Coil Winder

I've been meaning to get started helping in some way with this project and the direction Rylan pointed me was to a coil winding project on github:https://github.com/Rylangrayston/winder I'm not entirely sure about the coil's material and what generic specifications I should be looking to satisfy. I assume the idea is to put together a simple project that will allow all the beta testers (and anyone else) that wants to create different coils or additional coils to help prototyping different designs/constraints, but I'm not sure what kind of space the coils are fitting into. Can any of the beta testers at least provide me with an approximate AWG/composition for the wire? Any other information about the coils (radius, height, turns), and any additional information about the magnet (distance from coil, flux, composition, etc) would really help me get started in the right direction.

Thanks for your time everyone.

The coil we aim to make with our winding machine is a flat disk shape with a diameter of 11 mm and a height of 2 mm
we are using 38 gage wire. We wind the coil to 32 ohms.

I dont know if this is the best coil shape, more experimentation is needed and if some one uses different magnets or more power that can change everything.
Thats why we are taking the time to post the winder files so you can experiment with better coil designs

... they look a bit like this:

Attachment 1704

We had to make our own winder because we have been experimenting with wire as small as 52 AWG
which is so small that you cant even pull on it to make it come off its spoll, it just breaks instead of spinning the spool.
So we had to make a stepper driven thing that unwinds the spool at the rate that we wind the coil.... tricky stuff but it had to be done

Quote:

---

Originally Posted by rylangrayston

The coil we aim to make with our winding machine is a flat disk shape with a diameter of 11 mm and a height of 2 mm
we are using 38 gage wire. We wind the coil to 32 ohms.

I dont know if this is the best coil shape, more experimentation is needed and if some one uses different magnets or more power that can change everything.
Thats why we are taking the time to post the winder files so you can experiment with better coil designs

... they look a bit like this:

Attachment 1704

We had to make our own winder because we have been experimenting with wire as small as 52 AWG
which is so small that you cant even pull on it to make it come off its spoll, it just breaks instead of spinning the spool.
So we had to make a stepper driven thing that unwinds the spool at the rate that we wind the coil.... tricky stuff but it had to be done

---

Is 38 being chosen for price, availability and speed of manufacturing, or would you prefer something closer to 52? and what kind of wire? Looks like bare copper in the picture. I have some 34 I can use for now so I'll start with that.

34 AWG in my experience takes up to much space, and can handle far more current than we push threw the coil.

38 is small enugh that when we make a 32 ohm coil it dosent end up being way larger than the little magnet we are moving.
We only push 100 mA thru the coil and the magnet we are moving is very small so when we use a very small wire we get a coil who's windings all fit inside the field of the small magnet.

The copper wire isnt bare its has a thin varnish coating.

Do you have a coil winder machine built? if so post pics!
:)

No, I'm working on it. I got some 50 AWG polyurethane coated magnet wire coming and going to using the suggested parts in your git repo. Just trying to make it more accurate and precise consistently so other betas can print and use one easily.

I know it is a bit late to make suggestions, but here is an idea for you guys:

In my work, I have done a bit of research on PCB trace inductors. From your description it seems like you have a coil of 4mmx2mm side cross section with 0.1mm thick wire plus winding and compression errors. Without the errors this makes about 800 turns, and in reality it could be a bit less.

800 turns passing 100mA is 80 ampere turns within your 11mmx2mm space. I will use this as my target field strength.

If you go with standard spec PCBs, track width is 8mil minimum, or 0.15mm, and can carry roughly 0.45A at this minimum trace width. The spacing minimum is also 0.15, making each wire 0.3mm.

Assuming you get an 8 layer PCB, 2mm thick:

4mm/0.3mm= 14 wires per layer x8layers = 112 turns x0.45A per turn = 50.4 Ampere turns within the same space.

This is a comparable value to what you have, the only catch would be you might need an amplifier to match the impedance (but that can be as simple as a couple of transistors or Op amp).


Personally, I like the idea of using a PCB because it offers a reliable source for custom inductors and transformers which is not dependent on your parts being available to order. It also ensures a pretty reproducible, rugged and reliable result (no flimsy wires). In your case, I remember you guys compaining about a coil manufacturer being unreliable and having to wind your own inductors, so I could really see it working for you.




- UD

Quote:

---

Originally Posted by User_Defined

I know it is a bit late to make suggestions, but here is an idea for you guys:

In my work, I have done a bit of research on PCB trace inductors. From your description it seems like you have a coil of 4mmx2mm side cross section with 0.1mm thick wire plus winding and compression errors. Without the errors this makes about 800 turns, and in reality it could be a bit less.

800 turns passing 100mA is 80 ampere turns within your 11mmx2mm space. I will use this as my target field strength.

If you go with standard spec PCBs, track width is 8mil minimum, or 0.15mm, and can carry roughly 0.45A at this minimum trace width. The spacing minimum is also 0.15, making each wire 0.3mm.

