Rotating POV LED clock - ideas for electrical contacts

Projects General Guidance
1

I am trying to figure out how to get electrical power to the rotating part of the clock without having to have a separate battery or the electronics on that rotating part.

It would be an easy matter to make a couple of circular tracks from a copper top etchable circuit board.

But I can think of a way to make the 'brushes' other than a couple of pieces of bent wire. These are far from ideal because they are likely to wear down the copper track fairly quickly or lose contact with it one way or another.

I was thinking of creating some contacts out of some small ball bearings some how as these will roll around the track rather than drag across it.

I could create two sets of circular tacks, one on the rotating part and one on the stationary part and have the ball bearings in between the tracks.

But how could you mount a single ball bearing in a disk of MDF or PVC in a manner similar to the way they mount them in ball bearing races, e.g. used in bicycles.

Anyone got any ideas on this? Or perhaps another means of creating a 'brush'?

2

Several options... Google Images on this string POV slip ring contacts

3

Actually I just had a brilliant idea myself.

I have a cheap pen with a metallic tip here and I just checked with my multimeter that there is good electrical contact between the screw on bit and the ball bearing in the tip of the re-fill. So that would make an ideal 'brush', and it should be an easy matter to mount the pen with a small spring against my circular track.

4

mrburnette:
Several options... Google Images on this string POV slip ring contacts

Ahhhh! It is easy when you know what to search for. I think I like this idea better because it means I could have some data lines as well as power ones. Thankyou.

5

Perhaps you could get a small DC motor and take it apart to get the carbon contact brushes which are often mounted on a bit of phosphor-bronze spring.

You could IR to communicate data to the rotating part without needing data contacts.

...R

6

Robin2:
Perhaps you could get a small DC motor and take it apart to get the carbon contact brushes which are often mounted on a bit of phosphor-bronze spring.

You could IR to communicate data to the rotating part without needing data contacts.

...R

I realized I can't really use the center bore slip rings on ebay.

Assuming I mount the electronics inside a hollow disk, with the LEDs exposed on the surface, I would need the axle on the back of the disk and the slip ring on the front - not a good look.

I think I will have to run some narrow copper tape around the edge of the disk to form two tracks and some spring loaded brushes of some sort on the base for the power, and establish a wireless serial connection between an atmega inside the disk and another atmega in the base to transfer data.

I have already fiddled around with a couple of inexpensive bluetooth modem things with my arduinos and got them working.

7

Sounds like it is getting a bit complicated.
https://code.google.com/p/povglobe/

In this example project, the author just used a mini-headphone jack for the 360 connector. With the proper silicon electronic contact lub on the jack and plug, it should last for a while. Also, a product such as DeoxIT may be useful as it lubricates while preventing oxidation. I use it on all of my high-end projects to coat the IC pins before placing them into the sockets.

Ray

8

mrburnette:
Sounds like it is getting a bit complicated.
Google Code Archive - Long-term storage for Google Code Project Hosting.

In this example project, the author just used a mini-headphone jack for the 360 connector. With the proper silicon electronic contact lub on the jack and plug, it should last for a while. Also, a product such as DeoxIT may be useful as it lubricates while preventing oxidation. I use it on all of my high-end projects to coat the IC pins before placing them into the sockets.

Ray

Yeah I have seen that project Ray.

But I would be very limited in the size of my clock that an audio plug could support, lest it damage and/or wear out the plug

And it wont work that well for a rotating disk anyway.

I was intending to use a 12V rather heavy duty PC type fan (salvaged) as my motor and glue my disk (containing the electronics) to it. I will need some sort of spacer between the disk and the hub of the fan - shouldn't be too hard to figure out.

9

How about a rotary transformer? I'll bet you can make your own with a couple of ferrite toroids. Wind the wire around the outside ring of the toroid, place one on the stationary base and the other just above it on the rotating part.

You can then send power via high frequency AC, a few 10s of kHz, and send data encoded on it.

It will help power transfer if you sand the sides of the toroids off so they'll make a wider, flatter area next to each other. Even better if you get a couple of cup core toroids.

