Low-tech approach



  • the 2 magnets would just lock each other in place.
  • Unfortunately yes, it would simply be overpowered by the implanted magnet as ThomasEgi said. It's similar to the issue of using a reed switch and a compass spinner, the magnet will simply stick to anything nearby which can interact with a magnetic field. The lines I've been going on are that the compass spinner will be be the only magnetic or ferrous item in the project.
  • I like where your going with this idea. when I read this I immediately had 3 ideas come to mind tell me what you think.

    1: magnetize a disc magnet on the opposite axis as they are normally fielded then ad a small bump and coat the magnet in TiN. Because the magnet will not adhere to the tissue it could be technically possible for it to float and spin to magnetic north inside the body and the bump would allow you to feel the magnet shifting about.  I somewhat doubt this would work due to the tension of the skin against the magnet but if it does it would be a super simple solution. 

    2: float a magnet in a glass tag with an elecro-mechanical transducer at the "north" end of the tag when you point to magnetic north the magnet slides in to the transducer causing it to create a small electric shock letting you know that you are pointing north.  This is based off the same principles that make a sound-powered phone work.  issues being that you need a way to float the magnet and the transducer has enough sensitivity to produce a noticeable electric pulse. 

    3: This one is a bit more complicated and not as well thought out but I was thinking a gyro-compass powered by something similar to the rolex automatic winding mechanism could be used. it would be pretty difficult to manufacture something small enough that anyone would actually be willing to implant it but I dont think its outside the realm of possibility.  also I think that a mechanism similar to the automatic watch winder could be a viable source of low voltage power for a large number of implantable devices. 

    what do you guys think?? 
  • 1. too much friction, the magnet will not float but be held in place by the tissue.
    2.tricky, but it might work with a piezo. given it is not ferromagnetic.
    3. not exactly "low tech". if you drag that much complexity into it, you are better off with regular electronics.
  • Osteth brings up a good point about discs which are magnetized from one side to the other as opposed the standard face-to-face. A disc like that can spin pretty freely inside a low-friction container provided it is given a pivot point. Imagine a CD with a marble inserted into the hole. That sort of contraption would spin for a long time if given a twist. There is some merit to the simplicity of a compass with so few moving parts. Of course this type of spinner would have to be contained in a sealed vessel and probably filled with alcohol or deionized water. But it wouldn't be hard to prototype and see if it merits further testing.

    I'd like to hear more about what you have in mind with #2. As mentioned before anything that can interact with a magnetic field will likely make the compass stick to it.

    I wish I knew more about the auto-winding mechanisms. It would be beneficial to have a wound spring that is constantly tensed and then when a small compass spinner lines up and taps an internal switch it could deliver a large BUMP to the whole device via a hammer/anvil powered by the spring. I don't have an analogy for that idea.
  • after sleeping on it a bit I came up with a more thought out design this is based off #2 with some ideas stolen from McStuff's proto. here is some super expert level 3d drawings! https://www.dropbox.com/s/8ukqtlpt6ij9i4a/compass%20implant.png

    this seems a bit more realistic and practical than my previous ideas to me. just gotta find a magnet company to magnetize the disc magnets on the other axis and they could likely manufacture them with the divots in them as well the casing could be made pretty easily by a company that makes stamped shields for electronics components. 
  • I recognize the lines you're thinking along and they are much more practical so please keep your brain churning on these ideas. I've explored some of these ideas and I'll share my results in the hope that you can see a way around the roadblocks I encountered.
    #1. Trying to hold a compass spinner with pivot points on the top and bottom requires a level of precision I cannot reach with my skills and tools. I have tried and you can check out my results. Someone more skilled may see an easy fix and I welcome suggestions and criticism.
    #2. If you have a pin on your compass spinner and it needs to touch another pin there also needs to be some way for the spinner to pass by or it will simply hang there. A compass has a very weak pull, even with strong magnets.

    Super Magnet Man has Nd discs magnetized across the diameter. I think these could be really useful. Another option are the wedding rings that are magnetized the same way.

    I don't want to herd this discussion around what I've failed at because someone else may approach the same idea much better. Keep your thoughts coming.
  • I have all the tools and such to do this so long as I can get a a balanced magnet with a precisely centered divot on each side.

    I have contacted a couple magnet manufacturers to see if they can build the magnets. I am having a feeling that and electro-mechanical transducer small enough for enjoyable implanting will not generate enough electricity to be noticeable.

