Homemade powder trickler Help and suggestions plz

BMorse

Joined Sep 26, 2009
2,675
Yeah, It dawned on me after reading your last post about the commutator brushes, my bad!..

While these scales seem to be quite accurate, I don't know how long they stay accurate - and if they become inaccurate, do they report the failure?
Over time the stain gauges they use (load cells) tend to loose their "tension" and the unit usually self calibrates every time it is powered up, and I was not saying to completely destroy the unit to get to the load cells, but there are 2 points you can tie into to get the output of the load cells without disturbing the rest of the circuits, and it was just as suggestion, I had seen it done another way with a pic32 for measuring food and other stuff for diabetics insulin dose calculations (which is just as important as a good reload) The actual scale is still used to measure the powder and any UC used to control the feeder motor can go off of this as a guide when to stop or slow down. I would actually build in a redundant system that double checks itself, maybe you will have to pre weigh empty cartridge, load the proper amount of powder, and re weigh the cartridge again, and have a Go / NO GO signal or indicators to check for proper fill....

I may not be a "re loader", but I do enjoy my fireworks, and I do build my own mortars occasionally for the 4th, and I do understand the hazards of working with volatile powders (except for the Brushed motor thing, too busy at work and was not thinking properly! :) )....
 

gerty

Joined Aug 30, 2007
1,305
Which model? 550? 650?
I bought the Square Deal first, about 2 years later I bought a 550.
I started out with a RCBS Reloader Special, which I still have and use for depriming once fired military brass with crimped primers.
Do you remember the old "Herter" brand, home of the infamous Herter .401 Powermag.I had one of their presses for a while, didn't like it.
 

SgtWookie

Joined Jul 17, 2007
22,230
I bought the Square Deal first, about 2 years later I bought a 550.
I started out with a RCBS Reloader Special, which I still have and use for depriming once fired military brass with crimped primers.
I have a primer swaging tool that was made by C&H dies. It has a shell holder that screws into a standard 14tpi x7/8" die thread, and the sweging ram is inserted into a standard .30-06/.308/.45 ACP shell holder on the ram. It works quite well. That old Pacific press of mine has an adjustable amount of leverage at the pivot point.

Do you remember the old "Herter" brand, home of the infamous Herter .401 Powermag.I had one of their presses for a while, didn't like it.
I vaguely remember Herters. Dad had a Lachmiller press and primer seater. They've been passed on to my cousin, who decided to take up reloading.

Anyway, back to the topic at hand... ;) don't want to be hijacking our OP's topic.
 

SgtWookie

Joined Jul 17, 2007
22,230
Back to our currently scheduled topic....

It occurred to me that their must be a way to optically recognize the characters being displayed on the scales' LCD screen. This would completely avoid having to disassemble the scale, risk damaging it, and a possible lengthy period of deciphering it's inputs - if they are even available without digging through some sort of epoxy potting.

Processing a video signal input would be too much to ask for an 8-bit uC.

Trying to build a detector for all 5 digits displayed in 7 segments with decimal points from discrete components would be mighty complex.

Anyone have some thoughts on this approach?
 

beenthere

Joined Apr 20, 2004
15,819
We repaired and calibrated electronic scales until they went to uC's, and it became impossible to do anything sensible.

The mechanism may not be a load cell. The ones I got into have an optical sensor that detects the pan at zero as it is floated up in a magnetic field. Added mass forces the pan down. The logic adds current to the field until the pan is back at zero.

The control loop is a simple servo back to zero. The D to A control word also gives the mass added. If it's that kind of scale, reading the display is about your only option.

Even if it's a load cell type, messing under the cover can create problems. It might be possible to use an IA to buffer the signal out to a circuit, but then you would have to scale it with calibrated masses.
 

SgtWookie

Joined Jul 17, 2007
22,230
Thanks, Beenthere - that helps.
I did some fiddling around with "load cells" quite a few years ago, but these were of a stressed-beam capacitor type, and very non-linear. The software that monitored the system had to apply a correction to the readings.

