Here's a little bit from my experiences with the Acculab VIC-123 scale (since sold and I'm now a happy user of a Sartorious GD-503 analytical lab scale - when I feel my Chargemaster 1500 isn't enough). He asked that I put this out on some of the gun/reloading forums to answer some of the questions he gets...
Having finally had about a gut full of my Acculab VIC-123 and its flaky behavior, I’m starting looking for something better, realizing full well that that ‘something’ is probably going to cost dearly. I’m not a 100% sure that a true milligram scale is really necessary for our applications (powder charges), but I like the idea of ‘knowing’ that my charges are exact. Plus, if I get around to doing some of the experiments I want to do, it’ll be helpful to actually *have* a good reliable lab scale.
In the past I’ve done a lot of ‘window shopping’ @ balances.com. The site is fairly well organized, has what appear to be reasonable prices, and a pretty extensive selection. As an added plus, the owner (Gary Shane) will usually answer most questions related to sales of items he carries. Sometimes in an abrupt manner, but I think fairly and honestly as well, which means more to me.
I started asking Gary some questions concerning a few blurbs I saw on his site regarding the Acculab VIC-123 scales (which he does sell) and the strain gauges used in them. His responses were fairly educational for me. I added some emphasis to some key elements below:
I'll make it clear. Strain gauge is only good for the very best 15,000 divisions
http://balances.com/applications/glossary.html#strain. The Acculab VIC-123 has 120,000 division which is total impossible using a strain gauge. They make these divisions using software but as you experienced the software doesn't cut it.
A strain gauge is a scale and applying a load results in the deformation of the load cell. Over a point of time the load cell loses it's elasticity resulting in greater hystersis
http://balances.com/applications/glossary.html#hystersis
Bottom line stain gauge is good only for a kitchen scale 3000 division (3000g x 1g) or 5000 divisions (5000g x 1g)
If you want repeatable results you need to go with a magnetic force restoration balance. Visit
http://balances.com/applications/glossary.html and learn about balance terminology. Visit
http://scientech.balances.com/scientech/forcemotor.html to better understand a force motor.
If you want the best review
http://balance.balances.com/scales/559 GD503 - 1545.955 GN x 0.005 GN
If this is too pricey then you'll have to buy the ALC-320.3 at
http://balance.balances.com/scales/171
The next round of Q&A was similarly interesting…
(me) Would that also be the reason that occasionally my VIC-123 indicates an odd last digit i.e. 46.79gr instead of the 'usual' 46.78 or 46.80gr (normally everything increments in 0.02gr, but every once in a while it'll indicate an odd number last digit)? Some sort of glitch in the software rounding?
(Gary) The software is all algorithm based and the algorithm wasn't something they spent a million dollars on. What the scale reads one day verses the next is why I tell people it is more of a novelty or joke scale. Anything with a strain gauge more than 12,000 division is a joke scale.
(me) Do you happen to know what kind of load cell is used in most common 'reloading' scales such as those sold by Lyman, RCBS, Dillon, and others?
(Gary) These scales have lower quality load cells and the algorithm is worse. Ohaus Corp built some of these scales for Lyman, RCBS and Dillion from what i heard but I don't have any direct proof so this is only hear say.
(me) I think what people would be most interested in seeing is in realistic terms, how much more accurate is the VIC-123 than a common reloading scale that is commonly advertised as +/- 0.1gn? If they both use the same strain gauge technology and it is so limited in terms of divisions... is the +/- 0.02gn readability just smoke-n-mirrors?
(Gary) If you look at the specs at
http://balance.balances.com/scales/906 you need to look at "Linearity ± 0.004 g" Since the scale reads 0.001g that means the linearity (aka accuracy) +/- 4 divisions.
Now in grains the scale increments in 0.02 gn divisions so the linearity is 0.08 grains (0.02 x 4)
Remember this is in ideal environmental conditions and testing on a cherry picked scale built perfectly.
If you ask me you would be lucky to have even 0.8 grain accuracy.
I had to ask for a bit of clarification on that last point…
(me) Is that last section supposed to be '0.08gn'? Otherwise I don't follow the leap from 0.02gn x 4 = 0.08gn to 0.8gn.
(Gary) Again that is my opinion on the accuracy is like 0.8 grains but according to the manufacturer spec it is 0.08 gn.
