The Inaccuracy of Digital Meters

[AgX]

Again and again I read at Apug quite a nonsense relating to digital readouts of scales and thermometers.
This dramatic case hopefully makes you aware of the potential errors:


These days in Germany a healing practitioner is at criminal court for killing 3 of his patients by applying by negligence an overdose of a drug.

His aim was to apply by infusion 105mg of that substance and he had to weigh the substance before.

He employed a digital scale from a laboratory scale manufacturer (Kern PCB 200 at about 250€)

of

200g max load

and

-,-- g display

and

d= 0,01 designation


-) He did not realize that with any digital readout the last digit (10mg in this case) is insecure and can be +/- 1digit


-) He did not realize that, more important, a metering device by design may incorporate an error by far greater than that last digit, especially at the verge of its range.


A expert witness showed that the scale in question at 1160mg can have an error of 340mg. Thus 30%.



I looked up at the data sheet:

linearity: 20mg

minimum load: 20g


(Linearity here means the max. absolute error over the complete load range. In this case it is lower than the error shown by the expert. But that data sheet error only applies at the allowed load range, here thus above 20g.)




I hope this makes things more clear...




A pun saying amongst german electronic technicians is "Wer mißt, mißt Mist".
Which can be translated as "Every metering results in bullshit". This is applied on electronics metering where the effect of the circuit on the metering complicates things, but doubt on what one is doing should be applied on any metering.

 

[Vaughn]

In the States, we just say that two light meters will never agree. With plenty of room for exceptions, of course.

The health practitoner was poorly trained, supplied and is an idiot. Fortunately, photography is neither rocket science nor hangs a human life on a thread, so a high level of accuracy is not as important as repeatability.

PS -- I like it when I can just use a heaping teaspoon of this, that, or the other thing per liter/quart.

 

[AgX]

But the suppliers are to blame too, for negligence or even intent.

With intent I mean those who emphasize the several-digit readout, but give no hint at tolerances.
With negligence I refer to manufacturers like the one of this scale, who not add features. As clearly stating the max error next to the display and blocking the display at loads below the minimum one.

 

[faberryman]

Why was he using a 200g scale to measure 105mg? Therein lies the negligence.

 

[MattKing]

Why was he using a 200g scale to measure 105mg?

Because the scale purported to display results down to the mg level, and he/she was either too dumb or too ignorant or too careless to know/do better?

 

[Mainecoonmaniac]

Meters are just tools. If a photographer wants absolute the "perfect" exposure, there's no such thing. I use my experience and gut feelings for my exposure which isn't perfect either. A simple example is backlit subjects. I know my meter is going to lie to me so I'll open up 2 stops and let the background blowout.
Check out Fred Parker.
http://www.fredparker.com/ultexp1.htm

 

[Ko.Fe.]

Why "expert" witness is loading one kilo on 200 grams maximum scale?
It is like asking this "expert" to hold 200 kilos.

 

[Sirius Glass]

In one of my undergraduate Electrical Engineering classes, a professor was calculating a charge on a drop of water when he dropped the negative sign. One of the students pointed out the error and the professor said, "Who cares about twenty-four orders of magnitude." After that we said that exact answers are not needed on exams.

 

[trendland]

Again and again I read at Apug quite a nonsense relating to digital readouts of scales and thermometers.
This dramatic case hopefully makes you aware of the potential errors:


These days in Germany a healing practitioner is at criminal court for killing 3 of his patients by applying by negligence an overdose of a drug.

His aim was to apply by infusion 105mg of that substance and he had to weigh the substance before.

He employed a digital scale from a laboratory scale manufacturer (Kern PCB 200 at about 250€)

of

200g max load

and

-,-- g display

and

d= 0,01 designation


-) He did not realize that with any digital readout the last digit (10mg in this case) is insecure and can be +/- 1digit


-) He did not realize that, more important, a metering device by design may incorporate an error by far greater than that last digit, especially at the verge of its range.


A expert witness showed that the scale in question at 1160mg can have an error of 340mg. Thus 30%.



I looked up at the data sheet:

linearity: 20mg

minimum load: 20g


(Linearity here means the max. absolute error over the complete load range. In this case it is lower than the error shown by the expert. But that data sheet error only applies at the allowed load range, here thus above 20g.)




I hope this makes things more clear...




A pun saying amongst german electronic technicians is "Wer mißt, mißt Mist".
Which can be translated as "Every metering results in bullshit". This is applied on electronics metering where the effect of the circuit on the metering complicates things, but doubt on what one is doing should be applied on any metering.

In case of digital scales the accuracy is concerning to the price Agx - you know for sure!
You want to waight a chem of just 2g.....terrible task!
There you have the need of precision! A normal precise special scale have + - 1g and is expensive!
Of you looking trough the net there are lots of Chineese digital scales loser priced!
The precision is around 0,1 g - what does this indecate ? Thats Chineese business - because the correct precision seams to be from the class of + - 0,5 g what would mean you are correct about
0,7 - ~1g.....!
But it you are able to spent lots of money you are able to get the correct class :
precision 0,05 - 0,1 g!!!!!

That would mean precision around 0,11g...!
But look at it : you can messure up to max 20g!!!!!!!!

Not for use to weight 500g with precision of 0,11g! Easy going! But not easy to understand at once! I can't say the reason why Chineese digital scales came more and more cheap ?
Caused from the need of drug dealers?

