How Accurate is My Dillon Powder Measure?

[lovemywoods]

I enjoy testing different loads in rifles. I typically will prepare test rounds by carefully weighing each one on a RCBS ChargeMaster powder dispenser/scale combination. Once a “best load” is determined, I’ll make a few hundred rounds for that rifle.

So, can I trust the powder measure on my Dillon 550B to throw precise loads or do I need to hand-weigh these also?



The Test Procedure:

I created and ran a comparison between two different types of powder in the Dillon powder measure.

Hodgden’s CFE223 powder is a “spherical” type powder while their Varget brand is extruded into little rods. Just by looking at the two powders, it would seem that the CFE223 would meter better through the measure since it should flow better and because it had a finer texture. The rods of the Varget powder can get hung up on the reciprocating bar of the Dillon measure (or on the rotating parts in a rotary powder measure).

I decided to calibrate the powder measure and then throw thirty charges of each powder. Weighing them would show what variation I was getting.


I created this simple procedure:

By removing the brass locator pin in the #2 station on the press, I could insert, cycle, and remove the brass cases.

The powder in each case was weighed and the data recorded.

Results:

The data were entered into an Excel spreadsheet.

I like to see data displayed visually so I made a graph of each powder’s results.

-- Both powders produced an average very close to the target weight of 26.0 grains.
-- The Varget had a range of 3/10s of a grain. The CFE223 had a range that was half of that.
-- The CFE223 standard deviation was likewise half that of the Varget.
-- I would be OK making “everyday use” loads for a precision rifle with the CFE223 results. I’m less interested in making loads that are going to vary by 3/10s of a grain. Therefore, I would hesitate to make precision rounds on the Dillon if I was loading a “stick” or “extruded” powder.
-- Either powder could be used to load rounds for plinking or for a rifle class (short distance).
-- For competition or testing rounds, I will still load each round individually.
-- This simple test could be run using any powder that you are considering using.
-- Shooting stuff is fun. Making your own rounds to shoot is also fun!



Disclaimer:
This is not a scientific article. It is just for fun. Let’s not get into a discussion of significant figures or measurement variation. I took the liberty of expressing some quantities to two decimal places to more clearly show the differences between the two powders. I know that the initial data was significant to only three digits.

 

[BGDave]

Thanks. That was informative.

Electroic scales are really cool for this kind of stuff.

Any thoughts to comparing these results to a Uniflow or similar?

 

[TWEAKD4]

I like your analytical approach to your loads. Very good information. I hope to be loading soon and would like to mirror your method to keep accurate records. Thanks for sharing and keep up the good work.

 

[rvb]

you're also using a digital w/ readings only down to .1gr accuracy, so it could be better (or worse) than you're reporting.

If your target was 26, were the 25.9 really .1 low, or because the way the scale either truncates or rounds really something like 25.4 or 25.99? Same with the high end, was it really 26.1, or 26.19? Or some other variation depending on how your scale displays results....

This is why I don't put a lot of weight in how accurate the powder measure is or scale is.... I care what the load does over the chrono. Good low std dev in terms of fps and I don't care if the powder measure is off a half grain between throws.

I throw 5-10 charges and go by the average to get my setting. (for example, if my goal is 25, after 10 throws I know if my averaage is really 25.0, or really closer to 25.1 like 25.09). You can see this in your results. your average for the cfe223 is 30% of the way to 26.1 and your ave for the varget is 30% of the way to 25.9. your std dev might go down if you tweaked your measure a bit. The actual variation in the powder measure wouldn't change, but the results of your math would.

btw, I've not yet run varget through my dillon, but there is a lot of info out there how to get it to meter better from the dillon measures.

ps, w/ the dillon large powder bar, .02-.03 is "close enough" for me for my average. It's really hard to tweak it closer. W/ the small bar, I can often get it right on to the next decimal.

-rvb

 

[kludge]

Was this a rhetorical question?

rvb asks another good question. It would be interesting to know the basic resolution of your scale in grams. To do that trickle some ball powder and watch the increments in grams to see how finely they are divided. I have discovered that 0.01 gram resolution is not entirely accurate for reloading when grams is the native measurement (i.e. is measures in grams and coverts to grains, rather than measuring in grains).

However, because no one's scale is exactly like another, repeatability of the scale and not its accuracy, is more important for reloading.

Another question to ask is, in your test were you quantifying the repeatability of your powder measure... or your scale? The answer is both, and you need to do a separate test with, say a 30gr check weight first (the actual weight is not very important, only that is is close to what you are measuring) to gauge the repeatability and drift of the scale before you can know anything about the powder measure.

