Calibration Weights Needed

[Potashminer]

Apologies, South Pender - I chose word poorly - was not referring to "you" specifically - I should have wrote something like "what concerns the shooter" ...

The more I have read from shooters like Speedy Gonzales, etc., who shot many very small groups, the less to me is a "slam dunk" that charge weight, SD, etc. is always related to tiny groups - considerable evidence they do not always go together ... I suspect rifle construction, barrel quality, shooter skill may all contribute, as well as perhaps other unknown-to-me elements. I have witnessed exactly one 5 shot group in the "two's" fired - at 100 yards - was my rifle and scope, my hand loads, but I was not doing the shooting - and I never could replicate that, since. There are guys like Speedy Gonzales, Virgil King, etc. who worked their groups down from the "two's", through the "teens", into the "zero's" - maybe they knew something??

 

[South Pender]

Potashminer, no need for apologies! All's good. My experience shooting groups over several decades is like yours. I've shot very few really tiny groups--next to none in the 2s. But I've had very few rifles capable of such groups. My main interest these days at the range has been to try to get true hunting rifles to shoot better groups than is normally expected, and so, if I can get one of my hunting rifles to average .50" groups for 5 consecutive 5-shot groups at 100 yards (the testing procedure used by the NRA in their American Rifleman rifle reviews), I'm extremely happy.

 

[Maple57]

Sure, but what you are describing as necessary is consistency or repeatability. And I agree that this is important. However, if you can't be certain about the precise weight of the test weight, this is less than optimal. Your ball of tinfoil shows a weight of 60 grains on your scale and does so consistently, but in fact it really weighs 62 grains. If you then weigh a charge of powder until it registers 60 gr. on your scale, you're actually getting 62 grains of powder because your scale hasn't been calibrated. My question was more one of wanting confirmation about how a scale could be calibrated. And as I noted in Post #32 above, I think you'd want a precise calibration weight. I don't know whether a less-expensive scale like the RCBS electronic scale could be precisely calibrated, but something like the A&D FX-120i (which still falls short of your preferred level of precision) could be. So consistency is all-important, no question, but I'd want more than that; I'd want precision as well.

Seems like you are assuming something that is an absurd exaggeration of the point I'm trying to make...

The simple tin foil ball is a relative value. It is a fixed amount of weight. If the weight of that tin foil ball is the same as what you have developed as an ideal load, then it is the perfect "Validation Weight" for that load... Not to be confused with a "Calibration Weight".

The problem with only using a calibration weight is that you will never develop a statistically relevant sense of repeatability because you are only testing it while calibrating or validating a single occurrence, and it does not match your actual target weight.

If you validate every charge against a "Validation Weight", then you will see when the two values do and do not align. Again, develop a sense of repeatability. Learn to understand exactly how precise your scale actually is.

 

[South Pender]

Seems like you are assuming something that is an absurd exaggeration of the point I'm trying to make...

The simple tin foil ball is a relative value. It is a fixed amount of weight. If the weight of that tin foil ball is the same as what you have developed as an ideal load, then it is the perfect "Validation Weight" for that load... Not to be confused with a "Calibration Weight".

The problem with only using a calibration weight is that you will never develop a statistically relevant sense of repeatability because you are only testing it while calibrating or validating a single occurrence, and it does not match your actual target weight.

If you validate every charge against a "Validation Weight", then you will see when the two values do and do not align. Again, develop a sense of repeatability. Learn to understand exactly how precise your scale actually is.

You are obviously missing the point I was making. Repeatability is easy to evaluate. Whether you do it with a weight that corresponds to your "ideal charge weight" or a calibration weight is irrelevant. But I want more than repeatability; I want precision too. In other words, I want to know precisely what that "ideal charge weight" actually is (precisely XX.YYY grains)--not just that it corresponds to a "validation weight." Of course I want my scale to be consistent (yielding repeatable measurements), but I want to be confident in the actual weight displayed.

