Powder dispenser using balance beam

Status
Not open for further replies.

ottsm

Member
Joined
Sep 9, 2012
Messages
78
I've been working on a project to make my own powder dispenser. It's just for fun, I could have probably purchased a PACT unit by now considering I already have the electronic scale. Of course I have had mixed results with electronic scales, they are getting better but cost more and I end up using the balance beam to check with anyway.

I've attached a picture that shows my prototype. It's just some pipe fittings, pressure test plugs, brass tubes, motors and couplings, two IR slot sensors, and an Arduino board from DFrobots. The board from DFrobots has the H-bridge on the circuit board to control the speed of two DC motors. Normal refills are under 20 seconds.

Currently I'm using a Redding scale. I have a Ohaus 1010 that has the approach to weight feature that should reduce some of the overfills that I have now, the current setup uses a "learned" fast fill time. On the first few refills it learns what the fill time needs to be and trickles the remaining amount. Sometimes this changes and it overshoots or takes longer. Another guy on youtube has a better setup (search for "Auto Charge Powder dispenser. Balance beam based"), this gave me most of the ideas, his setup is the best one I've seen so far. With a RCBS or Ohaus 1010 you can eliminate the Arduino controller and just use a few transistors to latch the circuit in. If you use a scale without the approach to weight, a simple circuit will not work, too much powder is in free flight when the first slot sensor is reached without the approach to weight feature, timing is the only way to do it without the approach to weight. It is possible to add your own approach to weight system, search on youtube for "502 scales accurized RCBS" to see how this is done.

Anyone else working on a similar project and want to share ideas? I wish PACT would make an option to use a balance beam, I tired to get my hands on a used one that I could modify, It's basically the same dual tube concept.

IMG_2068_zps896d14e5.jpg
 
Excellent job - A good sensitive reliable beam scale is a joy to use. I'm pleased you like my idea of the gravity approach to weight system - it works well with any of the auto trickler systems.

Here's a more refined version, just a step on from the bent paper clip. :)
 

Attachments

  • 20130111_150154.jpg
    20130111_150154.jpg
    62.9 KB · Views: 60
  • 20130111_150130-1_zps9f25e42d.jpg
    20130111_150130-1_zps9f25e42d.jpg
    15.3 KB · Views: 62
  • 20130218_181034_zps69bdc82e.jpg
    20130218_181034_zps69bdc82e.jpg
    68.2 KB · Views: 62
  • 20130218_180742_zps8e761db8.jpg
    20130218_180742_zps8e761db8.jpg
    80.8 KB · Views: 66
Really cool. Quite a bit of ingenuity applied and a well thought out project. Kudos to both of you guys, ottsm and 1066.

I do have a question. I see you both went with the two motor design and understand why. I did watch the video which was also quite good. My question is why did you opt for the two motor design rather than a single motor running at a few speeds?

Again, kudos on a sweet couple of projects.

Ron
 
Ron - Sorry to disappoint you but I can't take credit for either of those twin tube powder dispensers.

I was commenting on the approach to weight system, first shown in my 502 scale tuning video. Here

The twin tube dispensers, as far as I know, first appeared in the 1980's, long before there was any microprocessor type control, just basic transistor technology - the system, given you have a good beam scale at the heart of it, works extremely well. A Prometheus dispenser, at xK$ is basically just a good quality beam scale, a measure to dump the charge with and an optical switch to cut the feed off at the right time.

Here is the patent for the original twin tube dispenser, dating from 1986.
 
I did try a single screw design before I started seeing other attempts by others on the internet. I took a standard Redding powder trickler and attached a stepper motor to the existing aluminum threaded tube. If you measure the inside diameter of this trickler I believe it was 1" and with a little bit of sanding I concluded it would accept a standard 1" OD clear tube with a little sanding of the tube (although I never tried). I had a foam coupling to attach to the knurled end and the stepper motor. Nice thing about it was I didn't need a geared motor and it was direct coupled. At the time I didn't know about tilting the entire assembly to speed things up. My stepper motor didn't want to go fast enough either. It did work but was slow. The last attempt before going to the dual screw was with a better coupling and I replaced the Redding tube with brass tube of my own. This did help with the alignment issues. It could be accurate also but by that time I started looking at the dual feed designs and going to DC motors.

One other thing I'd like to try but it wouldn't be easy. If I could get a linear optical encoder strip like they use on printers and attach it to the balance beam, I could increase the resolution and know how much under or over shoot occurred and make fine adjustments automatically. I've looked into this, however, the standard encoder sensors don't have wide enough of a gap for the play that is in the scales. I may try making my own A&B sensor pickup. Only problem with increasing the gap is that the resolution starts going down.
 
