Post up them failures

To parallel the “Post up them projects” thread, here is where you can post your failures, and hopefully what you learned.

Here is me learning to cut aluminum:

What : 2mm thick 6061 (I believe) aluminum

Damage: $50 worth of end mills and about twenty hours.

Process (so far) : Fusion360->import SVG->emboss upwards->machine with “trace”

Machine specs: 800 watt water cooled Huanyang spindle on a Woodworker 1F, compressed air blowing across the end mill tip (chip clearing and cooling), 600CFM dust collector connected by 4” hose to dust boot.

Learned:

  1. Cost: one 1/8” flat nose spiral upcut two fluted end mill and a 1/8” ball nose - snapped off my first bit because my origin was 2mm lower than it should have been. Remember I embossed upward… so on the first cut, the bit buried itself 3mm farther than it should of and snapped. The next one I got lucky because the aluminum sheet is not exactly flat and my z-probe height was high by a few mm and it wasn’t until the aluminum melted onto my bit and welded itself in place, that the second one snapped - my feed rate was too slow as the chips were not taking away enough heat and I was cutting too deep. My bit was wrong for the material and clamping was also not sufficient, which lead me to my second failure/learning:
  2. Cost: crashed the machine : tried a 1/4” flat nose spiral up cut two flute which tossed off extremely sharp chips and made it about half way around before it grabbed the sheet and pulled it right out from my clamps. The machine stalled out and I called it a day. I had yet to fix my origin problem being too deep.
  3. Cost : one more 1/8” flat nose spiral upcut two flute end mill - after fixing the clamping, zeroing my Z on the lowest part of the sheet, cutting less material on each pass and finally fixing the origin problem - my feed rate was too slow (1200mm/min) and the bit was wrong for the material. It cut about %30 of the job, got hot, welded in place and snapped.
  4. Cost : Success - After watching the video on this thread about cutting aluminum and getting the right end mill, a 1/4” single flute end mill made for aluminum (and costing $18) and with blue tape, glue and numerous clamps, 1500mm/min feed rate, .2mm doc, finally got it to cut.

Summary: Aluminum sheet is reactive - it gets hot, flexes, isn’t exactly flat, and bits will grind to a halt and snap. My initial success with wood, which is really forgiving with incorrect origins, depth of cut, wrong end mills and feed rates, made me overly confident.

9 Likes

Hey Sean,

great topic. Now it’s getting interesting.

3 Likes

@TugBoat - I love this idea! I’ll be following this thread and most certainly will be adding to it… lol

Here are some failure quotes I’ve said myself and heard over the years;

Success and wins are great and feel good, however our losses and failures teach us so much more and truly make us who we are.

We are only a good as our greatest failures and the lessons we’ve learned from them.

Failures are almost always costly and/or painful, learn from the failures of others and don’t be afraid to share yours.

If you don’t fail from time to time, you are not trying hard enough, pushing the envelope, pushing yourself or trying to developed your own original ideas.

Failure is not the end, it’s the beginning of a lesson in what didn’t work. Study and analyze it so you can find your path to a way which will work.

Don’t be discouraged from trying your idea and except someone telling you “It won’t work, we’ve already tried that, we’ve already tried everything.” Yes, you may fail the first time as they did, however you may see, hear, feel, smell or taste something they did not which may lead you to make adjustments which will work.

At this workbench failures are expected, inspected and respected. ~ Unknown

2 Likes

Poor clamping and poor patience. :slight_smile:

7 Likes

Well, been there and flat done that! Lessons learned for sure.lol

My setup… make sure your CNC isn’t a parallelogram or skewed. Unless you don’t mind cutting parallelograms when you mean to be cutting squares. After I noticed the issue I found even my Y axis was bolted at an angle.


3 Likes

Well this was not good. Smoked the collet nut and froze the collet and bit. Had to take the spindle apart and heat the outside with a torch to get the collet and bit out. Broke the bit taking it out. I tried to use too short of a bit to do a finishing pass on a deep 1.75inch bowl.





5 Likes

Ladies and gentlemen I have discovered that vibration is bad.

In the post mortem it was obvious that my bit had loosened and because it was an upcut bit, it dug in deep!

This was my first experience with the machine bogging down and losing steps. But to be fair to the 1F it was trying to drive a 1/4 bit through 1"+ of material at nearly 200 ipm.

5 Likes

Not sure what happened. I’ll take a look at the vcarve file. But I never got to designing a profile. And especially not that low. On the Y axis. Also xy origin previous to cut had been off by 2 inches, no explanation on that. 5 hrs down the drain. Last check was 14 mins left. So this happened during that time. Looks like 5 profile cuts

Finishing run complete. Went to rerun my roughing pass with a raster toolpath to cleanup a few areas and apparently my machine lost it’s point of origin in the process…

I work mostly with rough-cut lumber and getting an exact math from probing X and Y multiple times isn’t likely. Going to do a little digging to ensure this doesn’t happen again. Lesson learned.

3 Likes

I think it’s Raster. Try along the X or Y. I just tried a third time on a 3hr carve, changed from Raster to Y axis cut and was successful. Not sure why this is happening.

2 Likes

Thankfully I was able to cut that piece down and make some pretty nice coasters. On my next attempt I will give that a try. Much appreciated!

After having epoxy seep into my t-track, I shattered this baby trying to remove it. Sucks after a week of 9 different milling operations.

7 Likes

It’s always the oops after a multi-step process that hurts the worst.

1 Like

Placement is everything…

4 Likes

Very nice design. The next one will be right on. There is always the other side.:grinning::+1:t3:

2 Likes

I’m getting some pretty severe bleed when I pour my epoxy in the engraving. Not really a CNC failure, just the 200 lb gorilla at the controls. All suggestions welcome! Right now I’m ignoring this project because I’m frustrated with it lol

I haven’t tried this personally but I’ve seen people seal the surface with lacquer or brushed on epoxy first, let that cure and then apply the tinted epoxy. Also I’ve seen this technique combined with cutting the v carve deeper than you want the finished product and then sanding off the top to remove the bleed. I would suspect you could also use a facing type operation to remove .020" or so if you don’t have a drum sander available.

2 Likes

Good ideas. I’ve seen some incredible multicolor epoxy inlays by folks using the flattening technique to shave off the top 0.1" using a surfacing bit.

A planer could do it as well & probably a bit faster with sharp knives - helical knives will give the best finish.

Then follow it up with a random orbital sander up to 320 or higher grit depending on the ultimate finish desired.

Paint the area with unwaxed shellac, sanding sealer. Alternatively paint on a thin layer of clear epoxy before the colored.

1 Like