I remember at the time (late 1980s) seeing images of a different type of load cell, which sounds like it must've worked as you're describing.
 

THE_RB

Joined Feb 11, 2008
5,438
What about a "volume based" system? You could use a stepper motor driving a small disk (disk is horizontal) with one or more little holes in the disk. Then a top plate for the powder horn (which is above the disk) and the bottom plate with a release hole (going to a chute etc).

So every time a hole passes both openings it carries a set volume of powder.

The vibration from the stepper motor would provide good feed of the powder from the horn (and good release from the hole later) so it should be a very reliable system and could dispense accurately although minimum resolution would be one "packet" carried by one hole so the holes would need to be fairly small.

Even though it is open loop (ie no weight feedback) if it is reliable enough to dispense an exact powder volume it would do the job very well.
 

gerty

Joined Aug 30, 2007
1,305
Volume wouldn't work because there are several different types of powder.
Ball, stick and flake, to name a few. The different types would require different volumes for the same weight.To give you an idea, some pistol loads would take as little as 3 grains and some rifle loads as much as 70 or more
depending on caliber.
There is 7,000 grains to a pound.
 

SgtWookie

Joined Jul 17, 2007
22,230
The measuring by volume idea is basically how most "powder throws" work. There is a chamber whos' volume can be adjusted by various means, usually by some sort of precision screw arrangement.

Setting the powder throw correctly is an iterative process; you set the throw to an approximate setting, dispense two charges to allow the powder to settle (dumping them back into the reservoir), and measure out the 3rd charge carefully. If it is not correct, you adjust the powder throw accordingly, and repeat the process until you are within 1/10 grain of the desired weight. The reloader needs to be very consistent in how they operate the powder throw in order to achieve consistent results.

After the charge is dispensing at the correct weight, it is good practice to verify the weight of every 10th dispensed charge. If the weight is off, the last 9 charges are then checked.

To automate the dispensing of a charge, it would be efficient to start with roughly 90% or so that could be measured by volume, and then release small amounts of powder at a time with a closed loop feedback in order to stop the dispensing when the correct weight is achieved.
 

gerty

Joined Aug 30, 2007
1,305
The measuring by volume idea is basically how most "powder throws" work
Agreed, but 10 grains of stick powder would have a different volume than 10 grains of ball powder.It would have to recalibrated everytime powder type was switched, that's the point I was trying to make.
 

shortbus

Joined Sep 30, 2009
10,045
The reloader needs to be very consistent in how they operate the powder throw in order to achieve consistent results.
Would it be a easier to build a automated power throw? Maybe a stepper motor and two limit switches? That way your powder weight would always be consistent, after the initial setting.
 

SgtWookie

Joined Jul 17, 2007
22,230
Agreed, but 10 grains of stick powder would have a different volume than 10 grains of ball powder.It would have to recalibrated everytime powder type was switched, that's the point I was trying to make.
You are absolutely correct. Even if the powder is the same, and the load is the same, the charge still must be weighed at intervals to ensure that the weight being dispensed is correct. Any reloader would, or should know this.

If the goal is merely economy, dispensing a light charge that varies a small percentage could be safe. However, for optimal accuracy, and particularly for higher pressure loads, the exact weight must be carefully controlled.
 

SgtWookie

Joined Jul 17, 2007
22,230
Would it be a easier to build a automated power throw? Maybe a stepper motor and two limit switches? That way your powder weight would always be consistent, after the initial setting.
If it were only that simple! :)

You can get pretty close measuring by volume, but if you want the load to be exact, you have to measure by weight.
 