    I am working on getting some materials together for a proto right now could be a couple days to find everything I need and I am tech-ing  up my thoughts a little bit. I am currently thinking the best way to accomplish this is to solder a mirco pogo pin to a button cell battery about the size of the magnet. then I have some shields for pacemaker VCO's that I am going to solder another pogo pin in to and sandwich the magnet between them. I will them solder a fuzz button on to the magnet and laminate a piece of kapton tape and a peice of copper tape on to the north side of the shield and connect it to an LED connected to the opposite lead of the coin cell battery. when the fuzz button sweeps the copper tape it will cause the LED to light up and because the light would only be turned on when you are actively attempting to find north the battery should last for an extremely long time. It adds a bit of tech to the deal but it uses all parts that are readily accessible except for a custom magnet. still working on figuring out the exact configuration to make it as small as possible i'm toying with it under a microscope here t work though and I think I could work.     
  • Well, I found a huge issue with this approach.... the magnet is constantly pulling itself to the copper tape and completing the circuit... so that's a fail.... back to the drawing board.
  • Use double-sided tape and aluminum foil if your copper has iron in it?
    What kind of batteries are you using that don't respond to magnets?
  • https://www.dropbox.com/s/8v2dtewar1k3vkg/20140813_232527.jpg

    here's a couple pictures of of attempt #2 another fail I broke the wall of the shield off trying to apply the tape. 

    I am using a button battery that came spare with my micrometer its outside of the field because I stacked the boards and the magnet is weak. I'm thinking there is a better way to do this but I'm going to have to plan it out better because hacking the random shit I have laying around isn't working so well. 
  • Sooooo.... how do you feel about going high tech on this one?? take a peak at this bad boy:

    smaller than a grain of rice and a ridiculous amount of features at a highly sustainable power consumption

    just a thought, and I have easy access to a pick and place line to get it assembled... 
  • Nice workspace! Can you describe what we're seeing? What are you using for materials and how will it work? I've been working on pretty large scale for prototyping and it looks like you're working on a smaller prototype than I have made so far. That's exciting.

    I'm confident that using the random shit lying around is going to become a working prototype. I don't want to do anything to discourage anyone from venturing outside this mentality though.

    I would like to point out my criteria for a prototype and get input to further the list.
    A. The device must provide haptic feedback that can be sensed by ordinary contact.
    B. The device must provide haptic feedback while facing only the direction of interest.
    C. The feedback mechanism cannot cause damage if implanted.
    D. The compass must continue to work after being inverted.
    E. The design must be scalable for implantation.
  • On my second attempt I used a cylinder magnet soldered between two small pogo pins ( the kind used in electronics test fixtures)  with a fuzz button soldered to the north pole of the magnet to engage a small  piece of aluminum tape on the north wall of the shield. 

    I essentially scavenged a couple of VCO's that go in to pacemakers. The shield is a non-ferrous RF shield that normally goes on the VCO and the bottom is two PCB's for that VCO as well. I drilled one out to give me room for the magnet to spin inside and soldered it to a second PCB. the bottom pin ( that is connected to the magnet) goes through a via to the ground plane where a button cell battery was going to be soldered. Then I was going to solder a LED to the battery and run a piece of aluminum tape up to the inside of the wall of the shield where the fuzz button could engage it and turn on the LED. 

    My idea was that it could be sealed in bioglass then implanted and when the magnet faces north the light would turn on and your arm would light up. or wherever you implanted it. I think it would will work (so long as these magnets and actually right themselves to north) but the assembly is a total pain. 

    anyone got a high resolution 3d printer that would be willing to lend a hand??  
  • How high resolution? we have a group owned "peachy" printer, see one of the other threads for more details.
  • Im thinking somewhere in the range of 1-3mil line thickness would work fine just need to print a casing to hold the magnet into.  also I would need help generating a file to print because I have no 3d cad software and that stuffs expensive. funny thing is I have built 11 3d printers for people and have never actually used one lol. 
  • Minimum wall thickness of the peachy is 0.2 mm, so that should be fine. go read the thread to understand how to request a print.
  • Ok, thanks for letting me know I will look in to that. going to try a couple more rough prototypes to get all the possible problems nailed down too. 

    Do you guys see any problem with using a light instead of a haptic response?

    LED's are tiny and would allow the device to be pretty small but I guess I am brainstorming to try to see if I am missing anything. 
  • I can now confirm that an N52 magnet is strong enough to right itself to north if floated in liquid. I sealed one in plastic and floated it in water and pushed it away from north 100 times and it pulled itself to north each time.  Which means that any of the magnets people are using now for implants are suitable to making a compass. 

    So, now that that is out of the way I have come up with another idea that will defiantly work and be implantable. however it requires having some things made custom. 

    First here is a little drawing to help make sense of it: https://www.dropbox.com/s/s34uhcpmnl1k7su/compass%20implant.png

    The base of the implant is a bioglass capsule. It works on the same idea of triggering an LED under the skin then you point north.  The capsules size will have to be decided based on the smallest sized battery that we can find that will properly illuminate a tiny surface mount LED. 