Thank you both for the replies. While I had given some thought to playing around with a small single point load cell I had never given much thought to improving and automating a beam type scale. Ottsm, if you can find yourself a linear encoded strip with a reader that may work out well. A big part will be finding a strip at a reasonable cost and with the resolution you would want/need.

Some time back I went into a RCBS digital scale I had out of curiosity. the scale is about 20 years old so nothing fancy like a micro-controller. Fooled me in how the thing worked. I assumed a strain gauge single point load cell which there was but then things changed. They amplified the voltage off the load cell and passed it along to a V/F (Voltage to Frequency) converter rather than an A/D configuration.

RCBS%20Strain%20Gage.png

Note how the part number of the strain module is visible. BCL-300GM which got me to this data sheet. The remainder of the guts look like this:

RCBS%20Circuit%20Card.png

The nice thing is that scale is 20 years old. Today's micro-controllers make much easier work of building a powder dispensing system. Single point load cell modules seem to be available as they pour in from China with inexpensive pricing. I may start kicking the idea around again.

Anyway, thanks for sharing what you guys have been experimenting with. I won't drag on with the techie stuff. :)

Ron
 
I have seen the videos on the Omega unit. I do like the photo sensor setup as far as using it without modifying the scales. Of course one would need a dual sensor to fully automate the process. The vibratory type conveying allows one kernel at a time. With a thick wall 1/4" brass rod rotating sometimes two or three kernels drop off, normally I just scrap the charge and move on. I do like the video where he shows using the analytical electronic scales, but with the price of one of these units one could purchase several nice rifles and you still need to integrate with them to automate the process (some do have communication capability but it's not easily done without getting into more involved programming).

I do miss using my old Belding&Mull powder dispenser, it seems to be getting less use these days. I guess you can now get parts for these at;

http://www.beldingandmull.com/visible_powder.html

I might get the brass powder baffle. Hard to beat the Lyman 55 though.
 
Here's an Omega auto-trickler being used to good effect - He claims quicker than a chargemaster.

https://www.youtube.com/watch?v=NmFdmrz7njg&list=UUgG6OmZ2trXuSr7dvaSWQAw
I wish he would have had the trickle more in the foreground. The irony was his beam scale is the same one I have. :)

The method using a trickle will be faster than the Charge Master or similar because we start just a grain or two below what we want so we throw most of the charge then sort of top it off with a trickle.

The RCBS Charge Master uses the motor at 4 speeds.
Key ChargeMaster Parameters with Default Settings
HSB_A1 (15.68) Grains under target weight to go from full to high speed
HSB_B1 (3.42) Grains under target weight to go from high to slow speed
BSP_C1 (1.08) Grains under target weight to go from slow to final trickle speed

So we start with an empty pan so getting to weight takes time. I guess it depends on how quick we want to get where we are going. :) We could maybe use the powder dispensers like the Charge Master in similar fashion. I actually have a Charge Master that was a gift and I have yet to get it out of the box (been about a year now). I need to finish gutting what will be the new gun room and get things setup. :(

Ron
 
The Lyman M5 scale that is being used in the video, is IMO, one of the best reloading scales ever made. It's the forerunner of the 10/10 but more sensitive. Other scales I like are the RCBS 5-10 and the RCBS 304.

All now unfortunately obsolete. These old scales were made by engineers, all the current scales appear to be made by accountants.:mad:
 
I've never used a chargemaster but I've read the claims, something like 42gn of 4064 in about 12 seconds. Personally I like to weigh all my charges first, take a flashlight and look to make sure I didn't miss or overcharge any of the cases, then seat the bullets, it's the way I've always done it to avoid mistakes. If your loading powder and seating the bullet then 12 seconds isn't going to matter.

I cast a lot of bullets so a lot of my loads are like 16gr of 4198. The fast fill can dispense that in a few seconds and the scales not going to need stabilized. This is were an electronic scale would be advantageous, I've seen a few others dump directly from any old standard powder measure onto the scale, then the trickler is off to one side and top it off (this assumes you trust your electronic scale).

http://www.youtube.com/watch?v=dujvwUtmrFA

The first outfit to make a good electronic scale with a setpoint and output connection would have my interest, especially two setpoints and two outputs. The Omega would benefit from this as well. I still like a good old balance beam to verify things.
 
This is a good read on the RCBS Chargemaster. What is interesting is how they use the motor at 4 speeds (as required). The programming apparently looks like this:
Key ChargeMaster Parameters with Default Settings
HSB_A1 (15.68) Grains under target weight to go from full to high speed
HSB_B1 (3.42) Grains under target weight to go from high to slow speed
BSP_C1 (1.08) Grains under target weight to go from slow to final trickle speed

As I understand it those numbers can be tweaked as they mentioned in the link.

I really like using a beam scale combined with a trickler which is automated.

Ron
 
Status
Not open for further replies.
Back
Top