Due to some recent reading I decided to do a two part test on some powders, and this is part 1. How accurate do certain powders throw from a measure? I can tell you from experience it depends on the measure and conditions, but I'm here to share data. The testing I did today was from a Lee deluxe perfect powder measure, the hand measure, not the auto drum, although I have both. Do not knock them till you try them. I've had RCBS uniflow, and Lyman 55 measures in the past, and no longer have either. The Lee drum measures are the best powder measures on the market today for extruded and flake powders, at least for normal measures. I have not tried a Harrells or Redding BR or anything like that. I am mostly interested in powder measures for a progressive press.

I tested 4 powders today. IMR 3031, Winchester Staball Match, IMR 8208XBR, and Alliant AR Comp. Three of them were purchased in the last month, the IMR 8208 was bought back in October or November. Just to clear the air, the date code on the 8208 says it was manufactured 10/10/23, so I can guarantee they have made 8208 recently. It has not been missing forever as some claim. I wasn't on any waiting list or anything, I just happened to see it in the store one day. The weighing was done on my RCBS 505 balance scale. I was shooting for 30.5 grain charges because it seems a good middle ground for what most guys are loading from 223 to 308 and in between. The 0.5 grains was so I only had to move one slider when weighing. I first filled the powder measure hopper, set the measure to drop about 30.5, then gave 10 drops into the container to let the powder settle in. I then started the test.

Note: My Lee measure has the old style red hopper, and I am running a Titan Reloading baffle in that too.

I did 30 drops for each powder, then calculated the results.

IMR 3031
Average- 30.37
Extreme spread- 0.4
Standard deviation- 0.137

Staball Match
Average- 30.57
Extreme spread- 0.2
Standard deviation- 0.063

IMR 8208XBR
Average- 30.38
Extreme spread- 0.3
Standard deviation- 0.076

AR Comp
Average- 30.57
Extreme spread- 0.4
Standard deviation- 0.093

Now a couple of notes, the IMR 8208XBR came so close to some crazy results. For most of the test, it only varied by 0.2 grains! Right at the end, throw 29 of 30, it threw a 30.6 grain charge bringing the spread to 0.3 grains. In a similar fashion, AR comp only had a single throw at 30.4 right away throw 3 of 30, bringing its spread to 0.4. Without that it would have been a spread of 0.3. It makes me wonder if I shouldn't have gone all the way to 50 throws for more accurate data. I feel 30 at least give me what I want to know anyway. I'm really surprised by 8208, that's pretty much as good as ball powder. I may try some Ramshot ball powder in the future since I know that stuff is usually the finest of the fine ball powders.

Conclusions: There's no surprise IMR 3031 was the worst of the test, but I'm quite surprised at how good it did. I have to attribute that to the Lee measure. I really doubt you are going to get that from a Lyman 55 crunch box. The Lee wiper really works, 30 throws smooth as silk with 3031. Now remember this is hand thrown, these numbers will not be this good on a progressive press with vibrations and inconsistencies and all that. Still it goes to show 3031 is not as horrible as some claim.

The next part of this test I want to test 3031 and Staball Match head to head on target and for velocity with thrown measures. 3031 will have more varience in the powder charge, but how much does that translate on target? I'm not 100% certain, but I only know one way to find out.

Also for anyone interested in bulk, I did record the powder measure setting to get about 30.5 grains of each powder. Obviously I was a little off on the average, but pretty close. I have no idea if the Lee measure numbers correlate to actual CC volume or not, but that's how I'm recording it.

IMR 3031 25.8cc

Staball Match 22.9cc

IMR 8208 23.7cc

AR Comp 25.7cc

Fantastic!  I love testing.

I do this type of test every time I get a new powder.  

I use a box-stock, unmodified Dillon powder measure.  

Typically, I have done as many as 100 throws but certainly 30 is a large enough sample to tell the tale.  

I don't have the all data in front of me but here's what I checked and as I remember some.

