Control.com - Nerds in Control

Hi everyone,

This is the first time I need to ask help on a forum. I'm not very proud that I can't fix this problem on my own.

The problem is that I need to find a sensor with the following capabilities and I can't find a good solution on the internet. That's why I hope I can get some help here.

The requirements are: (I'm from Belgium so everything is in millimeters)

- A linear measurement over 1000mm wide
- Accuracy of 0.001mm
- Good repeatability
- Stable, can handle some dust
- Local readout + send data to a Siemens S7-300 PLC

Does anyone know of a sensor that meets most of these requirements?
Thanks in advance.

Everyone needs help sometime.

The most accurate instrument I know is a Magnetorestrictive sensor with a magnet that slides along a stainless tube. They are very accurate and quite rugged. To get the best accuracy you should use one with a digital communication rather than a 4-20 mA. There are dozens of manufacturers, just search for Magnetorestrictive Sensor.

http://www.sensorsportal.com/HTML/DIGEST/may_06/Positio n_Sensor.htm

Hope this helps you.

Regards,
Roy

Two common spellings (you have to know both to search). I believe the absence of the "re" in the middle is the formal form, but both seem common.

magnetostrictive or
magnetorestrictive

Carl

For that accuracy over that distance, I would suggest looking at a Heidenhain linear encoder. They are very common on machine tools.

http://www.heidenhain.de

I believe you can interface them with a Siemens encoder input card if you use the correct interface box (the interface boxes are sold separately from the encoders).

You will need to talk a Heidenhain distributor as to what parts you need, as figuring it out from their catalog information isn't very easy.

The easy thing to use is the magnetostrictive sensor as mentioned above. You can get them with 1 micron resolution but they will not be accurate to within 1 micron.

I would NOT use analog!!!! A 16 bit analog to digital converter will not provide the desired resolution. Chances you have a count or two of noise anyway. I would use SSI as you require a range of about 1000000 counts.

The easy path is http://www.balluff.com or
http://www.mtssensors.com/products/index.html

If you really need the accuracy then look at glass scales but you need an SSI interface or encoder interface with a vary accurate homing routine.

>A linear measurement over 1000mm wide
>Accuracy of 0.001mm

That kind of length with that kind of accuracy is a tall order.

Heidenhain can hit one micrometer, or rather +/- 0.5 um only at 270mm. The can do +/- 1 um at over a meter. That's an exposed linear encoder.

Sick Stegmann claims 1um at up to 16 meters(!) with their "Lincoder".

MTS Sensors can hit 1 um at up to 10 meters.

Others that have a solution that might apply, in no particular order: Renishaw, ASM Sensors, Balluf, Sony Manufacturing Systems Corporation, MicroE Systems, RSF Electronics.

There are probably many more.

>Good repeatability

Shouldn't be a problem with that level of accuracy, but you'll probably want to talk to the vendors.

>Stable, can handle some dust

I'd lean away from glass linear encoders, then.

>Local readout + send data to a Siemens S7-300 PLC

I think this won't be much of a problem.

-James Ingraham
Sage Automation, Inc.

Hello, The heidenhain LIP 500 Exposed Linear Encoder offers between 70 and 1440mm of range, at an accuracy grade of +/- 1 Micron. Feel free to send a request for quote with your contact info to

steven.rushing@cdiweb.com
Thank you kindly. Steve

For such environment and resolution maybe suggested cheaper (vs. optical scale) magnetic linear encoder - for example, from Renishaw:
http://www.renishaw.com/en/9802.aspx

And standard AquadB output may be connected to Encoder unit of Siemens S7-300.