to myself... "oh yeah, that cap is doing this and this op amp is biased like this to provide xxx" if that makes sense. I would be thankful if at some point you might re-post your latest schematic and give kind of a verbal breakdown of its parts and HOW it functions or rather how its components interact with one another... then again I realize this is a lot to ask of someone. I am also realizing through research that dual power supplies is probably going to be the way to go to eliminate a great deal of these phas issues by removing the need for a great deal of passive components. Link to comment Share on other sites More sharing options...
psw Posted February 13, 2009 Author Report Share Posted February 13, 2009 Good reply...I am still cranky for other reasons (partly health) so my tolerance/perceptiveness is low...ahhh...and it's my birthday and feeling older! Anyway... This is exactly why I am feeling increasingly strong that a high-accuracy reproduction of sound is a necessity of at least equal importance as the driver with regard to the whole system... hence I am growing increasingly anti-LM386, though my latest circuit has finally after hours of tinkering, stabilized it enought to avoid oscillation at full gain. Well...the LM386 is just a good basic indestructible battery powered amp...cheap, easy to use, relatively foolproof (although i did solder one in upside down and was frustrated for a while once...but even that didn't kill it). Interestingly, I have tried a few alternatives with less success. I am not sure why it should take hours of tinkering to keep an LM386 happy, if ROG hadn't deleted the data sheet recommendations, it would have been a lot happier. As for the fetzer ruby...I have always been vocal in my criticism of that....I have never used it myself, and it was suggested first in that tutorial, but not by me!...however, it has served reasonably well. I have no idea why all the stabilization caps were removed from the ruby though. The fizz thing is always a battle...but I do think that your set up is not quite right...I don't have any details on your bypassing and installation and such, and it is quite possible that this might be causing some problems. In the original sustainer strat I found that running the driver leads through the trem cavity helped a lot...just having these near the pickup wires cause some distortions. If you have not completely lifted both grounds and hots from the neck and middle pickups and bypassed properly, you will be having these problems, even if you have avoided squeal. Col and me have been a bit different in the kind of effect that we were aiming at...so there are some differences there. Col's driver is more like an Hb and is tending towards a cleaner milder more controlled sound I suppose, perhaps mine is a bit more primitive in effect...not to misrepresent anything intentionally you understand. As for the phase thing, I do now tend to think these days that close enough is good enough kind of...given that there are so many factors. LK had been kind of saying, and it seems to be true, that if it is a little out, the strings vibration will be pulled into sync and drift with the phase differences if it is close enough...in psw speak, fast enough...lol! If it is too far out, you tend to get poor response, potentially no response...or most likely only bass string response. However the commercial systems do do a good job with this extra circuitry. If the latter portion of that is the case, how can anyone (with the currently proclaimed DIY configuration) be getting fundamental or harmonic mode reliably all the way up the neck on all strings? Or are they actually not? How can there be anyone NOT encountering phase problems? I can assure you that people have gotten this to work on all frets in harmonic and fundamental mode...you can hear it in the clips...or believe me...but it does work. I understand why you may wonder why it should, especially if yours isn't...but it does work! Mine can work reasonable well and clean...mine can run at a lower intensity cleaner...but if you listen to my tele clips linked in red below...you can get a good idea of what the sound is like with a consistent almost direct sound on that guitar. As for the clean signal debate, this seems to have pretty well settled now...there were many, including LK who assumed a more squarewave signal would be a better idea...but that didn't seem to hold up for anyone. There is a danger in underestimating the drivers importance and over emphasizing the circuitry I feel...but everything is important...the installation and switching is the less discussed and critical factor as well. But, if all is well, it shouldn't need too much fancy circuitry. Many have felt that more power is the key (maybe for head room) but if the driver is efficient enough, it will need less power and will also address fizz and EMI issues in the process. The main problem I have by the way with the fetzer is that I feel the preamp is clumsy, especially the transistor byass trim thing...and it is not optimized for a clean sound at all. But, it is tempting to ramp up the preamp gain when using opamps that will distort the input of any amplifier chip. In fact, a buffer may well be enough for a lot of people, especially if they have great honking HB's in the axe! anyway...so late and the post interupted by phone calls...so I hope some of that makes sense....I'd encourage you to reevaluate why the basic version still is not working right. Perhaps post some details about the installation or symptoms and see if it can't be brought up to a relatively clean and consistent sustain, it certainly sounds like it is not working as it should and as a result you may be jumping to conclusions on the assumption that it is not possible to get such a result with the basic system. Sure, it can be "improved" but it can work better than you describe... pete Link to comment Share on other sites More sharing options...