And finally, and perhaps more importantly, does 0.5 to 01gr of variance change the point of impact of the load? A ladder test or OCW test will help answer where the cartridge, load, and rifle are all in sync. A certain velocity and bullet from one load in tune with barrel harmonics may not yield an accurate load at the same velocity with the same bullet and another powder if the load/load density/powder/bullet combination isn't burning consistently.

 

[rvb]

rvb asks another good question. It would be interesting to know the basic resolution of your scale in grams.... I have discovered that 0.01 gram resolution is not entirely accurate for reloading when grams is the native measurement (i.e. is measures in grams and coverts to grains, rather than measuring in grains).

yea, since he said he didn't want to get all scientific, I didn't get into the issues w. unit conversion. but certainly that could make havok w/ his std dev calculations (.01 grams = 0.154 grains) if that's how his scale works. It really comes down to the chrono data and the results on paper. I wouldn't let this experiment be what keeps me from using the dillon w/ varget, for instance. I'd get some groups shot through a chrono.....

-rvb

 

[OneBadV8]

I noticed this with most powder drops that I've used, spherical powder always drops better than extruded. But, that's why I ended up with a Charge Master myself

 

[mssmith44]

That looks like great data to me. You used the correct number of samples for a statistically significant result.
The next step would be to load 30 samples of each with the Dillon, then your individual loading procedure.
Then a double blind test with the rifle in question. Then analyze the targets with a two dimensional variance to determine the best load procedure.
Or just shoot and be happy.
I have noticed that the smaller the area of powder cutoff in the powder measure the more repeatable the weight of powder thrown.
I have an old Herters measure that is about a 3/8 diameter tube that throws all types of powder uniformly.

 

[88E30M50]

Nice work. I'll have to try recording individual results like that with each powder. I'm a data junkie and like the idea of mapping how well the powder dispenser works with different powders. Right now, I run 10 charges and average them to see where the dispenser is at.

 

[IndyGlockMan]

Nicely done!
I could have told you that CFE223 meters better than Varget but I never knew how much better until now.
I use CFE223 for all my 223 loads. Great powder!

 

[kludge]

It really comes down to the chrono data and the results on paper. I wouldn't let this experiment be what keeps me from using the dillon w/ varget, for instance. I'd get some groups shot through a chrono.....

-rvb

Yup... It's the only way to know.

 

[GIJEW]

You had better results using varget in a Dillon 550 than I did. Being a stick powder, it got clogged up in the 'drop chute' and then dropped a double charge all over the bench after dropping a partial load in the previous case! I've had others tell me that they reamed out the chute to stop that. I just switched to RE15 for my CMP loads.
Not arguing with your experiment or rvb and kludge's comments, but--not having a chronograph--I've found that simply working up several test loads and taking the best one will give me ammo that will shoot better than I can (4/5 rnds in 3/8in @100yd with iron sights).

 

[kludge]

No argument here. The chrono is my most important reloading tool.

 

[Broom_jm]

And finally, and perhaps more importantly, does 0.5 to 01gr of variance change the point of impact of the load? A ladder test or OCW test will help answer where the cartridge, load, and rifle are all in sync. A certain velocity and bullet from one load in tune with barrel harmonics may not yield an accurate load at the same velocity with the same bullet and another powder if the load/load density/powder/bullet combination isn't burning consistently.

It's common, and perhaps understandable, for folks to get hung up on how much a charge WEIGHS, because that is what we measure to determine how much powder is going into a case. However, any ballistician will tell you that it's the VOLUME of powder going into your case that matters. While your experiment shows the average weight of powder thrown for these two powders is essentially identical, what you're concerned about is the occasional charge of Varget that varies from the average. What you have really done is show that volumetric powder measurement is a perfectly sufficient way to reload metallic cartridges, because the volume taken up by a given charge of powder is more important than the weight.

Something else to consider is the percentage variation from the average, or deviation percentage. Your lowest charge was 25.8 and the average was 25.97, which is a deviation of roughly one half of one percent (.655%). To further put it in perspective, the highest charge weight of Varget thrown from your Dillon was only 1.15% greater than the lowest charge. If you really want an eye-opener, carefully measure the extreme spread and standard deviation in case capacity between the 30 pieces of brass used. Again, keeping in mind that volume of case capacity consumed by a given volume of powder is the key, you'll start to understand that the amount of variance is truly nominal.

If you were to shoot both of these powders in an OCW test, you would likely find an accuracy node for each, wherein the same percentage variance documented in your testing produces no discernible difference in POI change at nominal distances. In other words, until you shoot these loads you really don't know whether or not your Dillon volumetric powder measure, or your RBCS digital scale, or the combination thereof, is sufficiently accurate for your needs. However, what you WILL find, in all likelihood, is that both are more than sufficient to the task at hand.

Find your optimum charge weight with either powder and the percentage variance you noted won't amount to a hill of beans, or more than a tenth at the bench.