NASA uses this technology, and VCRs (remember them?) use rotary transformers to send signal across a rotating junction.

http://trs-new.jpl.nasa.gov/dspace/bitstream/2014/27360/1/96-1453.pdf

http://coefs.uncc.edu/mnoras/files/2013/03/Transformer-and-Inductor-Design-Handbook_Chapter_19.pdf

10

polymorph:
How about a rotary transformer? I'll bet you can make your own with a couple of ferrite toroids. Wind the wire around the outside ring of the toroid, place one on the stationary base and the other just above it on the rotating part.

You can then send power via high frequency AC, a few 10s of kHz, and send data encoded on it.

It will help power transfer if you sand the sides of the toroids off so they'll make a wider, flatter area next to each other. Even better if you get a couple of cup core toroids.

NASA uses this technology, and VCRs (remember them?) use rotary transformers to send signal across a rotating junction.

http://trs-new.jpl.nasa.gov/dspace/bitstream/2014/27360/1/96-1453.pdf

http://coefs.uncc.edu/mnoras/files/2013/03/Transformer-and-Inductor-Design-Handbook_Chapter_19.pdf

For starters it is a POV clock so the spinning disk will be vertically oriented rather than horizontally.

But I don't understand how to separate coils on two separate toroids would be magnetically coupled anyway.

11

boylesg:
But I don't understand how to separate coils on two separate toroids would be magnetically coupled anyway.

The term "toroids" may be misleading if you are thinking of the more common single toroid with the winding going through the centre.


Imagine a transformer with two "E" cores facing. Consider a winding on each of the "E" cores, so that you could in fact separate the two halves with a sheet of paper - a small air gap is tolerable in a transformer and even desirable as it reduces saturation of the core.


Now imagine that you can rotate one of those E cores whilst having it not quite touching the other, the centre of the "E"s remains (almost) in contact. If the outside of the "E"s is extended fully around the transformer core, then it will always be magnetically coupled as you rotate. If there is a hole through the middle of the "E"s through which a shaft passes, then the overall magnetic core with the windings now concealed inside, is in fact a torus.

12

Paul__B:

boylesg:
But I don't understand how to separate coils on two separate toroids would be magnetically coupled anyway.

The term "toroids" may be misleading if you are thinking of the more common single toroid with the winding going through the centre.


Imagine a transformer with two "E" cores facing. Consider a winding on each of the "E" cores, so that you could in fact separate the two halves with a sheet of paper - a small air gap is tolerable in a transformer and even desirable as it reduces saturation of the core.


Now imagine that you can rotate one of those E cores whilst having it not quite touching the other, the centre of the "E"s remains (almost) in contact. If the outside of the "E"s is extended fully around the transformer core, then it will always be magnetically coupled as you rotate. If there is a hole through the middle of the "E"s through which a shaft passes, then the overall magnetic core with the windings now concealed inside, is in fact a torus.

Oh I get it. It would complicate the circuitry however.

And I was going to use a heavy duty 12V PC style fan as my motor with a thick plastic rod glued to the hub of the fan and screwed to the hollow cylinder that will contain the electronics that drive the LEDs.

So such an axle is not going to fit through the centre of the torus halves that you are referring to.

I was thinking that the hollow cylinder will probably need some support anyway to relieve any pressure on the fan bearings.

I thought I might be able to put some sort of roller bearing on my base and then have the rotating cylinder run on these with two tracks of Al tape around the circumference of the cylinder such that the rollers double as electrical contacts for power.

13

boylesg:
I was thinking that the hollow cylinder will probably need some support anyway to relieve any pressure on the fan bearings.

You most certainly will. And getting it balanced in all axes will be absolutely critical.

boylesg:
I thought I might be able to put some sort of roller bearing on my base and then have the rotating cylinder run on these with two tracks of Al tape around the circumference of the cylinder such that the rollers double as electrical contacts for power.

In reference to previous suggestions and that, putting pieces of tape on things and having contacts run on them is not going to provide a reliable or durable contact.

Unless you are prepared to machine the slip rings, or use ones already assembled contiguous with the machined bearings, I feel you are doomed to disappointment.

14

Paul__B:
In reference to previous suggestions and that, putting pieces of tape on things and having contacts run on them is not going to provide a reliable or durable contact.

Unless you are prepared to machine the slip rings, or use ones already assembled contiguous with the machined bearings, I feel you are doomed to disappointment.

