EBME Forums

Testing Patient Scales

Posted By: andysang

Testing Patient Scales - 10/10/18 9:30 AM

Hi Guys

Has anyone come across an alternate tester for testing patient scales.

My hospital has 50 odd scales and I have lost the strength to be able to wheel around 250 kgs for my annual testing .
Something like a calibrated strain gauge or ...

Cheers
Posted By: Geoff Hannis

Re: Testing Patient Scales - 10/10/18 6:23 PM


Use plastic containers and water ... using the well known formula that relates volume of water to weight (mass). smile
Posted By: Malcolm

Re: Testing Patient Scales - 11/10/18 1:34 PM

Andy, Sounds like you need an apprentice !!
Posted By: kit

Re: Testing Patient Scales - 11/10/18 4:41 PM

We don't check the calibration of scales, however, what would be wrong with just weighing yourself on a calibrated scale then quickly going round and weighing yourself on the 50 odd scales you have and if the measurement ties up(within a few percent) it passes.
Posted By: Geoff Hannis

Re: Testing Patient Scales - 11/10/18 5:23 PM


See this earlier thread ... and this one. smile
Posted By: Showman

Re: Testing Patient Scales - 19/10/18 3:32 PM

I can see the auditors having a field day with all these suggestions, the only correct way is to have 3 calibrated reference weights and use them. You may break your back but you are covering your backside.
Posted By: Geoff Hannis

Re: Testing Patient Scales - 19/10/18 6:38 PM


Which auditors would that be? think

And how do you calibrate weights? Certify, may be. And what tolerance is allowed?
Posted By: Philip Page

Re: Testing Patient Scales - 27/09/19 4:10 AM

I have a compression jig that can measure up to 500KG - use a car jack to make the compression.

The accuracy can easily be established by using known controlled weights and measuring main scale with these - I currently have accuracies of 1g in 100Kg - not high level, like lab level scales, but OK for human scales.

The whole thing weighs about 40 KG, and when its used correctly - you can measure any digital scales that lock out after a period of time, ie I regularly test Seca 813's to their max weight of 200Kg.

I believe all scales ought to be weight check across 100% of their weight range to ensure there is no mechanical sticking or binding at some point.

Phil
Posted By: Neil Porter

Re: Testing Patient Scales - 01/10/19 9:53 AM

Phil, do you have a calibration certificate for your 'Jig' if not then all data is invalid.

JCI, CBAHI, ISO etc. require that all test equipment must be calibrated and certified annually.
Posted By: Neil Porter

Re: Testing Patient Scales - 01/10/19 9:55 AM

Weights need a calibration certificate on an annual basis the same as test equipment
Posted By: Geoff Hannis

Re: Testing Patient Scales - 01/10/19 11:41 AM


What tolerances do those august (money-making) bodies allow, Neil? think

And what techniques do they use to weigh the weights before issuing any certificate(s)?

Also, where are test weights tested, and how are they transported etc.?

But anyway (and more importantly), what tolerances are permitted clinically for patient weighing scales?

Bearing in mind that, wherever you are in the world (and as mentioned a few time before), 1 litre of water weighs 1 kg. smile

1 litre of pure water at 4 °C weighs 1 kilogram.
Posted By: Neil Porter

Re: Testing Patient Scales - 02/10/19 2:43 AM

Weight & Measures, surely you have heard of them, how many people have been conned in the markets due to 'dodgy scales'

All calibration centres must be certified by the regional government offices.

Tolerances are from the manufactures specifications
Posted By: DaveC in Oz

Re: Testing Patient Scales - 02/10/19 5:47 AM

We used to send weights for certification every 5 years not annually and I still think that was to much. Once they have been certified, unless you can see that a lump has been knocked off I really don't see the point in re-certification.

Any how, at 5 years our auditors seemed quite content.
Posted By: Neil Porter

Re: Testing Patient Scales - 02/10/19 11:40 AM

Dave,
It appears that the auditors have been replaced by accreditation assessors, if they allow more time between calibrations and certification renewals they will not make as much money.
Posted By: Geoff Hannis

Re: Testing Patient Scales - 02/10/19 2:09 PM


When I asked about "what tolerances do they allow", I was referring to the "accreditation" people. frown

My suspicion is that modern patient weighing scales may be more accurate than the techniques employed by the "accreditation" labs.

