PMEL Forum

K Sections => K5/6 - Physical Dimensional => Topic started by: GIJoe007 on 08-02-2010 -- 14:36:03

Title: Weight Scale Accuracies / Tolerances
Post by: GIJoe007 on 08-02-2010 -- 14:36:03
I'm having trouble truly understanding how to apply calibration tolerances to various makes and models of load cell/strain gauge based weight scales.

I found and downloaded the NIST Handbook 44 - Specifications, Tolerances, and Other Technical Requirements for Weighing and Measuring Devices but really that thing is so full of superfluous information I've not been able to really extract what I need out of it.

For example, a Mettler-Toledo Wildcat WS60LVR Floor Scale. . . 60kg span, 0. 02kg resolution. . . but what the heck's the accuracy?  Is it percentage of fs/iv based?  Is it a simple +/-kg value?

I don't know, but I'm really trying to do this "right," instead of just going "Eh. . . I'll calibrate it to a tolerance of +/-0. 04kg and call it a day. "

Any help would be appreciated. . .  :-D
Title: Re: Weight Scale Accuracies / Tolerances
Post by: MIRCS on 08-02-2010 -- 15:37:20
Does it have a class number, like III, IIIL...etc?
Title: Re: Weight Scale Accuracies / Tolerances
Post by: step30044 on 08-02-2010 -- 20:54:44
That one is Class III
Title: Re: Weight Scale Accuracies / Tolerances
Post by: MC Devastator on 08-24-2010 -- 11:22:31
I worked with Mettler Toledo for years and had many of the same hang ups you had.   What I used to do was gnats azz every calibration.   Then, I knew it was good.   Most scales take little or no time to cal, but you alreqady know that.

I got with one of our engineers and he summed it up like this, the indicator and the cells will have a cumulative error percentage.   Depending on the class of scale (Regrettably Hand book 44 is the bible on this. . . . . . . . <shaking head and weeping silently>)

I learned early on with scales that the best thing to do is to do a complete cal, snug in the load cells and get the cal as tight as possible.   Then there is no real argument.   Not much of an answer I am afraid.   But take the listed error for the cells, and the error for the meter.   That will ball park you.