Metal detecting sensor question.

[jkettu]

Sorry for slightly off-topic question, but since all the gurus seem to be here, maybe someone could help me out.

I need to reliably detect a 3/32" steel cable from a distance of up to 2". The cable is threaded inside a thicker and much longer synthetic rope, so an optical system can't be used. It would be highly desirable, but not absolutely necessary, if this could be done inside a steel square tube (10" x 10", wall thickness 1/2") when the cable is running approximately in the middle of it.

What kind of sensor would you recommend for this? Part number and source would be greatly appreciated.

[rich smith]

Maybe Harry Potters wand? Virtually impossible to detect from that distance or inside a steel casing even with highly collimated x-rays.

[DubbleD]

I was going to suggest a round-point steel shovel, but the wand would be better.

[AndyOne]

jkettu,

An actual real-world solution may be to connect the steel cable to the output of a low power transmitter and use a receiver to detect when it's close enough receive the signal at a specific strength.

Andy.

[jeffs555]

Don't know of any packaged sensor. You could run the cable through two coils and measure the magnetic coupling between them. Another possibility is to run the cable through two metal tubes and measure the capacitance.

[jkettu]

Andy & Jeff, thanks for meaningful answers. Unfortunately the cable is too far inside the rope to be connected to anything. Although more work than I was hoping for, toroidal exitation and sensing windings seems to be the way to go.

[microgyros]

I can't get a picture from your description but I can detect live cables through grounded steel conduit at close range with a "live wire detector".
It has a 4069ub positive feedback oscillator with ariel to pick up low frequency radio on the live cable.
I can't be specific but you will surely find circuit designs online based on capacitance change and radio pickup.
mike

[Ron W3FJW]

What's the purpose? Is it a messenger cable and you're looking for a possible break during manufacture?

[AndyOne]

A simple search coil as in a metal detector should also work, this is fairly old tech, no contacts required just proximity.

A.

[earlwb]

Unfortunately looking for a metal cable inside a metal tube is going to be tough.
The only thing I can think of is to use the "Fox and Hound" wire sniffer system.
This is a transmitter that you hook to a wire in the utility wiring room. Then you go to the other end and probe all the wires until you find the signal. Then you know you have the right wire.

Anyway, you may be able to get something like that to work. You can generate a RF frequency on the cable at one end and follow along the outer tube looking for the signal. if the signal disappears after following it along , then you have a break near where the signal was lost.

Some examples:
http://www.newtechindustries.com/newtech/catalog/cable_continuity_tester.htm

http://www.pimall.com/NAIS/e.counter.html

http://www.uoguelph.ca/~antoon/circ/foxhound.html

http://www.pic101.com/foxhound/index.html

http://www.shopwiki.com/search/SUB+ONLY+TONE+GENERATOR+AND+IN+TRACER,+FOX+AND+HOU ND+KIT

[jkettu]

I see I have failed to explain the situation well enough.. sorry about that.

This is going to be an accessory for a full-scale sailplane launching winch.
http://www.jik.fi/sivut/wp-content/photos/1_valmiinatihin.jpg http://www.jik.fi/sivut/wp-content/photos/5_kelataanvaijerisisn.jpg
To prevent the towline from falling outside the airfield it is reeled in close to the winch after release. The problem is that occasionally the winch driver gets distracted and pulls the chute and associated hardware through the fairlead, causing damage that takes hours to fix. I'd like to make a warning device that sounds an alarm when the chute is a couple hundred feet from winch.

So, essentially I'd like to detect a spot on a moving rope. The towline is braided dyneema rope, 1/4" dia, 1.5 miles long. The spot to be detected is 300 feet from free end. The rope might be moving at up to 50 mph. I thought to mark the spot with a length of steel cable because it can easily be threaded inside the dyneema rope and can be long enough to give ample time for the detector to sense it. Obviously, it can't be connected to anything to be used as an antenna - it is only a position marker made of a material dissimilar to the towline. The metal tube I mentioned in the opening post is that boom between fairlead rollers and winch drum. It would be mechanically convinient if the sensor could be installed inside the boom - there is no need to detect anything through the wall of the boom.

I think I'll wind two toroidal coils and run the towline through them. Feed one coil with ac and measure the waveform inducted in the other. Some experimentation will be needed, but I'm confident this should work.

Thanks everyone and sorry for the confusion.

[earlwb]

Well the simplest method is to use a bulge of some sort on the cable so that when the cable reaches the bulge, it trips a paddle switch turning off the winch motor. Or something like that.
One could also put in a small neodymium magnet (maybe not so small) to trigger reed switch or two or three to turn off the winch motor.
A fancier method might be to use a section of the cable painted white and use a photo-sensor to detect the color change, and it trips something to turn off the winch.

[tomp]

Interesting problem. My first thought was to use an inductive proximity sensor, or something similar. They're used all over the manufacturing world, just for purposes like this. They sense when a part on a production line is in position, or when a machine has properly done an operation, etc. The tricky part to this problem is the size of the wire you're trying to detect: 3/32nds isn't much metal to sense, although if you could constrain the rope through a narrow area it should be able to be sensed.

If you want to search for sensors yourself, try companies like Allen-Bradley, Keyence, Omron, Seimens, Danaher, Turck, Pepperl-Fuchs, and Namco Controls. Or better yet, find an automation integrator company (people who build manufacturing equipment) and ask their assistance. I'm sure they've done this before. Maybe you can trade glider time for their assistance!

Alternatively, could you measure how much rope is played out, then indicate when so much has been reeled back in?

Good luck!

Tom

[Ron W3FJW]

An even simpler solution to the one Earl suggested is to get a better winch driver... Or give him a repair bill whenever he screws up. Anyone that can't concentrate on the job at hand shouldn't be doing it. Good thing he's not handling explosives.

[dmccormick001]

Sounds like it should be fairly easy to install an optical sensor that "looks" across the width of the drum or spool on which the cable is wound, and connect it up so that, as soon as the cable has filled the drum enough to block the sensor's view, it cuts the motor off.