Extruded powders:  SD = 0.14 gr (IMR4895, H4350, Varget, IMR4064 and others)  note - Varget throws as well as any other extruded powder

Ball powders:  SD = 0.014 gr (W760, W296, TAC, Hunter, Magnum and others)


What I noticed while doing my very first series of tests was the importance of consistent technique.  Later, working on additional tests I found it was essential to give the powder time to fall into the case (you can hear it).  Go to fast and you lose consistency.

It may seem like a waste of time but I learned a lot doing it.

I wouldn't call it a waste of time. For example, before this test I had every reason to believe AR Comp would meter every bit as good as 8208, and it does not. Most of my testing has been with flake powders for pistols, but even there it has shown me things. I've also tested Lee powder scoops, and some might be surprised how accurate they can be. I wish I had taken better notes in the past to share.

Well darn, I knew I was out of Ramshot TAC, but I could have sworn I still had some Big Game, but it is not in the locker. Now that I think about it, I might have given it to someone who could use it. I did find I have Accurate 2700, I might test that, I might not. I know I haven't ever got results like you have with your ball powders. that is crazy good. I'm not sure I can even hand trickle that good without going insane.

Yes technique matters, but I don't try to pinch my nose, while reciting the pledge of allegiance, and hoping the moon phase is right anymore. That Lyman 55, wow, I just do not get why some people consider that the gold standard. All those knobs and meaningless numbers, plus that knocker on the front, and it doesn't even work that good. I too have found you have to be certain the hole is filled. My technique is just be smooth. Raise the handle to the stop, not too gentle, not so hard it bounces, hold until its full, you can hear it, maybe give it an extra second or two, then lower smooth until the bottom. Again not so hard or so soft. None of that double tapping at top or bottom or any of that other nonsense. I've been doing it for so many years now I just do it.

I have no use for 4198, so don't hold your breath.

You are going to have to fight me off to get them, ha.  Seriously, I bought a Lee because of all the threads saying they meter stick so well but it had the pointer on my 5-10 swinging far on both sides of the spot but seldom on point, with H4895 or Varget, so it sits in a box somewhere and my 55's continue to reign supreme.

Consistency of results is all about the user and managing the nuances of each measure.  
The manually operated DPM with the spring return used give good results loading IMR4895. Just like my RCBS Uniflow has for years.  
But some reloaders have no skills at all.  They just want "perfection" so speak...without any effort on their part!  

I am not going to double space every single line. Either fix the forum, or zoom in your screen. If you guys are not capable of reading an article, then I'm in the wrong place. I'm here to share data and knowledge, and some of that requires a "wall of text" I have not tried the Dillon measure, but the idea of it measuring "any flake powder" "exact" or even close to ball powder seems far fetched.

You obviously did not read my article, or did much for testing yourself. Staball Match is a ball powder, it did not measure exact. It barely edged out 8208. You could argue it's my technique, all three extruded powders threw more accurately than most people report.

I've been on this forum about a week, and this is the third or fourth person telling me to double space every line. Why the heck is this not just a standard option? Is it better if I increase the font size like this? I flat out refuse to hit the enter key twice every single line. That is ridiculous.

For those who want perfection, and for those who want perfection without any effort, there is the AutoTrickler.  It requires no technique or effort, at all, just $1k to spend.

If you do not have or do not want to spend $1k for very tightly controlled powder charge weights,...

Try this test.  
1.  Do 100 drops, weighing each one.  
2.  Write the weights down.  
3.  Review the numbers as you go.  
4.  See what happens when you chunk a kernel.  
5.  See what happens as you vary your technique a little.
6.  Make notes on what you changed and where you made the change.
7.  What did the change do?

I had to modify the Lee PPM to get optimum performance: Tuning Lee PPM

Well yes and no. Some brands are much better than others. I would have to assume most would be more accurate with something like IMR 3031, but from what I've seen, a lot of times the powder measure is more accurate than the auto trickler with finer powders. Plus they are SLOOOW. Some are faster than others, but most of what I've seen, you can measure, then trickle up by hand faster than an auto trickler. As you say, the good units are $1000, and for the really good ones they can be upwards of $7000. I don't know about the new one, but the old Hornady auto charge is a joke. I could turn the speeds up as much as possible, and it still took 20+ seconds for any charge of powder. And it wasn't even accurate. It almost always went over by a tenth or two.

