We are experiencing a lot of creep with the weight readings from our floor scales and were wondering if anyone has had any experiences or knowledge on the mechanics of load cells and why creep occurs.

We used a Mettler Toledo floor scale (Model 2158) that was rated load 10,000 lbf to weigh a piece of equipment that was approximately 5,000 lbf. I noticed after placing the equipment on the scale the weight crept a lot more than the .03% load/20 minutes specified in the manual ... it was more like .25% load/20 minutes. The unit is probably ~5 to 10 years old and in need of calibration.

A few questions came to mind:

1. What are the causes in creep in load cells?
2. What are some reasons that might be the cause of the large amount of creep we experienced?
3. Which point of the reading is typically more accurate: initial reading or after creep?

Thanks for your help.

D. Lewis, P.E.

Is there any linkage between the platform and the load cells?

when was the last cal ?

Call Mettler

There's a load cell at each corner of a 60" x 60" steel platform. Each load cell has a rocker pin suspension that self aligns and direct forces vertical onto load cell. There's a link to a PDF of the table below if interested.

http://us.mt.com/us/en/home/supportive_content/pro...

Calibration date was in 2000. No expiration date listed. Yeah the more I dig into it the more it sounds like I should just call Mettler. I was hoping there was a simple explanation that others might have encountered.

D. Lewis, P.E.

No expiration date on the calibration? That doesn't seem right.

--Scott
www.wertel.pro

If the "fixed end and side bumpers" have gotten gooified to the load cells they create a parallel load path that is potentially time dependent.

That’s 12.5lbs. w.r.t. 5000lbs. in 20 minutes. Are those four load cells and the entire scale system self compensating for temperature changes over the 20 minute period. Could that be the problem?

MintJulep... I can't see the side bumpers, but that's a good thought. It's not in the cleanest or coolest environment.

dhengr... Now that you mention it, we did actually have to open the bay door (about 5 feet from the scale) to swing the equipment around with the forklift. Being it's ~95F outside today and ~80F inside it might be compensating for the temperature change. Good call. I'll have to investigate.

Thanks,

D. Lewis, P.E.

MintJ:
Now you have really started something.... “gooified?” If the powers-that-be see that term, every damn code and spec. in the world will have an addition seven pages of verbiage and modification factors for “gooification” vs. time, temp., modulus of “gooosity” and time of the month, etc.

I was briefly involved in manufacture of load cells.

In them, you rely on a coat of tar (the best, or least awful, weatherproofing) to prevent degradation of a thin film of superglue (applied carefully after extremely finicky preparation) which is expected to repeatably transfer considerable strain from the beam to the strain gage, which is made from a material that's nonideal in many ways.

In short, it's a wonder the damn things work at all. Expecting them to last a decade is pushing your luck.

Mike Halloran
Pembroke Pines, FL, USA

Ha. Mike, I didn't realize they were so delicate. I did check a few known loads as well (< 200 lbs) and the error was still within the resolution of the scale for those items (2 lbs). So it does work relatively well despite it being so old.

D. Lewis, P.E.