In my opinion, nonsense such as this should be challenged whenever and wherever it appears. I believe it to be yet another scam based upon very little clinical need (or possibly none at all).

Newtonian physics should suffice! smile

Meanwhile, what about precision balances such as those found in clinical laboratories and pharmacies ... do the "accreditation" gang poke their noses in there as well? Or do they limit their interests to "low hanging fruit"?
Posted By: Geoff Hannis

Re: Testing Patient Scales - 02/10/19 3:43 PM

Originally Posted by Neil Porter

... how many people have been conned in the markets due to 'dodgy scales'


Only once ... at the souk in Jeddah. whistle
Posted By: carendo

Re: Testing Patient Scales - 17/10/19 12:17 PM

..all very well but how to measure exactly 1 litre of water, and also what weight are the vessels holding said water?
Posted By: Geoff Hannis

Re: Testing Patient Scales - 17/10/19 5:38 PM


Welcome to the forum, Andre. smile

I'll leave it to you to work out solutions to those questions for yourself.

Be sure to get back to us with your suggestions.
Posted By: Neoteny

Re: Testing Patient Scales - 18/10/19 5:57 AM

After repairing a set of scales that were out of warranty, and testing them with calibrated weights.

I was 'threatened' with prison by a certain scales supplier.

STILL waiting for the police to knock down my front door............
Posted By: carendo

Re: Testing Patient Scales - 18/10/19 11:42 AM

Thanks Geoff and glad to be here.

A large portion of my work involves weights and measures so was interested to see how you were measuring your water and weighing your vessels as was hoping to learn something, but to respond to your important question.

A clinical set of scales rated at Class III are defined as such in that the measurements taken over time can be used(in conjunction with other factors) to diagnose patients and prescribe drugs, and the tolerances we work to for this class of scale are:

Reading Discrepancy
0-500 Divisions +/- 1 Div
>500-2000 Div +/- 2 Div
>2000 Div +/- 3 Div

As a side note have attached a png showing the tolerances allowed for various classes of weights to help with your Q below...

Originally Posted by Geoff Hannis


What tolerances do those august (money-making) bodies allow, Neil?

But anyway (and more importantly), what tolerances are permitted clinically for patient weighing scales?



.[/size]


Attached picture weights.PNG
Posted By: Geoff Hannis

Re: Testing Patient Scales - 18/10/19 5:34 PM


With reference to the attachment you provide, I have no idea what E2, F1 et al mean. frown

But for purposes of discussion, here are a few more documents (hopefully they are still in current use):-

1) NAWI
2) LACORS
3) NMRO
4) GOV.UK

When it comes to allowable tolerances, I prefer to think about ± percentages. Taking a quick look at a randomly selected patient weighing scale (actually a Seca Model 769); it claimed an "accuracy" of "greater than ± 0.15 %".

In the NAWI document the "in-service allowance" for a 0-200 kg Class III gives an allowance of ± 200 g. I make that to be ± 0.2 % when weighing a 100 kg load (patient).

I take three points from that brief exercise:-

1) The scales have a greater accuracy than the "spec"
2) Modern electronic scales are very accurate
3) 200 g of water relates to 200 ml

Surely any "shade-tree biomed" techniques we use when checking weighing scales with water-filled containers can attain at least as good as ± four 50 ml syringes worth of volume!

Hopefully someone will check my workings here.
Posted By: Geoff Hannis

Re: Testing Patient Scales - 18/10/19 6:07 PM


Another important thing to be considered when checking patient weighing scales (and one that I place in higher importance than absolute "accuracy" per se) is:- repeatability of results (aka precision). smile

Accuracy is the difference between the measured value and the true value.

Precision is the repeatability of successive measurements under the same conditions.

Accuracy % = (Actual minus Measured)/Actual x 100%
Posted By: Geoff Hannis

Re: Testing Patient Scales - 18/10/19 6:44 PM


Originally Posted by Geoff Hannis

With reference to the attachment you provide, I have no idea what E2, F1 et al mean.


OK ... I've found something now.

It seems that Marsden use M1. smile

Note the ± 0.005% accuracy of the 5 kg M1 iron calibration weight (but note also the price)! This complies nicely with the attachment at Post #74654 (eg, 5 g at 100 kg - or 5 ml of water).
© 2020 EBME Forums: Biomedical and Clinical Engineering Discussion Forums.