Linear Encoder For Rotating Wire 1

[RMG143]

For a current project I am looking for a way to track thin wire linearly as it is unwound from a spool, ground in a grinding machine and then rewound onto another spool on the opposite side of the grinder. The kicker is that due to the nature of the grinder, both the spools are rotating in a single direction and therefore so is the wire. The wire is rotating about its own axis and does not vary due to a number of bushings it is strung through.

(spool)--wire--[grinder]--wire--(spool)

So put simply I am looking for a sensor or system of sensors which could act as a linear encoder for a wire which is rotating about its own axis.

Any help would be appreciated.

Thanks all

 

[ivymike]

how about running it over a wheel and measuring revolutions of the wheel? (slippage is possible)

could you magnetize segments of the the wire as it passes, then measure the movement of the magnetized portions?

 

[dhengr]

Btwn. the first spool and the grinder, I’d run the wire btwn. two rubber rimmed (or lightly knurled?) pinch wheels and count both wheel revs. and average them. You would want to be able to change the rubber rims due to ware caused by the rotating wire. Mount this counter to the support system for the first spool, so it rotates with that spool, thus very little relative rotation btwn. the counter wheels and the wire, re: rim ware. Maybe inboard of some sort of level wind follower. I assume the real wire rotation is happening relative to the grinder and not the spools and that the spools rotate together.

 

[IRstuff]

I don't get quite what you mean by "track linearly." Do you mean to trace its entire path from one reel to the other?

In which case, the only plausible off-the-shelf answer would be a machine vision camera. With backlighting, you could tweak the contrast to essentially give you a black wire on white background, and with software processing, you can verify that the wire is staying on its path.

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[tygerdawg]

What info do you need from the wire? Linear speed, and variation thereof? Linear amount over time? And what is wire material? Diameter? Surface condition? Material condition?

I would probably take a stab at rigging two hardened tool steel wheels face to face with wire through, like O|O . One wheel might have a central groove ground into it to help track the wire as it flows through. Build the assembly with springs to keep constant clamp force on the wire as it passes through the wheels. Then mount an encoder on one of the wheels.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[jedward]

run the wire around the intermediate spool that is attached to the revolution counter
-use slack tensioner
-no slippage
-use angle encoder and spool diameter

 

[RMG143]

ivymike:
Slippage using a wheel would produce too much variance for wire grinding which needs to be fairly exact. The magnetized idea is interesting though. I will have to look into it.

dhengr:
This is actually very similar to our current setup. Problems arise due to greasy wire slipping on the wheels. Various wheel materials are being tested.

IRstuff:
I mean the length of wire coming off of the first spool. This figure controls how the grinding machine creates profiles on the wire.

tygerdawg:
The length of wire coming off of the initial spool is the information I need. This translates to the grinding machine and tells it how it should be grinding at any particular time. If the encoder says we have dispersed 3 feet of wire, the spool is X from the grinding center => the grinding machine needs to be grinding the wire profile in accordance with 3 feet + X. Wire material varies but is typically a SS variant but currently is nitinol. Surface condition is stock wire with no secondary finishes or polishing. The wire is quite smooth and small amounts of great from the supplier add to the chance of wire slippage.

jedward:
If I am understanding you correctly it would seem that there would be a problem using the spool diameter in any calculation as the diameter of the spool is constantly changing in no predictable way. Wire spools have wire that is not perfectly laid layer to layer, side to side.

 

[tygerdawg]

I think what jedward has in mind is something like a capstan arrangement with multiple wraps of wire around it...seems like that would work also.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[asimpson]

Could a vision system do it by tracking details on surface of wire?

This is how optical mouse works.

Agilent may have something

 

[asimpson]

http://www.paperage.com/issues/nov_dec2003/11_2003laser.pdf