Scale resolution comparison

[Keith Glasscock]

Did I need a milligram scale?


While I am no stranger to accuracy and competitive shooting, I have always had to make hard choices about my gear based solely on my budget. It takes little imagination to believe that I considered buying a new scale for over a year before actually doing it. During that time, the available scale technology and prices changed significantly.
I purchased a new AND FX120i scale from Cambridge Environmental. Located in Canada, they do an outstanding job of cross-border commerce. My total, including shipping, was just under $400 US.

No sooner had the scale arrived when I started wondering about what real-world benefits it would provide. Experimentation commenced immediately. First, I left the scale on for three days with an empty powder pan zeroed on it. After those three days it still read 0.00 grains. I left a bullet on it for 8 hours. When first placed; 105.06 grains. After 8 hours; 105.02 grains. The second test did result in a zero drift - .04 grains. Rezeroing the scale brought the bullet weight back to 105.06.

Satisfied that the new scale was working well, I used my old scale, a 1990 Ohaus built RCBS 90, to weigh out the charges for a 22 shot string of F-class. I have developed a technique that I believe gets me the most out of that old scale, and I was curious to see the true value of the technique. The target was a 33.5 grain load of Reloader 15 for my recently built Dasher. The 22 charges were eerily good for a scale with a 0.1 grain resolution. The average was 33.49, the standard deviation .018, and the extreme spread (ES) .06 grains respectively.
As a comparison, I changed techniques to the “normal” way of using the old scale. Extreme spread jumped to .12 grains for the first 5 charges. Changing technique again, not allowing the scale to rest at zero between charges, the ES grew to .16 grains. Apparently, technique has significant effect.

Now I tested the FX-120i. Can charges be made with zero variation? In a word, no. The FX-120i doesn’t have enough resolution, and the variable kernel sizes of Reloader 15 would make it difficult at best. However, a .02 grain ES can be obtained easily. I’ll have to decide how get the best performance out of this new scale in the future, and whether I need better than .02 grain accuracy.

Did I need the new scale? The little Dasher has shown an average change of 10 fps for each .1 grain change in charge. If used as a basis, powder charges only contributed 6 fps to ES and 2 fps to SD over a 22 shot string. With a rifle that is intended to shoot only mid-range F-Open, that is a very small number. Even at 600 yards, it would likely be lost in the noise. Considering that I can likely only get a 4 fps ES improvement, it will have little effect on target –slightly more than the diameter of a bullet hole at 600 yards. If 1000 yard matches are on the menu, the equation changes significantly. That 4 fps ES is equal to approximately one inch of additional vertical – not something you want to give away in a class where lost points are rarely given back.

 

[Dos XX]

You can run that scale in grams and get three decimals which equals a little better resolution. However, the stated accuracy of the scale is .02 grains if I remember correctly, so I'm not sure the extra resolution can be realized in accuracy. Might be an interesting experiment. I have thought about it but haven't tried it.

I do run it in grams though.

 

[Jay Christopherson]

.02 should be enough resolution since the weight of a typical kernel of RL-15 is .025 grains. H4350 is closer to .03, but still, a single kernel should change the reading. If you are within .025, you should be good to go... Tough to do much better than single grain accuracy, unless you are worried about non-uniform grain sizes or cut grains from the thrower...

 

[normmatzen]

I thinkyou'll find the 0.02gr accuracy is for the grain mode as it probably increments in 0.02 grain increments.
If you put it on the gram mode, you will get 0.001g increments which are about 0.015gn.
I do this with my old Ohaus TS-200 lab balance and getting 1 granule accuracy is easier now.

By the way, my old balance has stability way better than you quote and I believe the balance you have is every bit as accurate as my Ohaus. Is your balance on a dead level, solid and vibration free platform?
I had purchased another Ohaus TS for a friend and I couldn't get stable tare. I finally took it off my cheap wooden file cabinet and tried it on my very level granite kitchen counter. Stable as a rock!