The other part of that is I think people put a little too much faith in an electronic display. A lot of those lower cost electronic scales are not nearly as accurate as people think. They might be fine right away, but they always seem to drift. If someone ran an auto trickle and it lets say it was set to 30.5 grains, and then took that and dumped it in a good balance scale, I think people would be shocked how much it actually varies.

No one is disputing that small lots do become available. You just can't get it in what a lot of us would call reasonable quantities. I don't buy smokeless powder in gun stores and I pretty much stay away from 1lbs bottles of the stuff unless there is a reason to test somthing.

I have shooting BOT set to send me notifications on 8208 as well as 4064 and Varget. It will tell you how many weeks out they have been and right now we are looking at 26+ weeks for 8 pound jugs of 8208.

Some of the availability issues have to do with European REACH compliance and the fact that non REACH compliant powders can no longer be imported to Europe. Since a huge section of the market dried up for these Australian & Canadian made powders, guess what they stopped making as much as they use to and put it on much slower production schedules.

Ah, perhaps we are not quite on the same wavelength here.  

You are saying "auto trickler" (generic description) and I am saying "AutoTrickler" (a product name), as seen at the link below.  They are both fast and "accurate to the kernel".  

The A&D scale that's part of the AutoTrickler does not drift.

Link to AutoTrickler


All that said, I agree with the rest of your post regarding electronic scale (in)accuracy.  Even my $300 AccuLab drifts and required developing a special technique to use to good effect (+/-0.01 gr).

Some people have argued that with me. I too used to buy in 8 pound jugs, probably will again too, but it just does not make sense at this exact moment. Things could be completely different 6 months from now, but at this exact moment you can buy eight 1lb cans of powder in a store for less than you can have an 8 pound jug of powder shipped to you. In my case my store sells jugs too. If you would have bought 8208 in 1 pound cans back in October, you could have saved money verses todays price increases. I think it was $40 per pound then. I'm not saying you should do anything different, but I am saying the idea of a jug being cheaper than 1 lb cans is not true at this moment.

Absolutely not. Fix the forum. I'm not writing a whole article and doing that.

I'm continuing on a line of thinking from my earlier post.

The reason I do not have an AutoTrickler is that perfect powder charges do not equate to zero SD on velocity or improved accuracy on my targets.  It really helps at 1,000 yards but up close,... not so much.

I say that because when I hand weigh the cases and powder charges so they are absolutely PERFECT, I still get a non-zero SD.  Last time I measured SD for hand weighed charges, SD was very low (6 fps) but it was not zero.  Add to that, I have developed a fast and easy way to trickle charge up to the desired powder charge.  Taken together, I cannot justify spending $1k for an AutoTrickler.

That type of precision reloading really helps at 1,000 yards but up close,... uh, not so much.

fgshoot, what type(s) of shooting are you loading for?

I load everything, so it really depends. Surprisingly, I find handguns are one of the most touchy to powder charge. I would have to assume it has to do with recoil, but powder charge can definitely change bullet impact quite a bit. It does not seem to be as much of an issue in rifles. Shotguns are basically a non-issue.

A lot of my rifle workups have been on a single stage press and hand weighed powder, but I wonder if that may be a mistake being as once I get a load, I throw from a measure. Some things such as my hunting ammo is always measured to perfection.

I too trickle my charges if hand weighed. I tried an auto trickler (a Hornady, not the autotrickler brand), and I just don't see it. I could invest in a really expensive unit, but why? Like you I throw a charge, put it on the scale, and trickle it up. I doubt it takes 10 seconds tops. I would only do this if I'm loading small volume, but for 50 rounds, it's not hard. I'm not going to spend $1000 just so I can watch it whirl around automatically.

I'm mostly testing my measure out of curiosity, it's not really going to effect anything for me. What I should do is test powders in the progressive press since I know that won't be as accurate. Like everyone I'm looking for the perfect answer where there isn't one. I think there is this idea that ball powders are perfect, and it sure isn't true from what I've seen. They measure good, not perfect. And ultimately what matter to me is the target. I shoot mostly extruded and flake powders because that's what gives me the results I want.

