What is a "crash"?

[mrquacker]

n00b question: So, what is the definition of a crash, how do I know when I've had one, and how do I know if anything non-obvious in the machine has been damaged?

To make a long story short: forgot to turn off tool offsets on my 770, and Z plunged a set screw holder straight into a block of aluminum. The force of it all pushed the 1/4" EM most of the way into the holder. I managed to hit the e-stop before the holder made contact with the block though. Even the bit survived, albeit with a long scratch where the set screw was holding it.

Programs that I've run since then appear to be accurate and ok. However for future reference I'm wondering if anyone has any suggestions, tips, or other examples of past boo-boo's?

[UUU]

Well I've just broken a tooth off an end mill when I moved in Z when I meant to go in Y. Fortunately the mark was in the workpiece rather than the table, and in an area that will be waste.

I'd define both our events as a "crash".

I seem to make most mistakes by moving at G0 speed when I mean to go G1. I break the cutter sometimes, but others I just get really big chips.

I'm hoping the stepper motors will stall if things get seriously gummed up, and before damage to the machine itself occurs. I have started to use the Z-inhibit feature in Mach3, to prevent the uninteded plunge that you describe.

[GLCarlson]

"loud noise followed by bad language"

Other signs and symptoms optionally present: broken tooling, scrap, blood in the scuppers.

Best advice I ever got (from Bob at Tormach): "Cut a lot of air, with a high zero plane, before you put tool in holder or stock in vise"

I would add: "with one hand on e-stop".

I have made it a standard practice to 1)review all code for maximum machine excursion in all axes (Mach helps here, use simulated run); 2)cut air at least a couple of times on any new program; 3) cut foam (2" thick insulation board is cheap, available, and easily made to match proposed stock); and, finally, 4) cut stock. I do this even when it's a "simple" program- those are the ones that blow up! And the foam model is a pretty good visual validation that I'm going to get what I wanted, too.

Usually, I'm not running long parts programs- a few hundred lines. And I'm not a production shop, so if I take an extra hour to make a part, nobody complains.

It's surprising, though, how many times "cut air then foam" has shown me a bad program, when I was sure it was OK.

[Steve Seebold]

If you gentlemen are getting bad programs, what are you programming with?

[TXFred]

I have made it a standard practice to 1)review all code for maximum machine excursion in all axes (Mach helps here, use simulated run); 2)cut air at least a couple of times on any new program;

I'm not a production shop, so if I take an extra hour to make a part, nobody complains.

The last shop I was at, the boss liked it that I aircut. Better to waste half an hour of machine time on an aircut then have the machine down for a day because it crashed and needs to be repaired.

I submit that any boss why doesn't let his employees aircut a new program, is a boss who doesn't see the big picture.

Frederic

[MetalShavings]

It does my heart good to read of your "Crash-Experiences." I thought I was the only one.

I've had my 770 for several months now and I still deal with the "Pucker-Factor" every single time I watch the tip of my tool make its way toward my work piece.

What is the, "Pucker-Factor?" It's when your Butt-Hole clinches tightly onto your underwear as if bracing for the absolute worst machining calamity imaginable.

I've already snapped, broken and crashed the obligatory 100-plus dollars worth of end mills and drill bits. I suspect it will be another several months before I can run my projects in a slightly more relaxed state of mind.

I don't want to come off sounding like I'm constantly crashing and breaking tools. It's actually only been a very small number but, these darn tools are expensive so the price can add up quickly; and yes, I cut alot of air before I ever chucked any tools at all. I guess it's part of the learning process.

The only difference is that we've admitted our mistakes. If there were a section on this forum dedicated to nothing but "Crashes," I suspect that the owners would have to add more band width.

MetalShavings

[underthetire]

A bump. Doesn't break anything other than the cutting tool.
A crash. Machine needs repair or alignment.

That's how I look at it.

[tbaker2500]

Heh. Most recently I was testing a slot cut with a 1/16" EM. I wasn't convinced I could do this particular cut without breaking the bit, so I had bought two of the bits. I chuck up the first bit in a ER20 collet, mount it in the machine and the proceed to zero for my test run. Forgot about a tool offset and run the bit right into the vise. Funny thing, you can't even hear those bits break. :-)
After much cursing, I wasn't happy. Instead of being careful, I leave the tool holder in the machine and replace the bit (I've heard you're not supposed to do this.) As I'm moving my hand (holding the wrench) away, I hear a "tink". The wrench head had brushed the bit and sent the tip flying.

I was glad the door was closed to the rest of the office, I was in a bit of a mood. I had managed to break both bits in 5 minutes WITHOUT getting to test if my cut would break the bit!!

I had 5 delivered to me the next morning. Tested the cut, no problems. Haven't lost that bit yet. :-)


I highly agree with the cut in air or foam first people here. Despite your best efforts, you will always get some prep move wrong, or a depth of cut wrong. Foam lets you see what you're doing first.

Also, general rule of thumb: If you've just bought a new set of vise jaws, your next part will involve you accidentally milling the jaws.

[Al_The_Man]

One of the things I have implemented on retro fits is to install a rapid overide as well as Feed O.R. which is not often done, this way when rapid-ing in to part for the first time the R.O.R. can be turned down and when close to the part, down to zero, this way the distance-to-go readout shows if a crash is likely.
Operators seem to like it.
Al.

[twocik]

The definition of crash for me is anything that causes you time/money to fix/replace, beyond that is a boo boo. Sounds like you have a boo boo, but if it was me I would check the alignment of everything. Last thing you want are a bunch of parts that you can't use. However these machines are pretty tough and most likely didn't do any damage.

Tips -
After I finish my cam, I always read my G & M codes line by line and illustrate it with my hands in so many words. It's super fun, but will save you a lot of money & time later (trust me). After time you'll be able to read the code like Tank from the matrix. lol

[mike sr1]

I am a new guy too and after plunging a 1/2" endmill into the vise body, I made up a portable e stop as per TxFred's pics here on the forum!

The e stop isnt in a good location on the panel for me, I like to start the program with the mouse in my right hand and keep my left hand on the portable e stop!
I cut air to start and if it looks Ok, then I do cuts in wood scrap, 2 x 4 cutoffs, cheap and easy on cutters, and readily available.
I am kind of a part time machinist and am used to the Bridgeport with my mind and hand controlling it, the 1100 doesnt go by what you think is right, once the start button is pressed the machine has a mind of its own, and will do whatever the program tells it to do right or wrong.........

[dammachines]

I always run my code through a simulator before taking it to the machine. I'm still using SheetCAM which doesn't have a built in simulator so I can also verify that all of my operations have been setup correctly and there will be fewer surprises when it comes time to cut.

CNC Simulator is free and works great for me.

Dave