Help Making A small ElectroMagnet!

[voyager2]

Hi All,
As some of you probably know I am working on a robot fish.
Here is some background info...
*Controlor: ATmega 8/168.
*Length:10-12cm
*Width:2.5cm
*Bouncy:neutral(hopefully )

I am faced with a problem, and have turned to SoR for help.
I need a small electromagnet with about 3000winds, if you know of a website that sells to Australia please post a link.
If you know how to make an electromagnet with out sitting and winding for days on end, please help

[Admin]

Not sure why you need one but . . .

Electromagnets are fairly weak, and mostly just heat up when you apply a DC current to them.

You can easily/cheaply make one . . . get an iron bar and wrap it with motor wire.

[voyager2]

Thanks Admin,
The coil is to drive the tail of the robot fish, so it does not have to be strong
You might be interested to know I got some of my inspiration from your robot fish

[waltr]

3000 turns, mmm, lets see, 1821, 1822, 1823.... what's that you say?.... oh no was that 1328 turns or 1832 turns. Oh well, unwind and start over.
Oh yea, been there.

If you can get access to a lath and set up a turns counter, winding this isn't that bad.

[Admin]

3000 turns, mmm, lets see, 1821, 1822, 1823.... what's that you say?.... oh no was that 1328 turns or 1832 turns. Oh well, unwind and start over.
Oh yea, been there.

You can measure the distance that 20 windings take, and extrapolate from there. Being off ~5 windings out of 3000 won't break anything.

[little-c]

depends, having looked at hoists and winches as part of a possible project. coils tend to stack up, as such the diameter changes. messes about with gearings. I'd use admins method but recalibrate every layer, or every other layer. depends on how wide the electromagnet is and the gauge of the wire. just keep a count on paper of how many coils you have calculated.


if your making your own electromagnet, its a pain in the neck. sucks, but then again not everything is nice. why do you think you get a robot to do it?

[Soeren]

Hi,

Usually, you calculate the required amount of wire etc.
You know the volume the coil can fill and you know the diameter of the wire (or can calculate the needed wire dimension when you need 3000 windings).
There are formulas with different constants for the amount of round wire that can fill a certain size slot (some will be air in a cross section). some constants are for meticulous winding each layer nice and even and some are for a more randomized winding technique.

After calculating, a hand drill can be used to wind the entire thing.

[Admin]

Another idea just popped up in my head.

Use a multi-meter to measure the wire resistance after wrapping it 20 times. Multiply that value by 150, then keep wrapping until you get that value in future measurements.

Not sure how accurate this will be, but probably darn close!

[voyager2]

Thanks for all the help...

if your making your own electromagnet, its a pain in the neck. sucks, but then again not everything is nice. why do you think you get a robot to do it?

...I would make a robot do it, but I don't think the $50 robot is up to it!
I'll try Admins multimeter method and Soerens drill method to wind the wire.
Will post back later to tell you how I did, keep posting!

Thanks

 Voyager2

[waltr]

The processor for the $50 bot could be coded to read a photo-interrupter coupled to the drill and count the number of turns. A serial interface could then display the total number of turns counted.

[paulstreats]

A stepper motor can also be used to make 3000 turns

[voyager2]

Bump...

Any other ideas?
I tried motor coil wire and made a vibrating orange thing with it.
No lifting capacity, just got hot (as Admin predicted).
And yes, when i tested the magnets inside a motor, they could lift a few (15) paperclips in a chain...

[Admin]

idea 1:
Separate the metal objects and the magnet by a small plastic plate.

This plate is connected to a servo . . . when you want to release an object, activate the servo to push the plate away from the magnet. After a cm or two, the metal objects will separate.


idea 2:
Use AC current.

[Soeren]

Hi,

Any other ideas?
I tried motor coil wire and made a vibrating orange thing with it.
No lifting capacity, just got hot (as Admin predicted).
And yes, when i tested the magnets inside a motor, they could lift a few (15) paperclips in a chain...

Wire comes in all kinds of dimensions.

I just had some 0.1mm diameter wire that I needed to get rid off, as I had coiled it up around a box when I took an iFlash inductive bicycle lamp apart (to find out how I could optimize it for charging a battery). The coil inside was 3,000 windings (yep, I counted each of them), on a former 7.75mm in diameter and 18mm long.


Well, I made a simple holder for a drill


and wound it on a 10mm diameter diameter bit holder with 2 cardboard endstops and something (a bit of MDF and a hose clamp) to hold the cardboard in place (carton pieces around 8mm apart)



and wound most of the wire onto it (minus a few meters, as I sped up too much from boredom and wanted to get it over with).


Don't mind the triangular piece of scrap MDF, it's only there to provide backing for the cardboard, so no reason to waste time on shaping it.







Afterwards the coil collapsed a bit when I removed the former and the cardboard sides...

but I taped it over, soldered on wires added flex and taped the wires in place (sorry no piccies of that, it's not that interesting to look at either - just imagine a small square wrap of beige tape with two wires exiting from a corner.


With a bit of steel and two small neodymium magnets (each 6mm in diameter and 3mm thick) and the coil in between, it works from around 1V (or less), but with too little power I think. cranking up the voltage, it does a good job, especially considering nothing was optimized and the gap where I put the coil was about 4..5 times the thickness of the (now compressed to around 5mm) coil.
Right now, it has been boiling for about 4½ hour on 12V (at 65.6mA) and measures a stable 55.6°C at its hottest point (measured with an IR thermometer gun), which is OK, as long as it's in free air (apart from a lot of window tape), but if sealed into a silicone body... Hmm, if made thin enough, it would actually be more or less water cooled 


Some of the points to note are...

• If your coils gets very hot, you are probably using (relatively) thick wire and pumping plenty of current into it. My coil was only luke warm at 9V.
• 3,000 windings is way too much (I think I got around 2,000..2,500 windings or less (but it's a very rough guesstimate) so if you want to run it from 3V, you need fewer windings, but do wind it bifilar (or even trifilar, quadfilar or ...), to be able to get more power packed into as little area as possible - a 0.1mm wire packs much neater than a 0.4mm wire, as less space is occupied by air.
• When I tested it at 12V (no H-bridge, just manual connecting the leads, it had quite a bit of oomph, so an optimized design, perhaps with a magnet going through the middle of the coil, should equip your fish with a hot piece of wiggly tail 
• Cardboard gets a little rough at the edges when cutting them with a hole saw
• Oh yes, the iFlash is grossly inefficient

If I had to make a tail wagging fish, I'd keep the coil steady and mount the magnet on the tail "wagger", so no wires had to break from bending fatigue.

Most important now is to experiment a lot before you make your final choices.
And please don't ask for more ideas until you have really tried out the ones you allready got - it just gives the impression that you don't really give a darn about the time people spend on your project.
So now you have to do some (read: a lot of) research and report back your findings - after all, it's your fish, so you eat it 

[blackbeard]

Ok here is a long shot but do you have a dot matrix printer or an electric typewriter with built in printer? They often have a selenoid powered hammer to depress a daisywheel onto the ribbon. Might work and it's free.

[Soeren]

Hi,

Electric typewriters (and some TTY terminals if any is still to be found) using a typewheel have got this solenoide, but I have never seen them in dot matrix printers, even if I have taken my fair share apart. Dot matrix printers usually integrates all the needle solenoides into a large heatsinked housing.

I don't see how such solenoids could be used for the fish tail though?
(I'd think it needs to be as flat as possible in the business directions).