With some experience in handling powder materials, I would suggest making the unit out of stainless steel to avoid corrosive powders and static electricity. Next I would attach a small vibrator to the reservoir to keep powder particles fluid and flowing. This could be nothing more than a 50/ 60 Hz. buzzer w/ amplitude control. The feed tube could be a 3/8" dia. stainless tube sawed at an angle and positioned in the center of the reservoir. Rotating this tube slowly by belt or gear drive and gearmotor would serve as the feed mechanism. Varying the gearmotor RPM, you could adjust the feeds from fast to dribble as your desired powder weight is approached. One last thought is to mount the reservoir with rubber feet to keep vibrations out of the rest of the unit.

Sound possible? DPW
 

SgtWookie

Joined Jul 17, 2007
22,230
With some experience in handling powder materials, I would suggest making the unit out of stainless steel to avoid corrosive powders and static electricity.
Powder measures designed for use with black powder ARE made from stainless steel; so you are quite correct on that point. Black powder is not necessarily corrosive, unless it has absorbed a considerable amount of moisture - at that point it is useless for reloading anyway.
Nitrocellulose-based propellants are not corrosive. Many nitro powders have a graphite coating to aid in dispensing, measure lubrication, and to help with static electricity issues.

Next I would attach a small vibrator to the reservoir to keep powder particles fluid and flowing. This could be nothing more than a 50/ 60 Hz. buzzer w/ amplitude control.
It would have to be of a type that had no brushes or contacts that could arc or spark.

The feed tube could be a 3/8" dia. stainless tube sawed at an angle and positioned in the center of the reservoir. Rotating this tube slowly by belt or gear drive and gearmotor would serve as the feed mechanism.
In my prototype powder trickler constructed of PVC pipe, I used a 1/4" diameter stainless tube that intersects the reservoir at a slight downward angle. There is a small diameter hole drilled crosswise in the tube in the middle of the reservoir, which admits powder at a slow rate.

A gear drive would complicate things and make it more expensive. A simple stepper motor is inexpensive, fairly easy to control in a precise manner, and does not require brushes, contact points, or other spark emitting devices for drive purposes. Most compact gear motors are driven by brushed DC motors that would be unsuitable from the standpoint of safety. A stepper motor could actually perform double duty as a source of vibrations.

Varying the gearmotor RPM, you could adjust the feeds from fast to dribble as your desired powder weight is approached. One last thought is to mount the reservoir with rubber feet to keep vibrations out of the rest of the unit.
The prototype trickler that I built is on a comparatively heavy base (an octagonal metal electrical box) that by some miracle of circumstances, has rubber feet stuck to it. ;)

The "fast to dribble" part is also easily done using a stepper motor and uC.

The real problem being faced at the moment is how to obtain the measurement reading from a scale that apparently only has a visual interface (LCD display) for an output. Once that particular problem can be solved, the rest of it should go rather quickly.

As of now, I don't have a good answer as to how to acquire the data from an inexpensive precision electronic scale that has an accuracy of 1/10 grain or better.

It would be great if such a scale could be found that has a data output port.
 

SgtWookie

Joined Jul 17, 2007
22,230
Continuing on...

It would be a LOT easier to automate the process using a manual scale, such as my Ohaus/RCBS 505 scale. There are no real mysteries to solve there; no hidden signals that would mean voiding the warranty, or other seemingly difficult problems.

After doing a fair amount of searching, I have not been able to find a precision (1/10 grain) scale that has a digital output for anything less than a complete powder measure is offered for.

Anyone else find a scale with that feature that is priced within reason?

Interfacing with a manual scale would really be quite simple in comparison to a digital unit that would have to be dissected and reverse-engineered. It would certainly simplify the I/O to the uC, as well - as the desired weight would be set by the scale itself. Just zero the scale, set for desired weight, and hit a "start" button on the uC.

The most complex part is now reduced to detecting the absolute position of the beam of the scale; and adapting the rate of powder dispensing so that the scale is rapidly balance, and the charge is within the set tolerance.
 

THE_RB

Joined Feb 11, 2008
5,438
Here's a thought. Rather than using a scale which needs to measure linearly within a large range 0-X, why not build a "single point scale" that only needs to detect a single point within the scale travel to say it is at the correct weight?

