OT: Load cell question

Last week we had to replace the head on a 40,000 pound scale. Used 2,000 and 1,000 pound weights. Calibration was a breeze, basically just a handful of button pushing steps (ain't microprocessors great?)

I noticed afterward that if I step on the scale the indication goes right up to my weight and goes right down to zero when I step off. When the forklift we used to place the weights goes on, the indication again goes right up to ~11,000 lbs. However, when the lift goes off the indication goes to -6 to -8 pounds, then slowly recovers to zero.

This deck is 20+ years old. Do load cells get 'tired'?

No what is probably happening is that the load cell bolts are tight but not tight enough for the weight that you are applying You will really have to snipe the **** out of them. A rule of thumb with working on scales is if it does not fluctuate at rest load cell is good. If it changes zero point from adding product to removing product its a mechanical based problem. One other thing to note remove any form of rust from the load cell mounting this will throw off the linearity of the weights as pressure is applied. Also make sure the scale is 4 corner balanced you should have 4 load cells that interconnect to a load cell communication board you may have to use the trim pots to balance the loads corner to corner. Hopefully all your load cells have the same internal resisntance. (do not cut back the cables) this is included in the resistance totals when they are manufacured. Cutting them back can throw this off. You can use a lighter weight for the balancing. I would use 200 kg place on one corner calibrate scale move to next corner use trim pot to bring to same value. Keep in minds once you change one trim pot it will affect the other corners on the averaging.. The balancing is probably the trickiest part of scale repairs and takes a considerable amount of patients and practice.

Short answer: yes.

Its a little more complex than that. Load cells are strain gauges bonded to beam element which is either placed in tension or compression by the load depending on the load cell design. As the beam element elastically deforms under load the strain changes. As long as the load remains under the elastic limit the element behaves like a spring. Any beam element can eventually yield (bend) either from being overloaded beyond its elastic limit or from fatigue from repetitive cycling. Additionally the strain gauge can become delaminated from the element over time.

Other factors can be in play as well, such as wear or stickiness in any scale deck guides, incorrect pre-load on bolts, mis-alignment putting a side load on one or more load cells, poor quality connections, etc. If you recently replaced parts on the scale I would look at the work that was done first.

Alaric mentioned something I have forgotten about is in essence with his explanation one of the strain gauges may develop in accuracies on the high weight side Although this is very rare I've seen it happen 2 times in 15 years.

One way to test for what Alaric mentioned is to disconnect all but one load cell take a look at your read out apply full scale weight. It should come back to original setting once done if so continue to the next one do the same test. By hooking up one by one and testing in between it will isolate which load cell is either bad, loose or binding

Another quick and dirty trick is what we call a hammer test. Basically you get a 20 lb piece of steel and stand at each corner and bonce the the steel up and down. If one of the load cells is bad then it will show as a slow responce back to zero.

Another thing to watch out for is rubbing. This is where the platform is resting against another object. IE if the scale is mouted in the floor it could be touching the wall of the well.

As Alaric explained your bending a piece of steel. It is the the amount of bend in a certain direction that decides what the responce is. THis is where age does have an effect. It is usually in the form of corsion.

As far as the trim pots go. If you play with those you better pack your lunch. You can easily make a bad situation worse. It is very rare you should ever have to mess with those.

As far as the trim pots go. If you play with those you better pack your lunch. You can easily make a bad situation worse. It is very rare you should ever have to mess with those.

Provided your load cells are well balanced one to the other. Their is a Rice Lake communication board Rice Lake JB4SS the advantage this board has over others is that they includea jumper that disables the trim pots these work extremely well for 4 load cell applications where all loadcells are naturally balanced.(be careful here some trim pots trim excitiation other trim the signal lines. The JB4SS trim Box is also a good purchase particularly in high wash down applications as it includes a pressure diferential breather vent to compensate for pressure differences created by clean up water. By the way when I referred to sniping earlier I typically use a 6 foot snipe on my wrench to tighten the truck scale we have its the only way I found to tighten those bolts enough to handle the truck weight

Here is a useful load cell troubleshooting guide

I didn't expect so many responses. Thanks!

I quess what I'll do is pull the covers, it's a drive on, floor mounted type, and check out the mountings and load cells. I did clean out a whole box of pallet splinters, nails, etc. from between one end and the ramp (no dead rodents tho').

As far as balance pots, no, I don't want to go there. I will check the corners though to see if that is what's needed.

Clay B - I've never heard of the 'hammer test'. If I can put my hands on a suitable piece I might give that a try just to see what happens. Shouldn't be a problem, there's usually a broken hunk of something lying around.

Again, thanks for all the responses.

mordred - thanks for the link.

The hammer test is as I said a quick and dirty test. I use it all the time at customer's sites because it lets me know real fast if I have an issue in one of the corners. I am not sure of the exact math but basically what you end up doing is an instantous weight 2 to 3 times the weight of the object your bouncing off the scale. Just don't swing it like an axe, your just trying to get a large movement quick.

Alot of times when we install new equipment the customer compares it to their exisitng scale and when the weights do not match up they say the new equipment must be defective. So I will alot of times go and check what our system is being compared to. You would be suprised what I find.

Clays hammer test is similar to my jumping up and down on each corner test LOL I think I may switch to Clay's method too hard on the shins my way lol

Excellent avatar Doug.



Next time someone asks about a flip flop we'll just point them to your profile.

Alaric said:

Excellent avatar Doug.

Click to expand...

Thank you.

Alaric said:

Next time someone asks about a flip flop we'll just point them to your profile.

Click to expand...

Really, the main reason I made it (and what a process that was) was so it would appear on the site and maybe forestall some of those recurring questions. Notice I also included a one shot, since that seems to come up every so often too.

Revealingly, shortly after I put it up, with Mr. Melore's assistance, I had a PM asking me to explain the theory of operation. I referred the sender to the help section of this site.