5K Ohm Limit Switch

I am looking for a potentiometer "limit" switch if there is such an animal. I have an application where a standard true/false limit switch with a lever arm and roller would work but i would like to get a value back more precice than 1 or 0. did i confuse everyone? to be more specific i would like to monitor the size of a "bed roll" on a paper winding machine to determine if it grows. this would tell me if the roll had a wrap on it allowing the machine to shut down automatically. A good old on/off limit may give me false readings due to vibrations and scraps giong through the machine.

thanks in advance

I beliveve the Allen Bradley Devicenet 802T limit switch can give you an analog type signal over Devicenet.....provided you have devicenet capabilities.

http://www.ab.com/sensors/products/limit_switches/devicenet802t.html

I don't suppose you're running a DeviceNet network on the machine ? A-B's venerable 802T limit switch has a DeviceNet version that gives you a feedback value relative to the angle of rotation of the limit switch operator. It's not terribly precise (0-255 over the arc of the operating lever) but it's a digital signal and will be stable and repeatable.

[doh- simulpost !]

Some ultrasonic sensors give you analog output options, but I'd never seen a limit switch do it until I got my hands on the A-B DeviceNet version.

hehe....Ken

We have a machine that uses a precision potentiometer (10K) that is driven by gears. The pot is 10 turn because of the travel involved so its set at zero for the middle of the table. I do not know if the whole unit can be bought pre made but it is kind of simple, the motor drives a screw shaft to move a table forward or reverse, there is a gear driven by the motor too that turns a gear that the pot is mounted to. The ratio on the gears you would probably have to develop on your own based on specs but the device setup is simple. I will see if I cant send a picture of it. This is the old way of positioning die cutter etc type tables.

Since paper caliper is usually rather small, I would think it would take many wraps to accurately detect a wrap on the bed roll. Wouldn't conductivity to the metal roll be better for early detection or am I missing something here?

Maybe arcing and ignition of paper dust is a problem.

Are you sure an analog sensor would be any less subject to false readings than a discrete one?

When I had to detect a 'nip roll wrapped' condition, I used discrete sensor with a timer to filter out false trips due to vibration or roll bounce.

If you're trying to detect when a rewound roll has accumulated to a target size, and you know the web speed, you can calculate diameter by measuring the time per revolution of the rewound roll. For that you may be able to use a discrete sensor and a target on the chuck.

Web speed (Inches/second) * Time (seconds) per revolution = Inches per revolution, which is the roll circumference. Divide it by PI to get diameter.

You could use a foot switch type of device. Square D has some pretty nice stuff which are designed to be foot switches but can be used on many other applications... they come in various resistance values... google the web for an analog foot sfitch or joystick... you'll find many limits switch with analog (resistance) values.

PS: Hey Ken what's been happenning lately, have all your friends been redirected from the RLS forum to here...

PS: Hey Pierre if you're talking about me, I've been around here for a long time...

the product we are making is a spunbond plastic which is usally thicker than paper. the bed roll is rubber coated and runs around 1500 meters/min. i found an AB sensor called colorsight 9000 that may work by detecting the color change from normal orange to white when wrapped. has anyone tried the colorsight9000 yet? Also no devicenet, its a siemens s7-300..