    The triggering mechanism is a custom bit of PCB with a Target looking trace on it.  The center circle of trace will have a via to a pad that gets soldered directly to the battery.  The outer circle gets routed to one side of an LED that is soldered directly to the other side of the battery.  

    A fuzz button is epoxied to the north end of the encapsulated magnet which is used to complete the circuit. because of the triggers bulls-eye design the orientation of the connection doesn't matter so long as the fuzz button hits the bulls-eye it lights the LED. 

    Fuzz buttons and PCB traces are made of gold so I have no concerns about corrosion but the PCB may need to be made or rogers ceramic instead of fiberglass to avoid the liquid permeating the board.

    I used this magnet for my tests so I know that it will work for this. 

    For my test I sealed the above magnet in a piece cut off the back of a ball point pen. I then put it in a cap of water an it pointed itself north every time. I then put it in a large bucket of water and it would scoot itself across the bucket all the way to the edge so it should move along inside the tube as well, to make contact with the trigger. 

    Questions, Concerns, Ideas, etc??
  • Will the fuzz switch make enough contact to complete the circuit between the bulls eye and the ring around it? What about a series of vertical lines of alternating contacts similar to the ones used under the contact pads of keyboards, gamepads, and calculators.

    So far I have been focusing on making a haptic compass to simulate the NorthPaw. The reason was to hopefully duplicate the additional sense of direction as opposed to checking a light mounted under the skin of an arm. My sense of direction is awful. @osteth is going another direction and that got me thinking. What if a magnetic disc was floating in a dark case? On one pole of the magnet would be a glowing node and when it passed under a window on the north side of the dark case it would be visible from under the skin. Tritium and phosphors glow for 10 years without batteries.

    Imagine this, you look at the underside of your arm and there is a small bump. You swing your arm around and when you see a little red light appear your arm is pointing north. There are plenty of problems with this idea.
    • Tritium is radioactive.
    • The tritium would have to be isolated separately from the compass spinner so it becomes bulky.
    • Tritium isn't available in bulk so likely would have to be purchased as a keychain. At least it already comes with the phosphor.
    • The phosphor included with a keychain is unreliably sourced and potentially more hazardous.
    • Grinder implants in the underside of the forearm have not been attempted as far as I know.
    • The glass casing needs to be coated so only a window is still clear. Presumably from the inside.

    This was some out-loud thinking but maybe it will give someone ideas.
  • About CAD tools. FreeCad will do for this purpose. As you don't need full parametric dependencies on everything. I could give you guys a hand with that. about magnets and compass and switches. The only reliable thing that comes to my mind is a spherical compass and an optical switch. Like having a section of the magnet transparent so an IR-led can shine onto an fototransistor, driving the LED. OR by using a reflective method. If that drains the battery too much, an ATtiny could be run at very low clock speeds to periodically check for said condition (it's about 130nA current demand running the watchdog only, using a CR1632 sized battery that makes a month of operation.). .Or use a magnetometer chip in first place.

    The Tritium is a very dim light, and green. My guess is most of the dim green light never makes it past the blood/skin.

    So far, liked the original idea of mechanical setup best.
  • Another thing to keep in mind is the location. Having an implant somewhere on the arm may be confusing if it is a passive haptic sense. Your hands/arms change positions all the time and often are out of sync with the direction you are actually facing. I spoke with the designer of the southpaw recently (the ankle mounted directional device) and he says they chose that location specifically because your legs are almost always vertical and pointing the same direction you are. 

    If we end up going with some sort of haptic trigger or electric buzz then perhaps the toes or somewhere on the foot would be a good implant location. 
  • @ThomasEgi, an image search of "tritium keychain" shows a color range from green to red. I would guess that the type of phosphor used determines the color. Green is probably the cheapest or brightest but red is possible. I had that same thought though.
    A tiny spherical compass would work as long as it's properly weighted. A plain spherical magnet in a lubed-up sphere will just point down as I found out a month ago.
    I have no plans to discontinue looking for a purely mechanical design. The simple LED version @Osteth is looking into has merit different from mine.

    @meanderingman, you are correct. I understand the logic of the ankle placement for a haptic system and I would still plan for a haptic version to be leg-mounted or at least torso-mounted. An LED doesn't do much good under your socks which is what I meant about mounting under the thin skin of the forearm. Your message bears repeating.