But in cinema films they are using normal household scales and get normaly shot during business!

So you example is showing - one comes into prison from wrong messurement here - in the US
you normaly get killed!

with regards

PS : It is the human stain - technic isn't allways responcible for human made failures!

Exeption : Challenger missfortune!

 

[jim10219]

You don't need to clutter up the scale with junk that any professional should already know. Any time a professional, and especially one who's mistakes can kill people, gets a new piece of equipment, they need to know how to use it. It's the same with any profession. And really, that rule should apply to everyone, not just professionals. But professionals should be held to a higher standard, because it's what they do for a living. Besides, putting that information on the front isn't going to help because if he didn't care enough to look it up in the manual, he probably wouldn't care enough to read any printed warnings on the faceplate. Plus, there's probably more to the scales specifications than would fit on a faceplate anyway. So the only logical thing to put next to the display would be "read and understand manual" which he shouldn't have to be told to do, to do.

And lastly, you can't tell me that if the scale was that far off that he didn't notice it bouncing all over the place as he approached 105mg. I've used inaccurate scales before. I know how to tell if it's off. That should have tipped him off that something was wrong. But rather than go the extra effort to investigate and resolve the issue, he just ignored it, killed some people, and then hired a lawyer to figure out a way to blame someone else for his laziness.

 

[AgX]

I very much doubt that all professionals who treat patients including physicians(!) have sufficient training on such matters of metrology. And I speak of own academic experience.

A scale not necessarily have to bounce all over the place if in insecure range, and in this case it seemingly did not (court report). Thus someone understanding the d=0.01 as indication of precision may be lead astray.

 

[guangong]

Reading the above comments, all sophisticated and well learned, I am satisfied with my scales. One is a balance scale that was given to me by the manager of a scientific supply house when I was 12 yrs old (I went to buy a couple test tubes. Years later I realized the things he gave me must have been returns). Pans too small for large quantities of chemicals, but perfect for very small amounts. The other scale is a beam balance scale with a large pan. No batteries. No electronics.
I enjoyed reading the commentaries. Learned a lot.

 

[AgX]

I too got a balance scale. With a running counterweight and a mechanical mixed digita/analog readout. I can read it much better than 0.05g. And possible causes for error are, at least for a somewhat technically minded person, much better to realize than the black box of modern scales.

 

[craigclu]

Even highly regarded scales can be suspect at their extremes for their load cell ranges. When forced to measure in a scenario as described (though not for medical dosages!), a moderate weight is applied, then zeroed and then will likely make for a more accurate reading of a small amount being weighed.

 

[BrianShaw]

Because the scale purported to display results down to the mg level, and he/she was either too dumb or too ignorant or too careless to know/do better?

Woulda jury concur with that as a good definition for negligence?

 

[MattKing]

Woulda jury concur with that as a good definition for negligence?

Maybe, but this is one of many such cases that I would have pushed to be heard by a judge alone.
Up here, we do that with most civil cases.

 

[RalphLambrecht]

Maybe, but this is one of many such cases that I would have pushed to be heard by a judge alone.
Up here, we do that with most civil cases.

most interesting thread but, can somebody knowledgable now summarize how to use a digital scale properlyplease?

 

[MattKing]

Pick one designed to be used for the weight that you are trying to measure, and understand that its accuracy is something different than how many digits show on the display.

 

[Kevin Caulfield]

Accuracy and precision are two different things.

 

[AgX]

most interesting thread but, can somebody knowledgable now summarize how to use a digital scale properly please?

-) read the manual

-) use it in the allowed range (not the possible range)

-) keep in mind the stated error within that range

-) keep in mind the last digit always is +/- 1


If there is no manual or these informations, do not trust it.

 

[RalphLambrecht]

Accuracy and precision are two different things.

how can I check my scale for both?

 

[AgX]

how can I check my scale for both [Accuracy and Precision]?

Well, both are terms out of metrology, I consider rather puzzling in this context, out of themselves and due to varying terms in different languages.

Precision means the range of a values yielded after repeated measuring of the same state.

Accuracy means the deviation of the median of the range of precision.

See also this graphic:
https://en.wikipedia.org/wiki/Accuracy_and_precision#/media/File:Accuracy_and_precision.svg


Thus accuracy is NOT the greatest error...

Now you likely will understand why I do not consider these terms helpful here.

 

[Kevin Caulfield]

how can I check my scale for both?

You need to measure against a known value to determine accuracy. To determine precision you do multiple measurements and check repeatability.

 

[BMbikerider]

Comparing my D700, F6, F100, Minolta Autometer 3 and my Weston V all set at 200iso (ASA) they all differ but not by much.
In testing them, I used the white vertical blinds in my living room back-lit by the natural daylight outside. It is a nice even light and with the cameras/meters used about 4 feet away they vary by no more than 3/4 of a stop with the Weston (not surprisingly giving the longest exposure...ageing selenium cell!) Even with the 3 Nikon's used with all 3 variations of metering pattern there is little difference.

 

[Mainecoonmaniac]

Does anyone have their own "personal" ASA? I think tweaking your film speed to suit your taste can make the box speed irrelevant. Ralph Gibson shoots Tri-X higher that box speed for his look as an example. But that doesn't make his meter inaccurate.