 

[normmatzen]

I believe you get 0.02gr accuracy as on grain mode the display increments in 0.02gr steps. If you weigh on the gram scale as mentioned above, you will measure in 0.001gram increments and the 0.001g accuracy is equal to about
0.015gr accuracy. I do this with my old Ohaus TS-200 lab balance and find weighing to the granule is easier now.
By the way, my Ohaus has better stability than you are quoting. I think your balance is every bit as accurate as my Ohaus. Do you have it on a dead level, solid and vibration free platform? I bought another Ohaus TS for a friend and I couldn't get the tare to be stable. I took it off my cheap wooden file cabinet and put it on my very level granite kitchen counter. Tare then became dead stable!

 

[Dos XX]

.02 should be enough resolution since the weight of a typical kernel of RL-15 is .025 grains. H4350 is closer to .03, but still, a single kernel should change the reading. If you are within .025, you should be good to go... Tough to do much better than single grain accuracy, unless you are worried about non-uniform grain sizes or cut grains from the thrower...

Agree, if you are loading those powders.

I am loading a powder that has small enough granules that you can see the .001 grams. Can I see the difference on target? I don't know, but I can do it so I do.

 

[amlevin]

The little Dasher has shown an average change of 10 fps for each .1 grain change in charge.

Out of curiosity, are you using the same case over and over for these speed variation "tests"?

 

[Keith Glasscock]

The little Dasher has shown an average change of 10 fps for each .1 grain change in charge.

Out of curiosity, are you using the same case over and over for these speed variation "tests"?

No, I used the same batch of cases, randomly mixed, on two shooting occasions. As with anything, barrels, cases, and bullets have significant effect on the outcomes. That said, ES muddies the waters more than a little. The point of the whole thing is that I would see, comparing my old scale and appropriate technique to the new scale, approximately a .04 grain improvement in charge accuracy over a 22 shot string. If that equates to 3 fps, 4, or 5, I think that an improvement in ES would be beneficial at long range, but not so much closer in (thinking F-class now).

All in all, I was pleasantly surprised with the consistency I was getting out of a 0.1 resolution scale. I now get better with a the 0.02 resolution, but will I really see it on target? If I shoot long range, I probably will.

 

[Keith Glasscock]

I believe you get 0.02gr accuracy as on grain mode the display increments in 0.02gr steps. If you weigh on the gram scale as mentioned above, you will measure in 0.001gram increments and the 0.001g accuracy is equal to about
0.015gr accuracy. I do this with my old Ohaus TS-200 lab balance and find weighing to the granule is easier now.
By the way, my Ohaus has better stability than you are quoting. I think your balance is every bit as accurate as my Ohaus. Do you have it on a dead level, solid and vibration free platform? I bought another Ohaus TS for a friend and I couldn't get the tare to be stable. I took it off my cheap wooden file cabinet and put it on my very level granite kitchen counter. Tare then became dead stable!

I do keep my scales on the most stable, level surface I have available. Unfortunately, my house was built new last year, and the newer houses around here exhibit significant floor flex. I'm sure that vibration could have contributed to the bullet weight drift. I use, the new A&D scale is rock solid; although, it does exhibit zero drift when a weight is left on it for hours on end. Fortunately for me, I only weigh my charges for a few seconds each. I have only had to re-zero the balance once during loading, and that was my fault.

My results with RL-15 are almost identical to Jay's, the granules are .022 or so grains each, but vary wildly in size. That is why I find it difficult to believe that I will have an absolute ES in my charge weights of less than .02 grains. I certainly don't have the resources to purchase a microgram balance in order to find out.

 

[johara1]

I would set up both scales weigh it on the old and check it on the new one and i will bet the old one is .1 gr. accuracy and the 120i is .02 accuracy and air movement will make them wonder. Can you see it on the target? You have a lot of other thing just as important to see it, but you did eliminate a variable. Now you can go work on something else …… you will like that scale …… jim

 

[amlevin]

If I shoot long range, I probably will.