Imagine a 50:50 ratio balance beam scale, so you put a reference weight on one side and the power will be measured on the other side. Then attach a simple opto or hall sensor to detect the centre position of the beam.

So now you just set up the reference weight on one side, and all the electronics has to do is detect when enough power is dispensed so the scale balances to the middle point.

The main advantage to this system is that you could easily adapt a standard beam scale by adding a sensor. Although it might be better to build it from scratch so you could incorporate a "release the power out" feature too.
 

SgtWookie

Joined Jul 17, 2007
22,230
Here's a thought. Rather than using a scale which needs to measure linearly within a large range 0-X, why not build a "single point scale" that only needs to detect a single point within the scale travel to say it is at the correct weight?

Imagine a 50:50 ratio balance beam scale, so you put a reference weight on one side and the powder will be measured on the other side. Then attach a simple opto or hall sensor to detect the centre position of the beam.

So now you just set up the reference weight on one side, and all the electronics has to do is detect when enough power is dispensed so the scale balances to the middle point.

The main advantage to this system is that you could easily adapt a standard beam scale by adding a sensor. Although it might be better to build it from scratch so you could incorporate a "release the powder out" feature too.
Well, there's a problem with the approach, in the ratio of the weight of the moving parts of the scale itself, and the powder that's being made. The scale beam has a fair amount of mass in order to be rigid enough to give consistent and accurate measurements over a reasonably broad range; from 1/10 grain up to 500 grains. The beams are usually made of metal. One of mine has a brass beam and a cast iron frame; the other has a plastic frame and a chrome-plated beam.

The mass of the beam is enormous compared to the mass of a small pistol load. The beam and weights might weigh a couple of ounces, or 1/8 of a lb. A standard starting load for using Bullseye flake powder for a .380 auto is 2.3 grains. As there are 7,000 grains per lb, that is 1/3043.478th of a pound! The charge is less than 0.27% the weight of the beam. It is not hard to see why precision manual scales take time to respond to such tiny weights.

If the inertia of the balance beam were eliminated, then what you propose would be nearly do-able. Except that there will always be some delay or lag between grains of powder being dispensed, falling into the scale pan, and the result balanced - if only due to the inertia mass of both the powder and the balancing weights.

With the electronic scales, the only lag is caused by the internal feedback loop, the mass of the platform itself and the mass of the object(s) being weighed. I don't know the weight of a typical compact platform, but I can imagine that it's pretty small.

So being stuck with a manual scale, the progress of the beam will have to be tracked.

Just to give you an idea on how sensitive these powder measures are...

I was working up a test load on the reload bench in my garage a few months ago, and I was having trouble getting the scale to zero. It seemed to wander up and down a couple of tenths of a grain of it's own accord. There were no air drafts, no fans on, no power tools running, no street traffic - I couldn't figure out what was going on, until I discovered a very small ant walking back and fourth along the back side of the balance beam!

That's sensitive.
 

gerty

Joined Aug 30, 2007
1,305
I was working up a test load on the reload bench in my garage a few months ago, and I was having trouble getting the scale to zero. It seemed to wander up and down a couple of tenths of a grain of it's own accord
You'll also get the same problem with a little bit of dust on the pivot bearing surface.
 

Thread Starter

Jacob J

Joined Jun 18, 2009
159
Hello again

Ive been away for a while and I am now getting back into reloading and I ran into the timeconsuming process with weighing powders, so then this topic got into my mind.

I have decided to take apart my scale and get going with this. I have decided to use a computerprogram, so I need to connect the scale to my computer via serial port or something like that. Then I have decided to put a brushless motor on a standard powder trickler to simplify the buildup.

If I take my scale apart and post some pictures of it in here, can you then help me to get it connected to my computer?

ive found this http://micah.navi.cx/2010/01/hacking-a-digital-bathroom-scale/

and maybe it can help

Hope you still are out there and still as interested as I am in this project.

/Jacob
 
Top