Those 8208 numbers dont surprise me, its damn near as good as ball powder.

Ive done several 50 drop tests checked on my FX120i and always found it to drop +/- .08 from my Hornady drops, so not even a tenth of a grain.  Ball powders go about +/- .05.  
I would imagine Benchmark and other short cut extruded would be pretty close to this as well.

Here's some proof you can't speak in absolutes. I talked to my loading buddy, and not only did he still have the Ramshot Big Game, apparently I had given him TAC as well, and never used either. So now I have both again. I repeated the exact same test as described above for TAC, being an Accurate/Western brand powder, I knew it should produce as good of results as I've seen so far, and it did. Realistically it measures as good as my scale can measure. I do not believe I can hand trickle this powder better than it throws from a measure. This was no surprise. Definitely look above at Staball Match before saying all ball powders measure perfectly.

TAC
22.3cc

Average: 30.58gr
Extreme spread: 0.1 gr
Standard deviation: 0.042



I also dug out some flake powders to show what they do, especially with the common belief that 800x is the worst ever. I thought about using a higher charge since shotguns would, but I figured most would rather see a smaller charge for handguns. I figured 10.5 gr would be a good middle ground from many calibers from 9mm to 44 magnum. Besides that the test was the exact same.

Bluedot
11.9cc

Average: 10.75 gr
Extreme spread: 0. 3gr
Standard Deviation: 0.085

800X
13.9cc

Average: 10.56 gr
Extreme spread: 0.5 gr
Standard deviation: 0.143


Now for some beloved target powders

700X
16.5cc

Average: 10.40 gr
Extreme spread: 0.8 gr
Standard deviation: 0.213


American Select
16.3cc

Average: 10.68 gr
Extreme spread: 0.5 gr
Standard deviation: 0.112



Conclusions-
First of all, I have tested bluedot before, multiple times in multiple powder measures, and it normally has a 0.4 gr spread, so this is actually better than normal. American Select had one lone charge at 10.4, otherwise it's extreme spread would be 0.4 gr. Realistically these two powders aren't a mile apart, although it appeard that Bluedot is a smidge better overall with the standard deviation. 700x is a prime example of why you have to be careful parroting online knowledge. To be fair I love 700x as a shotgun powder. It is one of the most consistent shotgun powder over a wide range of changes, and the lack of charge consistency does not hurt in a shotgun much. It even still produces good velocity spreads. Still, I have never liked 700x in a handgun. This should not be a huge surprise to shotgun reloaders, but 700x is one of the worst metering powders there is. It's a relatively small flake, looks just like blue dot or unique or anything else. The problem seems to be a lack of coating. 700x is gray looking, and you can see it kind of clump as it flows in a meter. American Select is a great alternative in a handgun, and it's my go to for target ammo in smaller rounds such as 45 acp.

Now lets talk about 800x. It's unfortunate it got such a reputation, because it is one of the most underrated powders out there, both in shotgun and handgun. When it comes to pure accuracy, 800x is one of the best handgun powders there was. Yet people were scared off from the big flakes, and now it's discontinued. Here we see it does not even meter that poorly. It's not good, but it's not that horrible either. I personally don't think it deserves the reputation it has.

Ok, one more test. I should really stop messing around and get back to actual loading. This is the same exact test as before. I chose some handgun/shotgun powders that are more middle ground.

