Automatic powder trickler

First off, let me say this is a great forum and I have been lurking for a few days (drinking from a water hose). Secondly, please prepare yourself for the most new newb.

So I am a ME by trade, preferring to steer clear of any wires containing "magic smoke" or 0's & 1's....that is up until a few days ago when I was introduced to Arduino. Immediately a project popped into my head, something I have been wanting to do for a long time just never knew of a realistic way for simple me to do it. This is also about the only thing that has ever given me motivation to work on this type of hardware/software.

I did find a thread from 2012 where a guy I have actually spoken to (memilanuk aka Monte), was working on doing the same thing for the very same reasons but I am not sure how far he got. I'll PM him now that I have registered.

Nonetheless, here goes:

I have a A&D Fxi300 scale http://balance.balances.com/scales/751 as well as a powder trickler that runs off of 4 AA batteries. I haven't taken it apart but believe it's just an electric motor/eccentric mass set-up. It works very well.

My pie in the sky end result would be something similar to the following; I use three reset style buttons to set the target weight (10gn, 1gn & .1gn) to say 43.6gn for example. I would bulk feed ~.5gn low, directly onto the pan & scale. The Arduino would have logic that would say ~IF pan weight >/=98.5% of target, run trickle loop/algorithm/PID, whatever until target is met via PWM output to the trickler. The process would then rinse and repeat. Even if this took 20-30sec, allowing me to seat the bullet for the previous round while the Arduino did it's thing would really speed the process up.

Some of my arseumptions:
-I won't need to output anything to the scale (at least initially) as I normally tare the pan before starting my reloading process and it very rarely drifts. I did look over the command list for the scale and can see that it "should' be able to receive parameters from the Arduino if/when I get to that point. Currently I set all of this before I start, no biggie. I give the scale 24h to warm up anyways.
-I can modify a 232 plug wiring so that I can connect directly from the scale to the Arduino RX/TX/GND...or is via USB the safer method? I do NOT want to fry this scale!
-Initially I would just use my PC to set the target weight.
-Later, I would need an LCD if I want this thing to be stand alone w/o the PC to visually set the target weight as I toggle up to it.

Obviously I am quite ignorant to all of this, but definitely am now not afraid of jumping into to learn. My plan was to buy a kit of some sort, do the example projects to get my feet as wet as possible, then hopefully begin this trickler task.

So if you have successfully made it through all of my rambling above, is there a "kit" you recommend that has most if not all of what I need as well as a vendor? I ask this because I am seeing some knock off boards and/or junk components. Concerning the LCD's, I have read most kit versions require a lot of the Arduino's resources to run however Adafruit has an LCD that supposedly only requires two wires which could be shared in the right application. When I read the description, I get lost in some of the lingo.

Thank you in advance for your patience with me as well as any guidance!!!

-Jarrod

mouzetrap:
First off, let me say this is a great forum and I have been lurking for a few days (drinking from a water hose). Secondly, please prepare yourself for the most new newb.

I have a A&D Fxi300 scale A&D Weighing FX-300i milligram balance Balance - Precision Weighing Balances as well as a powder trickler that runs off of 4 AA batteries. I haven’t taken it apart but believe it’s just an electric motor/eccentric mass set-up. It works very well.

I read the users manual for your scale, the interface is RS232C, You will have to add an interface circuit between the Arduino and the Scale, Just a simple ‘shield’ to convert from the 5v to the +/-15v like this one: RS232 Shield this shield even has an open area you could use to add the shaker motor drive.

If you use a PC to control the Arduino, that would be about all you would need.

To make it a self contained device, A character LCD, LCD - Keypad Shield

mouzetrap:
My pie in the sky end result would be something similar to the following; I use three reset style buttons to set the target weight (10gn, 1gn & .1gn) to say 43.6gn for example. I would bulk feed ~.5gn low, directly onto the pan & scale. The Arduino would have logic that would say ~IF pan weight >/=98.5% of target, run trickle loop/algorithm/PID, whatever until target is met via PWM output to the trickler. The process would then rinse and repeat. Even if this took 20-30sec, allowing me to seat the bullet for the previous round while the Arduino did it’s thing would really speed the process up.

That is not a very complex requirement. A simple UNO could run it without to much effort.

mouzetrap:
Thank you in advance for your patience with me as well as any guidance!!!