    I'd like to combine some of these ideas and some from other threads and put them here. A
    spherical compass spinner is outfitted with a rechargeable battery,
    charging coil, LED, and a vibration switch. Nothing inside would be
    smarter than the charging circuit. When the vibration switch is
    activated by tapping the LED lights up for a moment. The LED would be
    positioned at the north end of the compass and shine through the
    skin. Over the implant would be a tattoo of a compass rose centered on
    the implanted compass. The light would appear under the compass tattoo to indicate how
    far to spin in order to face north. It would not consume any power other
    than what the charging circuit uses when idle.
    With everything mounted on the compass spinner there would only be one moving part and all electrical connections could be permanently secured.
  • Ok, I did not fully realize the scope of the idea or need for haptic feedback,  The main reason I thought a non-haptic system would work is because it will be very difficult to make a device that can provide a noticeable haptic feedback, especially if you want to have it constantly. 

    so what your looking for is an implantable version of the north paw anklet. which to my understanding is not intended for permanent use but only as a trainer for your sense of direction. after wearing it for a while you should strengthen your sense of direction to the point that you no longer need to continue wearing the device. 

     I am not really seeing a viable way to make a low tech haptic compass that is implantable for several reasons 

    mainly, you want north to feel north at all times. because navigation via magnetic north is affected by altitude and it is being planted in a moving object a gyro is required for the device to be able to read north at all times. due to the size and weight of a mechanical gyro compass this limits you down to essentially only one option which is an electronic compass IC.  thus negating the low tech approach entirely. 
    alternatively you could maintain the low tech approach by developing a pointing compass  which is not as effected by outside factors because it requires that you orient the device in a particular way and then it will point to north. however, the easiest way to read this type of compass is reading it optically. which is not haptic feedback and augmenting a pointing compass for haptic feedback will take a small feat of doing. because you have to engage a trigger which causes problems for the compass:

    A: it will inhibit the movement of the magnet.
    B: the trigger has to initiate a noticeable haptic response, not a simple task when working on a micro scale with out the aid of electronics. especially when the trigger has to be sensitive enough to be initiated by a floating magnet and rigid enough to not provide false triggering. 

    having said this, your idea for whiskers seems feasible on paper but when implanted the whiskers would likely have to essentially be attached to nerve endings in a spiders legs type configuration around the device, and they would have to be brilliantly machined in a way that they allow the magnet free motion but can still be triggered. this requires machining techniques yet to be developed and likely surgical skills that are to my knowledge also yet to be developed. 

    and even if you managed to overcome those challenges the device would only work when you orient you body correctly and are not moving. 

    I absolutely see the merit in having a compass onboard at all times but for anyone to actually implant one it has to be economically feasible for someone to attain the device and have it implanted as well as having usefulness above and beyond carrying a button compass in their pocket or using the compass embedded in their smartphone.  

  • I've been following this thread and I know you want a low tech solution but I had an idea. I want to get 4 magnets in my hand. If you had this done you could build a glove with 4 coils. When holding the hand level the magnets would vibrate in correspondence to north. So if north was between two magnets, they would both be vibrating weakly. as you turned one coil would strengthen and the other would weaken. this could also be accomplished with just one magnet just have it send noticeably different pulses for south and north. say quick ramp ups for north and more square wave type for south. i know it's a bit more complicated but at least this way you could turn it off as I fear that having a thing that ticks everytime you're facing north could get distracting. Just an idea, not quiet what you are planning but something that could test the compass at the least. mind you the sensor for magnetic north would have to be well away from your hand or the magnet may intefere. 
  • Are you talking about picking up the vibrations of the magnets or using coils to vibrate the magnets??  
  • using coils to vibrate the magnets

  • @Osteth you might want to update on modern machining processes. After having seen stereolithographic products you'll see that producing even complex micromechanical structures is totaly possible (altho not really low-tech, I have to admit that). Also no need for a gyro, the magnet's inertia, excellent bearings and some small physical tricks are enough to keep the magnet oriented.

    There are also other options for mechanical solutions. Like having a small weight that can move around freely, but when the magnet is pointing north it'll lock the motion. So you like.. shake your hand a bit and you'd feel the small weight bouncing, unless your arm points north.
  • @thomasEGI  I will have to look in to stereolithographic machining, have not heard of that before.  but in order to maintain constant bearing on a moving platform a gyro or gimbals are necessary. and keep in mind that we are talking about an implantable device. a magnet small enough to implant has an inconsequential amount of inertia and good luck finding good bearings on that scale. 

    your mechanical solution is also another version of a pointing compass and will not provide constant haptic feedback to north's position as Mattguy is looking for he want to have a constant reminder that north is X direction. and you are correct a pointing compass is very possible via low tech mechanical means. but this will only work when your body is oriented correctly and you are not in motion. 

    The only viable solution to this is to use a compass IC chip but that is beyond the scope of the experiment. 
  • for a constant north reminder there is indeed no way around electronics. the idea of an implant doing that job was labeled southpaw. http://discuss.biohack.me/categories/southpaw
Sign In or Register to comment.