These would be ideal combined with a little conductive grease: http://www.ebay.com.au/itm/151082798502?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649

All I have to do is use some bolts and some sort of upright on the base to mount them and let my clock cylinder sit on them.

If Al tape tracks wear out then I could always use much thicker Al flashing and if that fails then I might have to get some machined rings.

The above bearings should not be too harsh on the Al tape with some conductive grease though.

15

For starters it is a POV clock so the spinning disk will be vertically oriented rather than horizontally.

What difference does that make?

But I don't understand how to separate coils on two separate toroids would be magnetically coupled anyway.

Did you read the links? I don't mean wind the toroids normally. I mean wrap the windings around the outside, so if you slipped the windings off, they'd be donut shaped, too.

Imagine a ferrite core, long and cylindrical. There are two windings, one on each end, separated by a small distance in the middle. Feed 50kHz AC into one coil, 50kHz AC comes out the other winding. With me so far?

Now imagine that ferrite core is short, fat, and has a hole in the middle. Same thing with the windings, AC in one winding still comes out the other. Imagine a steel shaft passes through the hole in the middle.

Now just split the ferrite between the two windings, leaving a narrow gap. Now when you feed 50kHz AC into one winding, you still get 50kHz AC from the other, but at a lower voltage. You can now fix one of those in place, and the other can spin, with the steel shaft supporting it.

Tada! Two toroids with the windings around the outside edge placed right next to each otehr.

Or imagine these:

http://www.dextermag.com/media/catalog/product/cache/1/image/9df78eab33525d08d6e5fb8d27136e95/D/e/Dexter_Magnetics_Pot_Core_16.jpg

With a winding on each half, separated by a short distance, with a steel shaft through the hole supporting one side on bearings.

Here is the entire book from one of the previous links, on Transformer and Inductor Design. A good read.
http://coefs.uncc.edu/mnoras/tr_design/

16

They are also called pot cores.

It helps if you don't have the center in direct contact with metal. A simple sleeve spacer would do.

http://www.alltronics.com/cgi-bin/category/74

17

polymorph:

For starters it is a POV clock so the spinning disk will be vertically oriented rather than horizontally.

What difference does that make?

But I don't understand how to separate coils on two separate toroids would be magnetically coupled anyway.

Did you read the links? I don't mean wind the toroids normally. I mean wrap the windings around the outside, so if you slipped the windings off, they'd be donut shaped, too.

Imagine a ferrite core, long and cylindrical. There are two windings, one on each end, separated by a small distance in the middle. Feed 50kHz AC into one coil, 50kHz AC comes out the other winding. With me so far?

Now imagine that ferrite core is short, fat, and has a hole in the middle. Same thing with the windings, AC in one winding still comes out the other. Imagine a steel shaft passes through the hole in the middle.

Now just split the ferrite between the two windings, leaving a narrow gap. Now when you feed 50kHz AC into one winding, you still get 50kHz AC from the other, but at a lower voltage. You can now fix one of those in place, and the other can spin, with the steel shaft supporting it.

Tada! Two toroids with the windings around the outside edge placed right next to each otehr.

Or imagine these:

http://www.dextermag.com/media/catalog/product/cache/1/image/9df78eab33525d08d6e5fb8d27136e95/D/e/Dexter_Magnetics_Pot_Core_16.jpg

With a winding on each half, separated by a short distance, with a steel shaft through the hole supporting one side on bearings.

Here is the entire book from one of the previous links, on Transformer and Inductor Design. A good read.
Transformer Design Handbook | Maciej Noras

Yeah I have got the general idea Steve.

18

I mentioned VCR heads as using rotary transformers. Why not just use a VCR head?

http://us.cactii.net/~bb/propclock/

19

polymorph:
I mentioned VCR heads as using rotary transformers. Why not just use a VCR head?

VCR Drum Propeller Clock

OK well salvaging some VCR heads would be a great deal easier than building a rotary transformer from scratch, so I am open to the idea.

However the author states that he can get 120mA at 5V.

What if I was to use, say, 12 RGB LEDs.

That would require up to 12 x 3 x 0.02 = 720mA max if all LEDs are on at once.

Depends if the author is stating whether that is the max you can get from VCR heads or is that is the max his display draws.

20

Will all LEDs be on at once? A big, wide, rotating glowy thing?

Even if your clock were displaying all 8s, all the LEDs are not on all the time.