Automated Pellet Sorter

Jump to latest Follow Reply
Smitty911

Smitty911

Member
Aug 4, 2018
703
494
Orangeburg, SC, United States
I've been way to busy at work and haven't been around much, trying to catch up.

On the plus side I've been doing some work on a little side project and it's just about completed. The software is done and working pretty darn well. The 3D printed parts will be here next Friday. I'm optimistic that I can get everything together and working within a week after that and do some of my own testing.

So I'm curious if anyone else would be interested in one. Possibly providing some feedback on the test unit as well.

The pellet sorter will weigh each pellet and sort it into one of seven (7) trays. Tray #4 is for Nominal weight or weight on the Tin (of course it's adjustable). Tray #1 and #7 will be for those pellets that are to far over or under the nominal weight.

One item that need to be locked down, significant digit. We'll be reading in Grains or Grams, I was thinking 0.1 gram so you'd have +/-0.2 grams or 0.1 gram per bin.

Suggestions?



Smitty911 
 
  • Like
Reactions: CampFussell
Smitty,

As a person who is constantly scrutinizing pellet quality, particularly their weight, I would use grains, it would be a finer unit of measure over grams, which is not used too much in the airgun world. 

When I weigh my pellets, I have a scale that weighs in .02 of a grain increments. I weigh primarily .177 caliber pellets. My scale would read 8.40, 8.42, 8.44...etc. In my world of seeking pellet perfection, I would use the above 3 weights (for an Air Arms 8.44 grain pellet), anything higher or lower than that ends up in the practice or plinking bin.

I've found with some big name manufacturers, anything from 8.08 grains, all the way up to 8.74 grains in the same tin.

This is also why I am experimenting in making my own, that can be followed on the pellet and projectile part of this forum 

Tom Holland 

Field Target Tech 
 
Upvote 0
xbowsniper,

The software can be changed to either grams or grains. It shouldn't be an issue sorting to 0.02 grains, I'll have to check, consistency and repeat-ability will be the driver. 

One of the issues that had to be resolved is the design of the load cell, strain gauges, placement, etc. etc. Getting repeatable numbers isn't as easy as one would think. Thankfully, I work full time in a Independent Test Lab, so I have access to some pretty precise instrumentation as well as some really smart people (I wish I was one). The load cell will only handle 100 grams, to handle the sensitivity required to measure into the third decimal place. I'll have to go back through my number and check them.

Smitty
 
Upvote 0
This would be great. Maybe you can combine it with a pellet size measurement also by having a pellet head sized hole on the scale surface itself. You could have a replaceable scale surface with various sized holes, or have several measurement surfaces lined up next to each other. Are the pellet going to be placed manually on the scale surface?
Click to expand...

I haven't gotten to sizing yet, there seems to be several factors that would need to be worked around, like head size vs skirt size. Like I tell my boss when they quote some pretty complex test, "Anything is possible with enough money thrown at it." I have an idea about it based on what I have seen Tom doing. A piece of glass at an slight angle and allow them to roll, the ones that have a consistent arc have the consistent diameter, the ones that shift left (starting with the front facing down) have a larger skirt, ones shifting right have a bigger head. The trouble is Lead is sticky, so coming up with a way to ensure all the pellets have the head pointed down and continue to feed is an issue to work out. I've seen fastener loading for rolling the threads, but vibration sorting and the diabolo shape don't make that easy, yes, I've been playing with it.
The ultimate rig would lift the pellet around the waist, measure it, hold it over a imaging system, measure the head, rotate 180º measure the skirt, place in one of 100 bins ;)

Smitty
 
Upvote 0
Smitty:

I was referred to your post by a top shooter friend of mine, I don't participate in this forum...

Weigh sorting is directly proportional to pellet caliber and gun power, by this I mean that when sorting .177's a difference of .02 grains is desirable and mandatory if seriously competing USARB or similar matches and specially with the LV class where 8.4 gr. pellets are shot under 800 fps. (12 lb.)...

Now, if the shooter is using a .22 or a .25 shooting 18.13's - 25.39's and even higher mass pellets at velocities ranging from 860 to 1000 fps, a .05 grain sorting is more than appropriate...Some people may argue that if the system allows .02 gr. accuracy then to use this accuracy for heavier pellets than .177's...I agree with this concept, but if you are tied to 1 specific mass +- 3 (up-down weighs) as you mention, then those 7 or 8 trays would not be even close to what you would be getting ...

When I weigh .177's, for the most part I get a normal curve distribution which means about 8-9 different trays for 95% of the pellets and around another 4 trays for the remaining 5% of the total number of pellets...I ALWAYS get a scrap pellet tray that accounts for at least 2-3% of the total pellets I weigh...

When weighing .22's, using .05 gr. increments will usually give you 10 different trays distributed in a normal curve as mentioned...If I were weighing .22's in .02 increments I would end up with 22-24 different trays and this is impractical and not necessary because I shoot high velocities where a .02 gr. difference is not relevant, specially at the distances I compete (40,50,60 and 70 yard metallic silhouettes)...

Weighing .30 caliber has usually given me 8 different trays (.05 gr. difference) which proves that the larger the caliber the more consistent the mass of the pellets....05 gr. increments for .30 caliber at the usual 850 to 975-1000 ft./sec. and shooting EBR or RMAC at 75 and 100 yards (paper) makes absolutely NO DIFFERENCE in accuracy...I actually have shot many .5 to .750" groups with 1 grain difference lots and shooting unsorted pellets directly from the tin I have found that the groups easily remain around 1 inch if I do my part...So unless I am going to seriously compete, I don't weigh my .22's and .30 cal pellets but I weigh ALL of my .177's even for practicing.

