Linear quadrature encoders

[eeinpa]

I was recently looking for a cost effective linear quadrature encoder for measuring position of a press with a PLC. Basically I wanted something resembling a pneumatic cylinder, relatively rugged, with a straightforward quadrature interface. Absolute encoding would be a plus... Stroke needed is 14 inches.

I was surprised how hard it was to find something clean. Inductosyn devices seemed like a nice option, but it seemed their output signal options were bizarre; translators would be needed to get to useful signals, plus their mounting is sort of awkward. There were some options that had a physical sliding connection on the SIDE of the cylinder... very awkward mounting! String pot style encoders were another option, but the string seemed somewhat delicate and how do you zero them? Linear pots require analog inputs, and noise could be a problem.

I ended up with linear digital quadrature encoders from an Italian company Elap (

http://www.elap.it/)

which are available through a US distributor (

http://www.ghbinroth.com/Elap/Elap_Linear_Incremental_Transducers).

They are < $400 for 14 inch stroke devices. They aren't quite as rugged as I would like, but they should last for a while. One nice feature is that they have an optional index pulse at midstroke. By simply cycling my press through mid-stroke I have in essence an absolute encoder, with no limit switch BS required!

I'm just curious if anyone knows of similar devices I missed. Plus it would be nice to have an alternate part available. The month long August holiday in Italy was a bit of a pain (since I wasn't included!).

Thanks.

 

[Skogsgurra]

It depends somewhat on what accuracy you need. But, since you include string/pot as a possible option, I think that your needs are moderate.

For that, I have used Tempsonic transducers.

http://www.mtssensors.com/index.php?id=130

They are absolute and rugged. Easy to install and easy to interface. Analogue output is the simplest. There are bus versions also. Not too expensive.

I used four of them with 1000 mm stroke to measure position of press tables in a press where temperature was an issue. Worked very well.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Hi Gunnar. I didn't see any obvious string sensors on that link. Were there any? Are those enclosed rail looking things string sensor based?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[Skogsgurra]

No. They work on the mechanical TDR principle with a short pulse being sent into the wire and the reflection occurs where the ring magnet is. Time for the pulse's round trip is measured.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[itsmoked]

Ah yes. I've messed with those before. That's why I was thinking, "Those don't look like string encoders!".

Keith Cress
kcress -

http://www.flaminsystems.com

 

[eeinpa]

Yes, the MTS parts look familiar. I believe I looked at them or something similar. As I said, their output signal choices seem odd to me; I guess they are geared towards OEM use with machine networks. Quadrature encoder (which is supported by virtually every PLC on earth) is apparently not supported.

Their mounting also seems awkward for a retrofit... a ridgid shaft with a ring around it? I see they do offer one unit enclosed into a package similar to a linear pot (resembling a small air cylinder). But that unit has very limited output signal options. I can go analog (current would offer reasonable noise resistance) but would need a 15 bit A/D to match the resolution of the Elap digital sensor. You won't get that on an entry level PLC.

 

[Skogsgurra]

15 bit resolution hardly matches a string based sensor. Accuracy of those sensors seldom reach better than .1 percent. So, a 10 bit A/D would suffice. And that, you get even in micro PLCs like the Siemens LOGO!. Do you need better than that?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[eeinpa]

I did not say I used a string-pot sensor. The Elap sensor I use resolves 0.002" (0.5mm) over 14 inches. That's 7000 counts. It's in a neat package which easily two-point mounts, too. It has a simple quadrature encoder interface. I asked if anyone could suggest something similar. Thank you.

 

[Skogsgurra]

Right. The encoder as such has that resolution. But the string and the mechanics does usually not correspond to the resolution. That is why I asked what accuracy you really need.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[amptramp]

MTS and Balluff make LDT's - Linear Displacement Transducers (not LVDT - different technology).

http://www.mtssensors.com/products/linear-position-sensors/index.html

http://www.balluff.com/Balluff/us/ProductsChannel/Overview/en/Micropulse+Transducers.htm

In the early days (1980's), the output signal was a design unique to the technology of LDT's - Magnetostriction. However, today standard output signals are available.

MTS invented the technology, but Balluff also has a very nice product line with different profiles and standard analog outputs such as 0-10V and 4-20mA.

They are easy to apply and accurate enough for most applications.

They are also rugged. I have used them in Plywood and Brick plants.

I love to use them whenever I have a linear application.

 

[amptramp]

I recant my comment on the inventor of the magnetostriction technology. MTS acquired Temposonics, Inc in 1984. I believe Temposonics, Inc. had the early patents on the LDT. I know when I was young , Temposonic meant LDT like Kleenex means tissue.

FYI, this link describes the technology:

http://www.mtssensors.com/fileadmin/media/pdfs/551019.pdf

 

[CcanDo]

What range is the rate of travel for the 14" cylinder? Resolution of 7,000 may suggest a high speed quadrature counter. Digital from a rotory encoder coupled to rack and pinion seems to be one option. The "shelf item" encoder is sometimes fragile and/or over priced,IMHO...Also,"home" may be CL, updated via laser.