Green Dot
16.5cc

Average: 10.54 gr
Extreme spread: 0.3 gr
Standard deviation: 0.102


HS6
9.6cc

Average: 10.47 gr
Extreme spread: 0.15 gr
Standard Deviation: 0.051


Unique
15.5cc

Average: 10.47 gr
Extreme spread: 0.3 gr
Standard deviation: 0.083


Conclusions:
It looks like the Alliant flake powders are metering really good, much better than the Hodgdon's 700x and 800x, which should surprise nobody. I am a little disappointed in American Select. The surprise to me is how good the other Alliant powders are doing, they have not done quite as good in the past for me, and a thought occurred to me. I do not know if I have tested these with the baffle before now. I may take the baffle out and retest bluedot and 800x to see how it does. HS6 has always been a mediocre powder to me, so I don't have much data on it. It did good in this test. You will notice I put the extreme spread at 0.15 gr, where I have been trying to keep it to tenths. There were a few throws where it was dead between 10.5 and 10.6. If it was just one throw I might have ignored it, but it happened more than once. Realistically it does not matter. HS6 is near the limits of what I could hand weigh, and for sure there would be no benefit to doing so. I only wanted to include another ball powder to show what they do. I never gave much thought to Unique, maybe I should give it more of a try than I have. It measured really good in this test.

I mean no offense, but I do not consider errors of 3 - 5 tenths as being accurate, not even close.  Still requires trickling and you can do that with any powder measure.

No offense taken, I mostly agree with you. I was mostly aimed at dryflash3's post, which he clearly hasn't tested powders in the same manner I am. He stated "Any flake powder (except 800X) = measures exact, same with the Dillon PM's. Yes this included Unique." Which is not true. No powder measures perfect, although your definition of perfect can vary. I tested a pretty good range of flake powders that are common, and not one of them was what I call exact. 800x is not the worst metering powder out there. And he seems to point to Unique being poor or average, yet it was the best metering flake powder I tested.

He also claimed any ball powder measures exact, which is not true. Certain ones such as TAC measure so good you can't really get any better off a normal scale the measures to the tenths. I do not own a scale that measures to the hundredths. Different ball powders do measure differently too. How much it matters is up for debate.

I'm sorry to pick on him so much, I'm sure he has a lot of knowledge, but he is a good example of why you have to be so careful what you read on the internet. You have to understand the difference between experience, and real data.

This is experience, not data talking, but I can tell you you can not lump all powders into categories. You can not state all flake powders are the same, or all extruded are the same.

Let me explain my thinking relative to this.

+/-3 sigma is about 1/2 grain of powder.

You get roughly 100 fps per grain (some are higher some are lower depending upon the cartridge).

Taken together, that means about +/- 50 fps due to using thrown charges (probably less, as errors don't work in that simple manner).

Running JBM Ballistics for 175 SMK at nominal 2600 fps says, +/- 50 fps means -

+/- 1 " vertical error at 300 yards (+/- 0.35 MOA).

+/- 5" vertical error at 600 yards (+/- 0.85 MOA).

So, you have a small chance (3/1000) of being about 1/2 MOA off due to your use of thrown charges.  It will be less than 1/2 MOA if you are in close and more if you go long.



Put another way, you have a 70% chance of being within 0.1 MOA if you are at 300 yards or less.

The above is oversimplified and probably worst case because error do not simply add (as was done in the above).  It would more likely be about +/- 0.1 MOA for a shooter with 1 MOA groups.  At 600 yards, that might be +/- 0.3 MOA.

YOU have to decide if that much error is okay at YOUR distance when it occurs 3 times per 1,000 shots.


P.S. - Unless I am going long (600 - 1,000 yards), I use thrown charges.  That's how I see things and do things.  Who am I anyway, I'm just your average Joe Blow.

What I'm saying is you did not post your results. You posted your opinions.

I didn't want to wrap your whole post in quotes, even tho it is quite well informed.  I do agree with your basic syn, but it isn't always about POI and velocity.

When I load 115 grain 9mm and 230 fmj 45, I am at max with 4.8 grains of TG.  I set my powder measure so it peaks @ 4.8 because I am at the wall, and will run 20 drops to make sure none exceed that.  A proper running 55 will drop that shit to a half of a 10th all day long so I don't give two shits about the under as long as I don't step over the edge.

No powder measure will be 100% accurate 100% of the time, IDK how tuned or slicked up it is.  If you have enough headroom to allow for a 3/10ths overage, or whatever + or - you are comfortable with, no harm done.  

Same goes for digital tricklers that don't cost 1k+

In my reloading, I always, always, always, leave headroom for high powder charges.  Especially for dropped charges but also because, "Margin is your friend."

An example -

Many people told me to use 24.5 gr of VARGET.  During testing I found it was indeed a good load but near the upper range of what I was comfortable with.