Hank McSpank Posted February 13, 2009 Report Share Posted February 13, 2009 (edited) Just out of curiosity...how many on this thread have access to test kit? (an oscilloscope & signal generator, etc) There seems to be a lot of discussion about circuit design, phase, bandwidth .....but little mention of measuring it. I still think to get in the 'initial' ballpark circuit area (amplification wise), it'd needs an approach something the following... 1. Knock up a basic power amp (LM386 etc or whatever) ....connect it to your driver mounted on the guitar (or 'guitar plank' test rig!) 2. Inject 82.4hz (Open E string) from a sig genny into basic LM386 power amp. 3. Crank up the power amp input via the Sig Gen until the open E string starts vibrating sufficiently - measure the input to the power amp. (& the power amp output across the driver - just for reference, power levels etc) 4. Repeat steps 2 & 3 for the other string...right up to th 24th fret for the high E string & then finally doubling to ensure you capture the required harmonic content) 5. Construct a circuit to provide the right input level for your LM386. 6. Refine circuit accordingly!!! Sure it's step 6 where all the magic happens, but to get you to a good basic stage for your own particular driver, I can't think of any quicker way? Edited February 13, 2009 by Hank McSpank Link to comment Share on other sites More sharing options...
col Posted February 13, 2009 Report Share Posted February 13, 2009 Just out of curiosity...how many on this thread have access to test kit? (an oscilloscope & signal generator, etc) There seems to be a lot of discussion about circuit design, phase, bandwidth .....but little mention of measuring it. I don't have access to any of that kit, all I have is two multi-meters and some old simulation software. Fortunately, the simulations so far have been close enough that when I build the circuits, I get very similar performance in real life (at least for the things I can measure). I wish I had a scope, but I can't justify the expense of buying one... I could get a nice resonator guitar or decent recording kit for that kind of money, so its not going to happen. The real downside of not having the test kit is to check for deterioration in performance. e.g. I have a suspicion that my LM386 is dying slowly because it's being pushed just a little bit too hard. A scope would make checking this kind of thing a breeze. Link to comment Share on other sites More sharing options...
col Posted February 13, 2009 Report Share Posted February 13, 2009 (edited) Col- ... I have listened to your clips a few times and they are very nice, my favorite being the atmospheric noodle. Do you still have some fizz ? Listening closely, I can hear some, unless it is my crappy PC speakers. I get very slight fizz at the lowest notes. Those clips were made before I found a fault in my amp sim box. If you look at the waveforms they are lopsided and clipping at times on one polarity (there was a loose earth connection). I didn't comprehend your statament about broken bends, VST? What is that? that was in case you went to my soundclick page. The first 'song' in the list is nothing to do with sustainers, its all done with a synth. Its a demo I made to highlight a bug in some VST software I use. VST is Virtual Studio Technology, the software in question is Orion The most significant phase shift occurs due to the space between the driver and pickup. this changes depending on how far up the neck you are playing, so if you are playing open strings, it is roughly 36º whereas if you are playing at say the 14th fret its more like 100º. Are you sure this is not just the fact that phase shift varies with frequency? What are you basing this assumption on? I can see how something like that might be easily/intuitively supposed, so can you provide a source for the theoretical to educate me? My intuition (though it is just that) would suppose that the macro distance is not the issue as there is a whole number of wavelengths within that distance and as they are whole waves, their effect on shift is inexistent. I would think it is the last, leftover, fractional, tiny, almost quantum portion of the wavelength remainder that causes the shift. Just conjecture at this point, but I want to know more. If you think about how the string actually vibrates (a kink moving longitudinaly), you will realise that the physical distance between driver and pickup causes an electrical phase difference that varies, not just with frequency, but with what fret you are playing at. This doesn't have an effect on the fundamental, but it has an impact on all the harmonic overtones that make up the sound. that's why your pinch harmonics don't work. If there was no gap between pickup and driver, the pinch harmonics would not be killed! Think of it like this... (simplified one kink version) The kink is whizzing along the string, it's just bounced off the bridge and is heading towards the nut. It gets to where the pickup is, at this point the pickup is sensing maximum amplitude, and telling the driver to pull at hard as it can. But hang on, the driver is pulling hardest at a part of the string where the kink hasn't reached yet!. by the time the kink gets to where the driver is, the pickup is not telling the driver to pull hard any more, so the driver has eased off and the kink is not being maintained. Pulling hardest at the wrong part of the string quickly smoothes out the kinks and drives a standing wave only. (both kinks this time - worst case scenario) Now, what if you are playing at the 20th(ish) fret? By the time one kink has reached the driver (where ideally it would be pulled at maximum), the other kink is at the pickup(or very close) on the other side of the string. So the pickup is telling the driver to push at close to max power when it should be pulling at max power. In this instance, you are getting maximum damping of all but the fundamental. The charateristic timbré of a guitar is due in part to the collection of harmonic overtones that are generated when a note is plucked. As the note decays and these die off, the sound becomes more like a sine wave (more boring). This is exactly why I am feeling increasingly strong that a high-accuracy reproduction of sound is a necessity of at least equal