I would be interested and glad to test your system and help you in developing something good and reliable, you can check who I am and my credentials as a shooter and as a professional engineer within the USARB ring of shooters...

Best regards,

AZ


 
Upvote 0
Smitty,

You are mostly correct, just one correction regards to the rolling method that I use. Above you stated that I roll them to determine the headsize, which I do not, just to clarify. 

Before I roll them, I measure headsize first. I use a Pelletgage that measures exact headsize. This insures that all pellets have the same headsize when I begin the rolling process.

As I roll them, I'll determine where the acceptance area is, usually a mark at the bottom around 3/8 of an inch margin. The pellets that fall within, or touch those lines, are keepers. The pellets that end up to the left of the left line, have skirts too large as compared with the acceptable ones. The ones that end up to the right of the right side line, have skirts that are too small compared with the acceptable ones. I use this method only to insure consistent skirts in relation to headsize. 

I do like where you are going with this, I'm intrigued......

Tom Holland 

Field Target Tech 
 
Upvote 0
Smitty,

You are mostly correct, just one correction regards to the rolling method that I use. Above you stated that I roll them to determine the headsize, which I do not, just to clarify. 

Before I roll them, I measure headsize first. I use a Pelletgage that measures exact headsize. This insures that all pellets have the same headsize when I begin the rolling process.

As I roll them, I'll determine where the acceptance area is, usually a mark at the bottom around 3/8 of an inch margin. The pellets that fall within, or touch those lines, are keepers. The pellets that end up to the left of the left line, have skirts too large as compared with the acceptable ones. The ones that end up to the right of the right side line, have skirts that are too small compared with the acceptable ones. I use this method only to insure consistent skirts in relation to headsize. 

I do like where you are going with this, I'm intrigued......

Tom Holland 

Field Target Tech 
 
Upvote 0
Strever,

Yes, you are correct........Harry is the godfather of pellet sorting.

I have a way to accurately measure headsize, but not the skirts. By doing the rolling process opposite the way Harry does it, I can compare the skirts with the proper headsize, not the other way around the way he does it. If you have an accurate way to measure either, you can, by rolling determine the other.

Tom Holland 

Field Target Tech 
 
Upvote 0
AZ,

Thanks for some clarification. 0.02 grains accuracy and repeatability are possible, but I'll know more once I get the 3D printed parts in my hands, that's what the testing is for. I measured several tins of 0.25 and 0.22 and my spread was somewhat larger than what you mentioned. Of course there are size and speed limits that keep this thing from taking up a whole table. I could narrow up a couple things and get maybe 11 bins. Two for over and under nominal +/-0.08 grains, one nominal and 0.02 spread on the remaining eight (8) bins. Possibly have the software ask caliber and have the software adjust the 0.02 for 0.17 and 0.05 for 0.22, 0.25 and 0.30, it's only bad spelling and punctuation at that point. If more bins are required, the size and shape of the current model would have to be changed 180º but I only love what works.

I'm finding out once again that development is expensive ;) Quality 3D printed parts, buying one of this and one of that to find out it didn't work the way you wanted it to. Good times.

I'll have the 3D printed parts here on the 9th, the assembly and software will need to be tested. I may have to make a silicon mold to duplicate the parts to make 5 to 10 test units for people who know things to test. Once the feedback is received and most people are happy with it, I'll see what Injection Molding tooling will cost and the price point of the units.

Any guess on what someone would be willing to pay for one, it's about a 2 second per pellet cycle, so a 350 count tin is 50 minutes. Might be able to speed that up some. If there is enough interest, the next item would be to load a bin and have it put one pellet in the sorter. That would make it so you could walk away and come back instead of manually loading it.

Open to ideas.

My other product is getting close to going to tooling now, Patent is applied for, design is 95% done, software will be completed by this weekend. I'll be launching this product at the SEMA show in Las Vegas this November, so if any of you are going stop in my booth and introduce yourself.

Smitty
 
Upvote 0
One concern is the materials temperature stability. Even a slight change, expansion and contraction will throw off your measurements enough to invalidate your sorting results.
Click to expand...

John,

I work at a test lab, I'm very familiar with everything from -150º C up to +750º in some of the test I've performed over the last 15 year here. I don't think the temperature differentials will be a big enough deal to "invalidate" sorting. Most people will be doing this at home in a semi-climate controlled area where everything has thermally stabilized to ambient prior to sorting.

So if someone measured everything at 23º C and then took them outside to 10ºC and allowed everything to stabilize and resorted them, you should get the same distribution. I could be wrong, I've been divorced so I'm capable of mistakes. ;) There is some thermal dependency on the load cell as they are aluminum, the LCD display won't like really cold weather either. I have several environmental chambers in my Lab, so one I get this thing working well, I may throw it in one and resort and see. I'm not sure how you would compensate for it without adding a temperature/humidity sensor, and compensation software which while possible might not be worth the money.

I'll kick it around some and some testing and see if it's a factor.

Thanks for the suggestion.
Smitty
 
Upvote 0
No worries. Though can you guarantee results between say 60f-80f? Not everyone will be sorting indoors let alone in a climate controled area. A .01mm variance in size up or down can for precision shooters be the difference between a 9, 10, and 10x. So if your sorting for 4.52, and in reality they are 4.53 it can be an issue. 

I'm not trying to put your idea down or disparage your work. Though being a picky shooter, and knowing others who loose thier poop unless everything is perfect or as advertised my suggestion is only a heads up.

On the other hand if everything is stable and works properly I would keep the machine for myself and charge $50 for 500 perfect pellets. Seems more profitable if all you need to do is fill the hopper and walk away.
 
Upvote 0
You must log in or register to reply here.
Top Bottom
Back
Home
Forums
Resources
Menu