What is the least expensive distance measurement device? 1

[ERT]

Hi all,

I am designing a machine that has a pneumatic cylinder and ram system. I want to measure the cylinder stroke with a system that is relatively inexpensive ($200~$300 would be great).

The stroke is 30 inches, and it cycles only about once a day. It moves very slowly in extend, and retracts in about 60 seconds. I would like accuracy to be within ±1/8 inch, although worst could be considered if cost saving warrants.

I know there are linear encoders and glass scales but all of these offer much better accuracy than I need, and cost more than this machine model has budgeted for this option. LVDTs are of interest, but I thought there might be better ideas than this.

Packaging is also a consideration, a typical LVDT arrangement makes sense, although I would likely have to add spherical joints to both ends to avoid binding.

Servo pneumatics really don't serve my application. I also don't think direct cylinder piston sensing will work for me.

Any ideas would be appreciated,

ERT

 

[MikeHalloran]

Consider a cable pot:

http://www.midoriamerica.com/r_g_conductive_cable.html

You will of course also need a small power supply to excite it, and an ADC to read the position.



Mike Halloran
Pembroke Pines, FL, USA

 

[patprimmer]

Would you be able to set up mechanical limit switches. They can be real cheap if you just need to stop at the same place each time.

Regards

eng-tips, by professional engineers for professional engineers
Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

 

[ERT]

Mike - thanks for this, I had heard of a linear pot, but the cable pot is a new idea to me.

Pat - thanks as well, but I do need a measurement system along the entire stroke. Within the discrete realm, I had thought of using several proximity sensors or photoeyes on flags staggered at the ±1/8 resolution, but this would be messy and require homing.

Any other thoughts are appreciated.

ERT

 

[GregLocock]

You haven't defined what lifetime you need.

Do you need the same accuracy in both directions?



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[ERT]

GregL,

Yes good questions, say 5 year life. Which would mean a life criteria of only 1000 strokes.

And yes, I only need the accuracy on the slow stroke. I have had second thoughts of how this device could be arranged - I think the slow stroke when the measuring occurs is better suited to the retract direction (regardless of cable pot, or piston type device).

I should also mention this is a heavy industrial atmosphere. Temperature is ok at 15°C to 40°C, but there is a fair bit of dust in the air.

Thanks again for any suggestions.

ERT

 

[GregLocock]

OK, from left field, really cheap, really nasty.

Take a power hacksaw blade (cost 30 dollars), a hardened steel pin ($1) and a cheap microphone from RadioShack ($10)

Arrange the blade so that it clicks over the pin

Use the mike to count the clicks.

You'll need some signal conditioning, say $5






Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[MikeHalloran]

I haven't seen a hacksaw blade that long, but a broken bandsaw blade would work.

Given the low speed, I might use a switch to turn it into an incremental encoder. You need a second switch to sense that the ram is at one position, say one end. Both switches have to be debounced, but you could do it in software with a PIC or something similar.

The only disadvantage of an incremental encoder is that it loses track of where you are when there's a power glitch, so you'd want to run the electronics from a battery. A 9V battery can run a PIC for a few days at least.





Mike Halloran
Pembroke Pines, FL, USA

 

[GregLocock]

RS make some nice little optical encoder sensors that wouldn't break the budget, that might be an easier way to get a TTL pulse.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[ERT]

GregL,

Is that RS Electronics, or someone else?

ERT

 

[GregLocock]

Yes, tho I can't find it in their appalling on-line catalogue.

Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[danbarber]

Have you tried Accu-Coder encoders? They are pretty cheap compared to foreign made ones like Kubler.

http://www.ctiautomation.net/Accu-Coder-Encoders.htm

 

[MacGyverS2000]

Ever had the doctor measure your weight on a balance scale? I'm thinking along those terms for simplicity. As the ram extends it pushes a marker along a ruler. When the ram retracts someone could take a measurement of the farthest reach of the ram by looking at the marker. Since it only does this cycle once a day, it would be easy for some floor monkey to walk by once a day just before lunch, write down where the mark is, and reset the mark against the ram again. It doesn't matter that the ram has already reset and is starting to grow again as it will be many hours before it hits the end of its travel again.


Dan - Owner

http://www.Hi-TecDesigns.com

 

[sreid]

For inexpensive, high quality optical encoders, see Avago (originally HP then Agilent then Avago).

http://www.avagotech.com/products/motion_control_solutions/?linksource=homepage

I like the bandsaw blade incremental encoder idea. Use two Hall Effect proximity sensors set at 1/2 the tooth pitch to get A quadrature B signals.

 

[JakeAdkins]

macgyvers2000 beat me to the punch. That would be the easiest way.

 

[ERT]

Thanks to all for your input.

Yes, we have tried the floor money approach, not really practical for us as we have over 100 of these systems.

We are currently trialing the string pot idea. We actually went with a low resolution encoder on a wire drawn unit.

Regards,

ERT

 

[JakeAdkins]

When I was in High School I used to test concrete cylinders for a civil engineering firm. They had a temperature recorder for the concrete curing room. It was a wheel that turned really slow with a pen attached to the thermometer gage. As the wheel turned the temperature was recorded. You would need to build something that would feed paper slowly and have a pen track the movement.

 

[MacGyverS2000]

I remember my father using those polar graphs to record temperature and humidity in greenhouses when I was a young'un... I think we had one stored in the garage up until a handful of years ago.

Dan - Owner

http://www.Hi-TecDesigns.com