Assuming you get an 8 layer PCB, 2mm thick:

4mm/0.3mm= 14 wires per layer x8layers = 112 turns x0.45A per turn = 50.4 Ampere turns within the same space.

This is a comparable value to what you have, the only catch would be you might need an amplifier to match the impedance (but that can be as simple as a couple of transistors or Op amp).


Personally, I like the idea of using a PCB because it offers a reliable source for custom inductors and transformers which is not dependent on your parts being available to order. It also ensures a pretty reproducible, rugged and reliable result (no flimsy wires). In your case, I remember you guys compaining about a coil manufacturer being unreliable and having to wind your own inductors, so I could really see it working for you.




- UD

---

That is a nice idea. Removing a fiddly manufacturing process.

I may be way off base here but, by using the PCB idea, if it turns out that the peachy can be used to etch photosensitive PCBs people would easily be able to make their own coils for experimenting with when they get their Peachys

Without a multi-layer board, you cant really get a large number of turns. So to with a single sided board it would be more or less impossible to get the ampere turns ratio without a massive current. The current limit for USB is about 1 amp, so we couldn't get the required field strength using a USB power supply.

However, doing PCBs with Peachy remains an exciting possibility. The beta testers have not yet bottomed out the possibility of using resin as resist, obviating the need for UV board.

Quote:

---

Originally Posted by Aztecphoenix

I may be way off base here but, by using the PCB idea, if it turns out that the peachy can be used to etch photosensitive PCBs people would easily be able to make their own coils for experimenting with when they get their Peachys

---

It can be done by layering those very thin PCBs, insulating them and then lining them up to solder. The process would be way more tedious than simply winding your own home brew coils and connecting them to a breakout board with epoxy to mimic the peachy PCB coils.

My suggestion is more for the peachy team as a means of mass producing coils with minimal headaches.

But honestly, I believe in diy up to a certain point. The peachy team could easily sell these coil boards to DIY people and save them tons of time, or people can organize group purchases of them from PCB manufacturers (we arent exactly living in the stone age here :p). Also, if you're buying your wire to wind the coil, you may as well just buy a coil embedded PCB. As makers, we need to be realistic about what is easily done for us vs absolutely doing every bit of work yourself. No man (or woman) is an island, so let's not get carried away!




-UD

That's a brilliant idea. 0.45A is a problem, but maybe things could be re-worked to not need as much current.

Quote:

---

Originally Posted by ScottWalde

That's a brilliant idea. 0.45A is a problem, but maybe things could be re-worked to not need as much current.

---

Well, you can always use a 16 layer PCB, but then it starts becoming inaccessible! Realistically, two 4 layers stacked on top of eachother to make the 8 would be good. Or likewise you can stack two 8's to get twice the turns and half your current.

Worst possible case is that you run a buck converter from the 5v to get 1.5v or some voltage ideally matched to produce 0.45A. Then you can run your current amplifier from that 1.5v.

This would ensure a solid regulated Vcc for the coil drive, and it may even help with your ground loop problem on the speaker wire! (also, arent ground loops fun?!).

There is a nice little chip I use for work:

http://www.digikey.ca/product-detail...3-1-ND/4147125

Inexpensive, simple to use, no real problems with it. (just follow their example). It will add more to your board space and BOM, but at least it is a possibility that accounting can have the *option* to shoot down!

There are also tons of chips like this, so dont be shy to shop around!


- UD

I just finished watching the assembly video for the Peachy again and from the looks of it, using a PCB coil instead of the wire wrap would make assembly easier and would be more conducive to the "snap fit" assembly.

How goes the progress on the PCB coils?

I just stumbled upon this:

http://www.digikey.ca/product-detail...2265-ND/703454

Digi-Key Part Number: 240-2265-ND

It`s a 20mm square ferrite plate. Id imagine if you stuck it to the back of one of the coils, it might better concentrate the magnetic flux and increase the coil`s influence on the magnet. Bringing the magnet closer to the coil would further help coupling.


- UD

Quote:

---

Originally Posted by User_Defined

How goes the progress on the PCB coils?

I just stumbled upon this:

http://www.digikey.ca/product-detail...2265-ND/703454

Digi-Key Part Number: 240-2265-ND

It`s a 20mm square ferrite plate. Id imagine if you stuck it to the back of one of the coils, it might better concentrate the magnetic flux and increase the coil`s influence on the magnet. Bringing the magnet closer to the coil would further help coupling.


- UD

---

No progress on this yet, were still working on solving the problems that cause holes first.
The ferrite plate is a great idea, so long as it dosnt become slightly magnetized, could cause historysis, or memory in the printer.

Hey guys,

I just wanted to let you know that we posted an update about Winder on KS and IGG today. You can check it out here.