 

[Maple57]

You are obviously missing the point I was making. Repeatability is easy to evaluate. Whether you do it with a weight that corresponds to your "ideal charge weight" or a calibration weight is irrelevant. But I want more than repeatability; I want precision too. In other words, I want to know precisely what that "ideal charge weight" actually is (precisely XX.YYY grains)--not just that it corresponds to a "validation weight." Of course I want my scale to be consistent (yielding repeatable measurements), but I want to be confident in the actual weight displayed.

False, Repeatability is not easy to evaluate and you are clearly in the denial side of this point.

How many guys are going to use a 100 grain (for example) validation weight 50 times to get a sense of repeatability. Its an almost pointless exercise because 100 grains (or whatever...) does not likely match your target weight.

When you have something that weighs the same as your target charge weight, you can now verify with every single charge if it matches the requirement. I don't care if the validation weight is 3 paper clips that happen to weigh correctly. Its a perfect match to the target weight.

You can have a scale that is calibrated perfectly, within the accuracy of that scale, but if you weigh the charge 20 times, you may very well get 10 different answers. That is where having a test weight has value. Now you can see when the scale produces inconsistencies that you will never see when referring to your arbitrary calibration weight.

 

[Grizzlypeg]

My scale has a 5 and 100 gram test weight. I calculated the conversion to grains, and wrote the # on the side of the weights with a sharpie. I often place the 5 gram weight (77.16 grains) on the scale just to see if it reads the correct number in grains.

 

[South Pender]

False, Repeatability is not easy to evaluate and you are clearly in the denial side of this point.

How many guys are going to use a 100 grain (for example) validation weight 50 times to get a sense of repeatability. Its an almost pointless exercise because 100 grains (or whatever...) does not likely match your target weight.

When you have something that weighs the same as your target charge weight, you can now verify with every single charge if it matches the requirement. I don't care if the validation weight is 3 paper clips that happen to weigh correctly. Its a perfect match to the target weight.

You can have a scale that is calibrated perfectly, within the accuracy of that scale, but if you weigh the charge 20 times, you may very well get 10 different answers. That is where having a test weight has value. Now you can see when the scale produces inconsistencies that you will never see when referring to your arbitrary calibration weight.

Sorry, Maple57, but it may be that each of us is "in the denial side of" the other's point. I don't agree with your characterization of consistency, and you appear not to agree with my emphasis on precision. So, let's just agree to disagree or, perhaps more accurately, agree to have different priorities.

You seem to be suggesting that repeatability is meaningful only if evaluated with a validation test weight (e.g., 3 paper clips) corresponding exactly to the powder charge you are planning to use. I'd disagree with that view. If my calibration test weight of 150.32 grains (10 g.) gives an identical weight on 20 replications, I believe that it will do the same with my "target weight" (say 60 grains). If this were not true, you'd have to have dozens of test weights each matching a different charge weight to be consistent with your example. Most advice I've seen just recently re calibration is to use a single calibration weight that is much heavier than the range you will be weighing in. One source suggests a calibration weight that is at least 2/3 the maximum weight that can be recorded by the scale. Although I don't know it to be a fact, this recommendation may be based on the notion that any discrepancy at a very high weight would be accompanied by proportionally smaller discrepancies at the much lower weights actually used. However, if no discrepancy appears at the high weight, then, as noted, I believe none will occur at lower weights. We may disagree on this point, but I've seen nothing to suggest that discontinuities would be expected at different points on the weight continuum.

In the month and a half that this thread has been running, I’ve acquired an A&D EJ-54D2 scale (at a rock-bottom price from Cambridge Environmental), with readability to .0002 g. (roughly .003 grains) and repeatability (standard deviation) of .0004 g. And as mentioned earlier, I’ve acquired an OIML F2-level calibration weight (from Transcat) with .0006 g. accuracy. Actually, in use the scale moves in .004-grain steps as I add kernels of powder—so about double the step size of the high-end analytical balances you’ve mentioned in another thread. As a result, I feel I have more-than-enough precision and consistency for the kind of handloading I do. I’ve calibrated the scale to the calibration weight and check accuracy with the calibration weight before each loading session.