As a result of my testing I chose to use 24.0, in part because then even my +3 sigma powder drop (+0.5 gr) does not go over 24.5 gr.

This is why I do not own an AutoTrickler ($1k).  I want one but really don't need one.

I was following everything you wrote until this. Why is it a 3/1000 chance?  If you shot 100 shots out of a machine rest, does it not mean that ~ 4 shots will be in the 99% confidence interval?

That being said, I throw all my short range HP ammo. 75Hornady, 77SMK & 77TMK and don't feel that I'm at any disadvantage. Lots of high X count cleans at 300 yards. However I have a V4 that I use for all my mid-range shooting and before that I was always a trickler for 600 and out shooting and have been happy with my scores.

Easy enough with larger capacity cartridges, but when you have a  window from low to high of only a few tenths total (9mm) there isn't a lot of room to leave.

With rifle cartridges I have always found the "accuracy node" short of max load so have a built in margin for error.  

The idea I was trying to convey was that 3 shots out of every 1,000 (99.7%) will fall outside the +/- 3 sigma limits on charge weight.

If you drop your sample size down to 100 shots, it is unlikely any will be outside the +/- 3 sigma limits (I'm round that 99.7% off to all 100 shots inside the limits.)  As you are shooting the full 1,000 shots, you will, sooner or later, see those 3 anomalous rounds.

Does that make sense?

Yes, my posts have a decidedly rifle cartridge bias.  

I haven't looked at my pistol charge weights in so long, I do not remember what they are, or care.  I say that only because I use that same philosophy - leave some margin, margin is your friend.

Also, I do not use any of the pistol powder(s) that are hyper-sensitive to charge weight.  I use W231 for most charges and don't hot rod, at all.

I had to scratch my head a little bit, but I gotcha now. Thanks.

I won't judge others, but 3 tenths over max is not going to cause harm in a 9mm or 45 acp. The problem I find in handguns is they are much more sensitive to powder charge variation than rifles. For one the charges are smaller, a lot of 45 acp target loads are 5 grains or less. A lot of 9mm are under 4 grains. The difference between 4.0 and 4.3 grains is WAY bigger than 40.0 and 40.3 grains. The other issue is handguns are much more sensitive to velocity changes. 45 acp is ususally pretty forgiving, but something like a smaller 9mm, say a SIG P365, I've seen 4.0 to 4.5 grains change POI over an inch at 25 yards, and that's with hand weighed charges. It can play havoc with your accuracy when your charges vary a lot.

This gets even trickier, because it's not just charge weight that does it. Even perfectly weighed powders vary, some more than others. For example, HS6 in every handgun I've tried it produces higher velocity spreads than a lesser measuring powder, unique for example. Even more bizarre, sometimes high velocity spreads, still produce better groups on target anyway.

There is a pistol powder well known to many here that is very sensitive to powder charge, especially to over-charges.  IIRC, several blown cartridges and blown pistols have been shown over the years and they all used over-charges of this particular powder.

I do not know the powder or its limits.

That would be one to look at very carefully - at what happens if you over-charge a case, the probability of over-charging a case if you throw charges and do that in the context of a lifetime's use of those cartridges - 10,000 or 100,000 cartridges, or more.  Be sure that the +3 or +4 sigma charge weight (or more) does not blow up your pistol.

In all my years on the internet, as full of stories as it is, I have never seen a pistol blown up from a simple overcharge of say .5 grains of powder. 40,000 psi isn't going to blow up a 9mm, a lot of shooters load that intentionally. If you put some of the loads in Quickloads that some shoot to make a 9mm make "major" power factor, some of them come over 45,000 psi, and they shoot thousands of them. I'm sure they load from a powder measure too.  As long as you are loading within a load manual, you will be fine. Don't try to make a 9mm make major.

Clays is a great accuracy powder for me.  I stay away from the top end as it has a reputation for getting very high pressures very fast.  It’s not worth it to me to get close to the spooky behavior.   I am thinking if you load clays to max and then accidentally throw a half grain over max you may have some sporty feeling loads.