importance as the driver with regard to the whole system... hence I am growing increasingly anti-LM386, though my latest circuit has finally after hours of tinkering, stabilized it enought to avoid oscillation at full gain. The LM386 is not the problem. the problem is the way the whole system works. It doesn't matter how hi-fi the power amp section is, the same problems will still exist. The only thing that you could improve radically by updating the poweramp section IMO is efficiency e.g. by using Class-d amplification. what actually happens is that when you release the string at the 'pluck' two 'kinks are created. They fly on opposite directions, one towards the bridge and one towards the nut. when they hit the bridge(or nut) they are reflected back with opposite polarity (down the 'other side' of the string if you like). If we adjust the phase of our circuit to adjust for the phase difference (between pickup and driver) as the kinks are travelling from bridge to nut, the phase offset as the kinks travel in the reverse direction is increased. What we gain on the swings, we lose on the roundabouts. If the latter portion of that is the case, how can anyone (with the currently proclaimed DIY configuration) be getting fundamental or harmonic mode reliably all the way up the neck on all strings? Or are they actually not? How can there be anyone NOT encountering phase problems? The cancellation of those 'kinks' just causes the note to quickly lose all its overtones and settle on the fundamental. unless you are using some harmonic mode that cancels the fundamental in which case it will settle on what ever it's base harmonic is, all richness due to extra additional overtones is actively removed from the signal In otherwords the fundamental (for that mode) is the only frequency that is not killed by the gap between driver and pickup. (I guess that there will always be some level of harmonics overtones present because the vibrations of the guitar will be adding them as the system removes them, but they won't be half as strong as they should be. And of course artificial ones added by your hard earned playing techniques will be wiped) EDIT: here are some links that will verify the stiff about 'kinks'. slow-mo video This is the low E unfortunately, but if you watch carefully frame at a time, you will see the two kinks. One goes to the nut, the other first bounces off the bridge (very quickly) and then follows the other kink to the nut. together they form a kind of moving zig-zag up and down the string. try and imagine the difference if a note was being fretted high on the neck, and then think about how that relates to the space between our driver and pickup. explanation with diagrams Its a shame that the explanation here is with the string being plucked in the centre rather than close to the bridge, but it still shows the basic idea. If you pluck closer to one end (like we guitar players do), the string will for more of a zig-zag effect as in the video, rather than the result in the diagrams which form more of a standing wave type of shape. The difference can be heard by plucking at the middle of the string and then near the bridge. Col Edited February 13, 2009 by col Link to comment Share on other sites More sharing options...
Hank McSpank Posted February 13, 2009 Report Share Posted February 13, 2009 (edited) I don't have access to any of that kit, all I have is two multi-meters and some old simulation software. Wow...I'm impressed you've honed your circuit without access to at the very least a scope! (genuinely!!) They're actually not that expensive nowadays - I paid about £110 for a dual trace Digital USB storage scope ...this is an external box solution which is hosted by my PC ( http://www.darkwire.com.au/html/dso-2150_usb.html ). You could always buy a secondhand one for the project (they're about £50 on Ebay for a Tektronix 40Mhz model) & then flog it on afterwards - that way there'd be zilch in the way of financial loss. I'm fortunate to have picked up a HP Sig Gen that my work were literally throwing out...but again, you can get software to run on a PC that serves as a sig gen (using your soundcard). I can't even begin to imagine how anyone could dedicate the amount of time needed to get results without test tools. Sure, you can get quickly the generic results by following the blueprint by many on here, but it'd be rather difficult to take this project down any new avenues easily, without at least seeing what the signal levels are across the driver (along with seeing if the signal is distorting etc) - without a scope you're literally are stabbing in the dark (or at least stabbing in a very low light room!). How else are you going to see the frequency response of your amp/driver - see what phase changes there are between input & output...& so on? Is anyone on here using one? Edited February 13, 2009 by Hank McSpank Link to comment Share on other sites More sharing options...
col Posted February 13, 2009 Report Share Posted February 13, 2009 (edited) EDIT: here are some links that will verify the stiff about 'kinks'. slow-mo video This is the low E unfortunately, but if you watch carefully frame at a time, you will see the two kinks. One goes to the nut, the other first bounces off the bridge (very quickly) and then follows the other kink to the nut. together they form a kind of moving zig-zag up and down the string. try and imagine the difference if a note was being fretted high on the neck, and then think about how that relates to the space between our driver and pickup. Oh dear, I'm quoting myself :-/ I was watching this clip again, and it looks like if you pluck the string about halfway between the neck pickup and the bridge (roughly where many folks hit their guitar strings) then the two opposing kinks will be almost exactly the distance between driver and pickup apart (obvious really when you think about it). This happens in both directions. I'm surprised I've not noticed this before, but basically it means that the 'worst case scenario' for active damping of the harmonic content is also the most common scenario in normal playing. You would have to play in an unusual way for this not to happen ( Edited February 13, 2009 by col Link to comment Share on other sites More sharing options...
Recommended Posts