-Jarrod

Looks like you are a shooter. One of the projects I am working on is a reactive target system that responds like the Army KD targets. It raises and lowers the targets, detect hits, networkable with upto 16 targets mechs per segment. It uses Foam targets so no ricochets like Steel targets. All of the exposed parts are disposable, Target Arm, Target with integrated sensor. I hope to have them available for sale this fall.

Chuck.


Check out my Kickstarter Project Memory Panes an expansion RAM Shield for Mega2560’s. It adds 1MB of RAM for those projects where 8KB is not enough.

The problem with direct PWM is that it assumes instant feedback. However your system will have a delay between the powder leaving the spout and landing on the scale. The scale itself will also introduce a delay because it's probably not outputting 100 readings per second.

You will have to characterise the response of the trickler and maybe plan to stop it when you are 0.1 grams away from your target.

For your application, the LCD is most of the work. Getting the motor working and stopping at the right point is relatively easy. The user-interface is going to be the biggest part of your code.

WOW! You guys are great! Thanks for the quick and detailed responses.

ChuckTodd, thanks for the links. I think for now I will just plan to use my PC as the "simple" stuff seems daunting to me as it is. I am sure I will have a mountain of tweaking to do to get the tuning right once the set-up is working properly.

Yes I am an avid shooter (obsessed is probably more accurate)!. Everything I do is KD, specifically FTR so 600->1,000yds. Your project sound very sellable! I am not sure exactly how the end product will be but I can tell you for me, when I practice, I have to either use DIY shoot-n-see's or steel so I Can see my hits. The shoot-n-sees are only good out to about 600, maybe 800 if the mirage is calm. If there was a way for hits to be plotted, say on my phone or some other device near, that would be the cat's meow. I guess this could be done mechanically, depending on how the targets are sized and meshed together. Nonetheless, cool project and please keep me in the loop!

MorganS, you bring up a very valid point...hadn't thought about the delay. Stopping .1gn short, then sending a single kernel burp to the trickler, allowing time for gravity, scale response/stable, scale output, arduino process, etc sounds feasible & repeatable as I can manually trickle one kernel fairly consistently just using the low speed button on the control pad...purely time based ON time for this should work.

Thanks again guys!!!

mouzetrap:
WOW! You guys are great! Thanks for the quick and detailed responses.

ChuckTodd, thanks for the links. I think for now I will just plan to use my PC as the "simple" stuff seems daunting to me as it is. I am sure I will have a mountain of tweaking to do to get the tuning right once the set-up is working properly.

Yes I am an avid shooter (obsessed is probably more accurate)!. Everything I do is KD, specifically FTR so 600->1,000yds. Your project sound very sellable! I am not sure exactly how the end product will be but I can tell you for me, when I practice, I have to either use DIY shoot-n-see's or steel so I Can see my hits. The shoot-n-sees are only good out to about 600, maybe 800 if the mirage is calm. If there was a way for hits to be plotted, say on my phone or some other device near, that would be the cat's meow. I guess this could be done mechanically, depending on how the targets are sized and meshed together. Nonetheless, cool project and please keep me in the loop!

Currently for impact location I use a small camera module, to relay a jpeg back to the shooting stand. When this mode is engaged, the target is dropped forward and the camera module mounted at ground level snaps a shot upwards of the target. The image is compared with the previous image and differences are highlighted. It's not perfect, but you can usually tell where the impact was.

Chuck.


Check out my Kickstarter Project Memory Panes an expansion RAM Shield for Mega2560's. It adds 1MB of RAM for those projects where 8KB is not enough.

Hi there,

How did the project end? I’m looking to do something similar.

I’m looking to drop powder charge out of my lee powder dispenser,

Could you guys help me with the code and how to wire it please?

Don't know about the OP's project... but it'd be pretty hard to beat Adam Macdonald's AutoTrickler and AutoThrower combined with an A&D FX-120i scale. Unless, that is, you have a spare $5k for a Prometheus...

FWIW, Adam started pretty much at the same point... he's just taken it a tad further, with 3D printed trickler base and controller housings.

It's been wildly successful... enough so that there is somewhat of a cottage industry of 'upgrade' parts over @ Area419.com for people (like me) who prefer shiny billet aluminum with a bit of heft to it, over 3D printed plastic.