Friday, May 12, 2017

Machining The GB-22 Is Done - Pic Intensive

Just an update on the progress on the GB-22.  A really good question is, "what's taking you so long?" and I don't have a good answer except that I'm not a real machinist, I just play one on the Intertoobs.  It's taking me time to try to decide how to do things so that I break the least amount of parts.  Hopefully, none of my work holding schemes fall apart and send dangerous projectiles around the shop.  Also hopefully, I don't break delicate tools.  Well, I achieved the first one. 

My concern is that long "diving board" spring that forms the trigger.  It's a piece of steel about 1-7/8 long and 1/10" thick - that ends in a big, chunk of metal 1-1/8 by 3/4 (the trigger).  I didn't want that to be unsupported at any time.  In my last update, I said I'd cut four passes: (1) a 3/8 wide end mill (EM) to remove most of the metal, leaving a thin skin on every surface, (2) a 1/4 EM to get final sizes and shapes, (3) a 1/8 EM to cut the slot and back of the trigger, and (4) a pass to cut the slot in the frame for the trigger to move. 

After a lot of consideration, I thought I should cut the 1/8 wide slot first. Nothing unsupported here.
For orientation, that's the top screw in the grips, and the four things in the upper right are 1/8" spring pins.  I broke the first of two cutters here.  I was cleaning chips out of the groove and bumped the emergency stop button for the spindle.  It took me a couple of seconds to realize what was going on and hit the start button again, and when I did, I think it broke a flute off the 1/8" EM.  A little while later, the cutter snapped off, and while picking up the pieces noticed that the flute was broken off.  I was able to change bits and get going again, with just a minor delay to re-zero the Z-axis. 

Next I questioned why I should bother with the 3/8" EM pass at all.  I spent a bit of time with my feeds and speeds calculators and my CAM program getting different tool path to compare and seeing what difference it makes.  Cutting the whole thing with the 1/4" bit is slower than just the 3/8, but the 3/8 is followed by the 1/4 and the combination doesn't really seem to be appreciably faster than just going with 1/4", so that's what I did. 
This is one pass along the top of the frame with the 1/4"EM fairly deep; right to left at this point.  At the end of this operation, the extra steel was removed.  This is a milestone!  And it freed up the raw material for the next GB-22 frame.
(note my high-tech cover to keep the exposed ballscrew from getting coolant spray and chips on it - a ZipLoc bag and duct tape)

Finally, it was time to cut the most delicate cut, that 3/32 wide slot in the frame.  Like I said, this cut made me nervous, and I thought the way to fix that was to add a threaded hole in my tooling plate and add one of the small clamps from my Sherline.  I drilled and tapped a hole for the cap screw, that ordinarily rides in a Tee-nut on the Sherline.
With that in place, I needed to stick the 3/32 EM farther out of the holder than I'd like, more like 3/8" than 1/4".  This was done with a simple file that just went to points at the start and end of the slot, advance the cutter from start to end, raise the cutter above the work, go back to the start, lower it to the next level and do it over.  When I completed the cut, it was at the wrong depth, and while trying to figure out why it hadn't cut to the proper depth so I could fix it, I broke that cutter.  I was able to complete the cut and with it, machining of the frame.
There's some work to be done before I can assemble it: some file work on the seer to get the shape right, some corners to square up here and there, minor stuff.  

A minor word on software.  This is a software intensive hobby, between CAD, CAM, the Machine controller and so on.  I mentioned a "speeds and feeds" calculator that I'm using and I'm impressed enough with it to recommend it.  Speeds and feeds refers to the RPM of the cutter (or work, in the case of a lathe) and the rate that the axis motors feed the work into the cutter.  This is a recurrent problem for all sorts of machine work and has generally been solved with tables, graphs and look-up data of all kinds.  I had been getting pitches from CNC Cookbook for a while for their GWizard software and always put it off, thinking it was too fancy for my Sherline, hobby-class machine.  When I got the big mill going, I figured it might be time to take a more critical look.  They offer it on sale regularly and I picked up a copy of both the GWizard and the GCode Editor as a package last time.  It's on sale again for the next week, so you might want to check it out.   If you use a spindle motor of 1 HP or less, it's a one time purchase.  If you want to enter higher horse powers, you'll have to buy annual (or life) subscriptions.  (No financial connection or interest, yada yada)


  1. Lookin' good!

    Some of my earliest memories in the shop are my Dad teaching me about "Speeds and Feeds".

  2. Damn, that's a lot of work.

  3. Making good progress and advancing along the learning curve of breaking end mills means learning how not to break the end mill next time.

    "a ZipLoc bag and duct tape" I think you meant to say, "a polyethylene flexible moisture barrier secured by scrim backed pressure sensitive tape" or maybe it's just that I worked for the government for way to long.

    1. I'll stick with duct tape. It needs no justification. Duct tape is like the force; it has a light side and a dark side, and it holds the universe together.

  4. Was that gun design intended to be CNC cut by laser, plasma, or waterjet? A cutter without a lot of side force, which wouldn't be bothered by the long thin pieces?

    1. I believe he cut the original out with a waterjet which does make lots of sense.

      The stated intent was to produce something that could be made in any sort of way. That YouTube kid, Royal Nonesuch, made one with a hacksaw, angle grinder and several "blunt instruments". The drawings, though, list dimensions to 2 and 3 decimal places, with tolerances of .01 or .001. It gives the impression of being more precise than what the kid does.

    2. [I am not a machinist, I just play one on the web.] For the small end mills, maybe first make a bunch of non-overlapping plunged holes like you would with a drill. Then move over half a diameter with another set of plunges to remove more of the web. Only then do the back-and-forth cleanup with the tip. Weren't you trying out a mist cooler? You have that fine enclosure, crank up the air flow to blow the chips away so they don't jam.

      Seems to me the steel pipe submachine gun designs are the easiest to build with equipment costing thousands rather than hundreds of thousands. Maybe there exists a greasegun-type fire control mechanism which has a long history of being semi-auto only? Imagine your speech to the jury: 'I picked this action because of this and this and this feature, so that I could most clearly comply with this and this law.'