Semi-automatic parts weighing/counting machine.

As I said in an earlier post, this part is really mostly a lathe job, but I wanted to do it in the mill to see what the surface finish looked like on the chamfer when finished with a ball end mill.

Here is a picture I snapped while warming up the spindle this morning. This shows how I'm holding the fixture plate in the vice. I machined a 0.125 x 0.125 step in my jaws. This is more than enough grip for this and even more aggressive operations. These are soft jaws, made of cold rolled mild steel. I just make these as I need them, and have several sets in both steel and aluminum made for various projects. I don't even own a set of hard jaws for my vises.

1615677851266.png

This was done with a 1/2 HSS, 4 flute, ball end mill, 300FPM (about 2000 RPM), 13 IPM, 0.020 step down and 0.020 step over, about 0.0015 tooth load. About 1 hour for the finish pass. Could have done it on my lathe in about 1 minute and had a much better surface finish. As you can see, the step down should have been a bit smaller, maybe 0.005 or so, that would have pretty much eliminated the horizontal ridges. But the biggest problem is the chatter. I couldn't hear it like normal, but I could feel the vibration in the machine. You can clearly see the chatter marks in the surface. Maybe the chatter could have been eliminated by increasing the feed speed, and using a shorter end mill, I had about 2 inches of stick out from the collet. It is also possible that conventional cutting would have better in this case rather than climb cutting because it loads the machine a bit more.

Here is the resulting finish, if I had more energy I would have played with it a bit to see if I could improve it. The finish is actually smoother to the touch than this picture would indicate and is acceptable for this application. But not what I would like.

1615678996515.png

Overall it doesn't look too bad, and is acceptable.

1615679145411.png

OK, I just had lunch, so that means it's nap time. :grin:
 
Not a bad finish. You could throw it on the lathe and hit it with some emery paper. But don't get your finger chunked on those thru holes!
R
 
Jim, this is a nice project you are working on and fun to watch your progress. I can't help but notice you have gone from 3 hrs of shop time per day to 4 hrs per day. You coming out of retirement? ;)
 
I think for those profiling ops a small diameter (1/4" or less) short flute carbide endmill is the way to go. Run at crazy speed for smaller steps if needed, though to be honest the "I profiled this on a CNC mill" finish is actually a selling point for alot of consumer stuff (bike parts for example). Even a corner radius endmill would work, as long as the steps are smaller than the radius.
 
I haven't updated this in a while, I have mostly been working on the software which is pretty extensive. If it was just controlling the machine it would be simple, but it also involves parts a parts management data base and other fun stuff.

One of the biggest problems I ran into was the weighing scale itself. I bought a rather expensive counting scale that should have been up to the task, but the some of the parts are so light weight getting an accurate count was near impossible. It really wasn't a problem of the weighing hardware of the scale, but rather the on-board software was just not up to the task. I screwed around with it for about a week before I gave up and decided to just build my own scale and write the software to run it. So the search started ''how to build a scale'', that has about 5 milligram resolution. So after searching and trying a number of different pieces of hardware I finally found the right bits & pieces to make it all work, and more importantly integrate with my software. I tried a couple of Arduino solutions, but the interface was more difficult and required about 3 pieces of interconnected hardware.

So what I finally found was this cute little device from Spark Fun. They say it won't measure in the resolution that I want, but ....... Hold my beer and watch this :grin: It's all about the software and filtering algorithms. With a 1Kg load cell, I have a resolution of 4.78 mg. Right where I need it, the smallest part I need to weigh is 30mg. Powered by the USB cable, so only one little piece of hardware and the loadcell.

https://www.amazon.com/gp/product/B01MG28IZT/ref=ppx_yo_dt_b_asin_title_o02_s00?ie=UTF8&psc=1

1618809410956.png

Connects to the load cell and it outputs raw data to a serial port. Very easy to integrate into any Windows based program. Developing an algorithm for the temperature compensation was a little taxing, but I have it working and the tare is stable within reasonable limits.

I built this little test fixture to mount the load cell.
1618810864270.png

and a side view
1618810928822.png

The load cell was $10 from NightShade Electronics https://nightshade.net/product-category/sensors/

So now that I have the scale working as I want, I'll integrate it into the mechanical design and get the hardware built.

More later.....;
 
Last edited:
I see you use MDF as a one off fixturing method. What are your thoughts on this? In particular:
  • Is it flat enough out of the box, or do you face it?
  • Are threads strong enough in it?
  • Looks like it gets wet with coolant, does this mess up the flatness or strength?
  • Do you re-use the fixture ever?
I'm always looking for good fixture ideas.

Thanks,

Mike
 
macardoso said:
I see you use MDF as a one off fixturing method. What are your thoughts on this? In particular:
  • Is it flat enough out of the box, or do you face it?
  • Are threads strong enough in it?
  • Looks like it gets wet with coolant, does this mess up the flatness or strength?
  • Do you re-use the fixture ever?
I'm always looking for good fixture ideas.

Thanks,

Mike
Click to expand...
1) It is normally flat enough with a very consistent thickness. If needed I will take a light facing cut.

2) I through bolt to the table T-nuts where possible, but MDF takes threads well. Many times I use drywall, deck, or sheet metal screws for hold downs if possible.

3) I'll say that it is somewhat resistant to water based coolant for time periods long enough to get the job done. Seems to be immune to petroleum based coolants. Water based coolants do not penetrate too deep and only seem to affect the exposed surfaces with some local swelling.

4) I have built some MDF production fixtures for holding plastic parts where I'm not using coolant, they seem to hold up very well and are stable with repeated use.
 
You must log in or register to reply here.
Back
Top