Winter Project . . .

[Foozer]

An attempt at a desktop 3-axis to be GRBl controlled ... or an epic failure in the making. Time will tell. A 12" x 20" footprint shooting for an 8 x 12 or so work area. Bunch of 1/2" x 3" bar stock, some 1/4 " angle stock, lots of 1/4 shoulder bolts set in .247 holes. Base assembled, slide rails tentatively installed, little tap here and there and best as I can measure they are square +/- .001. At least the slide moves, no noticeable binding with 'Pinky' (technical term) pressure.

 

[Foozer]

Still goofing with the Gizmo. Just a matter of adding more metal, setting a tolerance that keeps me from chasing my own tail seems to be around 0.003 thou. Figure 3 tho over 12 inches is as good as I can do, so got the bed level within that and moved on to the next thing, the upper. A little of this and some of that ends up with the lower slide rail being about .002 off with the bed. So with that, time to make the shims for 'tween the rail and mounting brackets. Cumulative error? Will deal with that after the Z-Axis assembly gets put together. End of the day it's just a putt putt.
One of these days I need to get a better camera . .

 

[Foozer]

Winter again and still putting with the gizmo - When in doupt, add more metal . . .

Now what to do with the free pen the bank gives out?

Silly thing actually worked . . .

 

[stevehuckss396]

Of all the CNC guy's I know, we all wrote our names with a sharpie marker to test our mills. When mine was finished I didn't have the smarts to draw pictures.

 

[Foozer]

Of all the CNC guy's I know, we all wrote our names with a sharpie marker to test our mills. When mine was finished I didn't have the smarts to draw pictures.

Thought I'd try the Inkscape jpg to svg, import into Fusion 360 and see how the engrave - trace functions work. Or to see just how it would draw -> Text letters came out well - not to bad to a freebie ball point pen. So to impress the Bride - did one of her cat . . .

Trace works better - Pic is engrave - Letters by trace - Still learning Fusion - Now to add the spindle mount and make a pen holder that will slip into it . . .

 

[Foozer]

Did I mention how much I dislike wiring? soldering? If I knew what I was doing, I'd dig a hole and bury this thing.

So hopefully have closed up the box for the last time (Well still have limit switches to do) - Spent way too much time chasing the EMI ghost. UGSC would randomly freeze up, twas the cheap USB cable, swapped it out for one with Ferrite coils attached and so far no more lock ups.

Have a lot stuffed into a little box - It behaved on a two hour test run.

Did I mention how much I dislike wiring? soldering?
Di I know what I'm doing? Heck No . . .

 

[Foozer]

But will the gizmo do anything? Well... it can destroy a dremel tool - Ok sorta expected that by pushing the feed rate - Dremel 0.020 doc & 0.100 step-over? Nope . . .
The *cough* Chinese 500 watt spindle thing showed up... Treat it like a Dremel - Well this contraption is just a heavy Dremel mill, Now can I turn drawing into item.

Gonna try a Beam 6 x 1 x .375

And so far

Mounted using painters tap and super glue
Nibbling away 1/8 dremel endmill @ doc 0.010 step-over 0.030 bought 20 ipm (fast need not apply)
Coming out better than expected - The spacing between the web cutouts and beam wall appears consistent.
Now on to the near 2 hour time to contour the outside - Flip it over and hope the locating plan works . . .
Outside contour has a three degree to it so it's 3D contour, Dremel ball mill and lots of time to kill while it nibbles away - Only took me an hour - Fusion 360 - (Ok more than that) to catch on that had to patch the top surface (new body) to limit the tool path to the outside edge...

 

[Foozer]

Spent a bit of time on the nitty gritty stuff - Is the table aligned with the spindle? Tram - Nod - Shingle? Who comes up with those terms . . . Anywho - cobbled up a holder for a DI - Checked five points , center and corner - Initially right side was low bought 0.005 or so - Surprised me as construction was by hack saw and drill press - Bit of shimming and four corners read .000 - Went for consistency of reading - Accuracy is set aside by the inconsistency of the ball screw - (Chinese Drunk'n Threads) - 0 to .5 reads .498 - 0 to 1.0 reads 1.0 - Now on to centering up the spindle soon as I find the indicator . . .

