Recording Weight Values on Calibration Certificates

Hershal

Metrologist-Auditor
Trusted Information Resource
The best is to record the weight to one place beyond the balance it is used on, and in the next column record the uncertainty.

Gage blocks can have the same template.

You need to use gloves when handing either the weights of blocks, and do NOT put sitckers on the weights.

Why?

First, finger oils that will change the values, and secondly there is a thermal transfer that must be accounted for in addition to the difference between calibrate and ambient temperature.

Handbook 44 is valid for its intended use, but the finest points of measurement science are not its intended use.

Hope this helps.
 
A

Automatic

*I have a very good understanding of principles of weighing, however specific regulatory codes is not my expertise, so please keep this in mind. *
Generating high quality data and regulatory compliance often overlaps, but they're mutually exclusive. It would be more useful if you can state what kind of application and specific needs that must be met.
The requirements for academic and corporate R&D laboratories that use analytical balances in pursuit of scientific inquiries is vastly different from forensic laboratory. In scientific research, you would explain why things were done a certain ways and why this method would improve the quality of data.
In forensics laboratory working on drug evidence, environmental liability scandal and in other places where the primary concern is government regulation compliance and having courtroom tough lab results, complete traceability all the way to the primary standard without a gap is a greater importance as the results are expected to be scrutinized by any possible reasonable doubts by a panel of expert attorneys looking to dismiss unfavorable evidence.

For example a class 1 weight of 200g the certificate would add decimal places to show it meets a certain class tolerance. Would it be proper to list it as a 200g weight or a 200.002375 weight. I'm trying to understand why some vendors list their weights using these 2 formats and which one is correct.
SECOND POST:
Well the reason I asked is we had a 200g balance calibrated. The weight was listed as 200.0004260 the balance read 199.9997. My supervisor stated that this balance was out of tolerance because 0.00072 was over the stated
This is fairly common misunderstanding which leads to user error and resulting data quality degradation. A 0.3mg skew on one time check with an audit is perfectly fine for many analytical balances. It isn't ok in a perfect world, but they're not a perfect device.
I wasted time trying to explain something, because I thought your mass is off by 2.4mg which would be beyond unacceptable for ANSI Class 1. (your profile says India, but you didn't mention which class was used, so I assumed, for analytical, ANSI/ASTM class 1(US customary standards) or OIML E2 is used), then in your later post you said the deviation is +42.6?g, a deviation less than what a normal analytical balance can reliably discrimnate. This is within specs of ASTM class 1, but would be out of specs of E2.
As a general good practice, I suggest always listing:
Balance make, model, type and serial number. Other identifying information i.e. biochem lab rm 23, asset tag 6969. Technician name and signature. Room temperature. Any unusual conditions noted by tech (i.e. print key broken)
Weight Class (OIML F1, F2, etc) and its serial number(so its record or the actual piece can be checked out if needed later), room temperature, technician name, date/time.
When you note the class, it can be assumed that uncertainty of the test weight is within what is permitted for that class and the idea is that the accuracy if reference is more certain than the instrument.
 
A

Automatic

The advances in electronic weighing technology is making it possible balances to measure masses very accurately.
In a short answer, yes and no.

It's more about sensitivity (resolution) and convenience. Electronic scales require much less user skills and the cost of production is dramatically lower.

A high resolution (million d + ) scales require constant calibration adjustments(which is automatic on many newer models) and suffer from linearity issues. These are issues arising from imperfections in technology/practical product limitations.
 
K

KeithM

Thanks for all the replies everyone. Since the original post we've made some changes to our cal program and now perform all of our Balance calibrations in-house. We made the following changes to our procedure so even though we do not do an uncertainty calculation the equipment owners are happy and satisfied.

We have a rounding procedure in place.
We record the weight values as stated on the weight cal certificate for conventional mass.
We have process tolerances usually well outside of the cal tolerance.
Though there are a couple of balances with tolerances that are too tight (we will eventually deal with change on this) our process tolerances allow some room for error.

Again thank you for all of the responses!
 
Top Bottom