Or not. Long range adds so many other variables how will you be able to single out the "better resolution of the scale" from air movement, shooter, rifle, etc, etc.

If one were to put 100 rounds each (loaded using old scale and new scale) downrange and then measure the average dispersion of the strikes, then one could deduce that one was better than the other or not.

 

[Keith Glasscock]

I agree, it takes some serious data acquisition capability to see small differences - especially through the "noise" of long-range shooting.

Even when considering the effect of charging errors on ES, one can easily imagine that some things will be additive and others may cancel each other out. From a statistical standpoint, you can estimate the effect of a change by adding the squares of the known errors and taking the square root of the result. In the hypothetical case of an ES of 9 (22 shot string) and a known scale error of .04 grains (once again assuming a 10 fps per 0.1 grain error), the result would be that we have a resultant average improvement of only 1 fps. That said, I fully believe in tolerance stacking, as I have seen it happen. That 9 fps ES could be 17 on the next 22 rounds...

When I say I would probably see an improvement, I would have to shoot a lot of rounds in identical condition, as you suggest, to get an average determination of improvement. Personally, I'm not going to do that as I don't have to prove that it works, instead, all I have to do is determine if my current load performs well enough for my needs.

In the end, did I need the new scale? Was it money well spent? The answer to both is a resounding NO. I might be able to academically see a slight improvement in rifle performance, but I shoot F-class and my rifles were shooting more than well enough before the change.

 

[johara1]

I was just down there loading for a match and see the numbers thrown out here, but i can see a change in the group when i go from 33.305 gr. of RL-15 to 33.355. That is a half a tenth, if you don't think a accurate scales is worth it then i think you should look somewhere else in your loading or your gun is lacking. To see the advantage of the quality equipment you need quality equipment to test it in. You better have a gun capable of a tenth at 100. I use a GD 503 but that 120i is the best buy on the market…… jim

 

[Keith Glasscock]

In Benchrest (capital B), I agree with you Jim.

I've found that F-class has the same score, and no tie-breaker built in whether you shoot a one-hole group, or a .25 MOA group in your target - if it is centered, you're golden. The real challenge is dealing with the variable of target pulling time and wind changes.

My Dasher is shooting better than before, a few changes to the stock helped. I never use wind flags during load development (not allowed at my club - it also forces me to use my wind-reading skills). Despite that, I was shooting consistent groups in the low 2's (only .020" difference between the best and worst group - something is consistent). Looking at my groups from the other day, I appear to have been about .1 grain too hot. It definitely had some vertical in it at 100.

That said, I'm not throwing the new scale away. I like it from a mental perspective. I know that I don't have to contend with the little bit of vertical that varying powder charges might cause. When I get the kid's 223 F/TR rifle going, the little scale will probably pay for itself.

 

[Raythemanroe]

Busdriver, wanna trade for a beam scale?


Ray

 

[Keith Glasscock]

10-10 or 505?

 

[Raythemanroe]

10-10


Ray

 

[johara1]

Busdriver, I think you are missing the point. When you say it's good enough, it isn't. What you need to think is if it shoots better it may shoot through conditions better, a nine may turn into a ten. I need one that shoots under a .1……… jim

 

[Keith Glasscock]

Jim,

You have a good point there. I can see how that could help. Everything is a balancing act, as you well know.

 

[johara1]

Busdriver, Yes sir i bang my head against the wall lots of times but when i start to get lazy and don't do something i know works, i will hang it up. I only move one thing at a time, it either helps or not then i don't visit it again. When you get to the point that nothing improves it, i screw on a new barrel. I went in and out on seating depth and up and down on powder and different neck tensions and i'm right where i was at the end of last year. Only a better quality barrel will help……. never quit trying …… jim