Shall be glad when it warms up - Got a truck dash that needs restoring . . .

 

[Foozer]

Shimming is a non-caffeine adventure . . .
You get to a point where you're chasing your own tail - When tightening down moves the needle more than it reads off, it's time to call it good. Used the torque wrench 30 inlbs to snug the fasteners - - 1 tick off - fore to aft - 1 tick off - left to right - Call it good - Ran a test facing 280 fpm 1 thou per tooth 4200 rpm 1/4 3 flute stubby Sherline end mill (them suckers are sharp - Cute little blood drawers)

No finger nail clicks across the surface - Time to finish cable management . . .

 

[Foozer]

Back to the order sheet . . . Condition being not enough room to run a limit switch cable in the Z-axis run. Solution - get a larger cable chain. Thought the larger one was of the type that snaps open to lay the cable in - Nope - Could unsolder the wire ends, run the cable and re-solder - Don't mind spending time redoing what already has been done but sometimes you just have say - Not Again -
So with cable chains - bigger is better - ability to snap open to lay in a new run is up on the list . . .

 

[Foozer]

New drag chain showed up - Much better - Now when/if I come to the point of adding another cable(s) for the X or Z axis can just pop open the run, lay it in, close it up easy as pie . . .

 

[Foozer]

Think it's testing the limits of tape and super glue, but the part didn't fly out so . . . Fillets came out lookin Ok . . .
Made a jig - locating device - Gizmo ? to hold the piece in place so the facing and fillets could be done one side - flip it from A to B and do the other side.

What Fusion generated for the roughing pass


The end result


Now to tape er down and do the other side

Gizmo took a 0.002 inch clearance for the piece to fit in snugly - Just have to remember to change the setup orientation . . .

 

[Foozer]

Ok - Ran the fillets on the other side - They be So-So but as I have never done this before I'll stare at it before I chuck it into the scrap bin.


This got me on the adventure of tool paths - This one is Blah too much yellow, too much unnecessary movements.

How do I fix it - Ended up altering the model to get a path that goes round and round, minimal retracts (it can cut air) Just follow a path and step over along that path. Scallop -

Lot cleaner looking - Just use that path on the actual model

and hopefully it'll be better - I will keep telling myself that
If there is a better way to achieve the similar method - Work on that after another go round with this learning curve . . .

 

[Foozer]

Another adventure - Another Ok But . . .
So setup another piece to practice on - Clean the jig surface - Acetone -

Apply tape - Super Glue spread out with acid brush

Set in place and add weight - It's what I had . .

Raspberry Pi running UGCS to Arduino-GRBL to

OK But - - - Need to come up with a new holding jig - maybe get rid of the side wings-use a pin in the hole with the aft riser to locate the side to side. Would like to use the same bit to contour the fillet and sides - Ball Mill? Not this way, some sort of flat bottom mill with a radius edge - that won't break the bank - But time to take a break and let it all sink in - Tolerance - Smoothing - Model faces - Touch/avoid - step over - inside to outside - outside to inside... Contour vs. Scallop... Ah the buzzer buzzed Corned Beef is done . . .

 

[Foozer]

So looked at this little thing and wondered 'What would a full size kinda look like?' Say if it was cast - It would have some draft to it, sounded good to me so then how do I cut it all from one side - Guess the term is undercutting - other term is 'What was I thinking - -' So a new pattern jig to allow the ball mill clearance on the undercutting.


The 'Tape' parting line is there somewhere. After a few hours of slow and go pulled the part.



Tried a little engrave, was a why not.
Now to change the driver stepping - currently at half stepping and noticed an irregularity along the sides - it's wavy - Shall see how the next one does with quarter stepping - Least it'll be easier - Combine paths into one op, turn it on and let er run . . .
Insanity or obsession - Both are fun . . .

 

[nel2lar]

Foozer
Some people just have too much fun. I am glad to see you are still smiling on the curve of learning. Very nice work so far, keep the shoulder to the grind wheel and it will be an running engine before you know it.
Looking good!
Nelson