Do those Iscar plunge cutters come in imperial sizes, or metric only?

I'll have to check that tool out. We do a LOT of plunge cutting. Currently using ingersoll DHU series for 1"-1.5", and reground bradpoint drills for 5/8-7/8. (Don't judge, they last FOREVER, even if we can only feed them 15ipm in 495B)

3D contours, all the way! But my typical application usually only involves 5-10 plunge cuts, patterned around a radially symmetrical part, and the parts are designed with a 1 degree draft angle so the rapid retract works fine. If you didn't have a draft angle, you could always add a perpendicular wire entity .01-.02 long or so to move away from the wall before retracting, only needs to be enough to account for any deflection you may have plus a hair or two.

Good idea. I know it's not going to shank out with the cutters I typically use, especially with the 1 degree taper. And it is almost always the shank it has a "collision" with. It's not dangling dust or chips... if I watch really close it looks like it's colliding with the material that is being cut. Hard to explain what I'm seeing, but the cutter goes to the endpoint of the move before the cut material disappears, and the material where it is registering the collision has this weird, almost fractal appearance for like half a second before it disappears. "Just theory but am thinking turning precision up actually helps limit false reports? I have run into this time to time. we call it space dust, little false fragments left in graphics area?? IDK. dont want to ad to your frustration but Vericut is the ultimate solution to this. sounds to me by your low end system this is not going to be an option $$$ wise. frustrating when company's trip over dollars to save penny's. I feel your pain" They're actually not too bad about buying us whatever tooling we want. We're about to go through and standardize our resident tools on the horizontal mill cells. TBH though, with most of what we make here, I could get away with writing a few macro programs for each machine and calling it good... if our engineer would standardize features in part families rather than just winging it every time. But then I'd be out of a job.

No, my computer sucks. I'd have to leave it for an hour to verify a part with precision all the way up. But I will try turning it down more to see if that affects the frequency of the false reports. Thanks for the suggestion!

I've been having an intermittent issue in verify with X9, and was wondering if anyone else had the same problem and had figured out a solution. Most often, I am having the false collision reports when verifying a plunge cut. It looks like the tool is stepping to the next move, but the material that should be cut is not removed as the tool steps through. So it will then register a collision with the shank/shoulder of the tool. I can see that the material is still there that should have been cut off, and a few seconds later the material disappears and it shows the cut as it should be, but not until after it has registered a collision. Does not show even a speck of red anywhere after the cut material is removed from the model. I know it's not an issue with the program; these are super simple cuts it is happening with. Plunge cutting with about 10% of the cutter diameter, and tapering away from the wall being cut at 1 degree. I've run several programs where mastercam showed a "collision", and there is none. No galling on the shank, not a peep or chirp while cutting whatsoever. It definitely looks like a software issue to me. Any thoughts? EDIT - Worth noting that it does *not "collide" at the same point every time I run verify.

Hahaha.... haha... haaaaaaa... glass filled PEEK sucks to machine. I did the same damn thing my first shot at machining it, used a 1-1/2" AB shear hog for facing and turned the part into a frigging potato chip once it came out of the vise. Everyone has had good suggestions so far, skip around a lot so you're not concentrating heat anywhere, sharp (pos rake, radial relief) tools work well, but better find something with a good coating, since that glass will destroy uncoated tools. Rough and finish in separate ops, reduce clamping pressure to absolute minimum- if you're going to do it in a vise all in one shot, at least put in a M00 and re-clamp with as little pressure as you can get away with. About the only thing I can add, is that you might allow the parts to just kinda sit in a warm place after roughing to allow the plastic to relax a bit. A lot of plastics will stress relieve at room temperature if you give them time.

Tap/mill an M32 internal thread into a block (squared up very nicely first of course), drill clearance holes halfway through and thread for the rest around both sides of the thread, split the block most of the way through down the middle of the thread with the thinnest slitting saw you can get away with using, and you now have a custom made clamp for hanging onto the thread. Make an extra to speed up part exchanges, so you can have the next blank clamped and loaded and just switch out the blocks. -If you want to get really fancy... drill/mill/threadmill an internal M32 down the centerline between some soft jaws clamped on a 1/32" shim. Faster loading, and you could probably get away with ganging up 3 or 4 per cycle in one vise.

I've only ever seen A51s in action, but for what we did at that shop they were fantastic. Consistently accurate, good rigidity, and very reliable. Held tenths all day and night. In the two years I worked there, the Makino cell was only ever down due to issues with the pallet changer/storage system behind the cell... well, and that one time one of my idiot co-workers forgot to draw in all his stock/fixtures when running parts on all 4 sides of a tombstone and smashed the spindle into a vise in full rapid.

Yeah, definitely a misnomer.

Anyone know any tricks to force mcam to post out A axis moves for plunge milling radially symmetrical features on a 3+2 horizontal @ b0? I tried changing lw_y_lin_lim and up_y_lin_lim in the post file to restrict movement to 1/8" above and below y0 to force it to reposition in A, but it made no difference. Axis combinations don't work with a 5 axis post so I can't shut off y axis there... one of the machines here that we use quite a bit for these parts is suddenly (well, maintenance may have dropped the pallet + tombstone against the way covers...) having trouble with chatter in y axis cuts, but not x. It's an old Kafo, so the odds of getting the machine properly repaired are slim to none. Any advice would be greatly appreciated. If I were to try to mill those features at B.5, and use a transform rotate op, would that force a axis movement instead of using xyz? -E edit - Milling the features at B.1, using transform rotate to pattern, did the trick. Figured I'd share on the off chance anyone else might be able to use it. There is some error on the radius created with the plunge tool, but on these particular parts it is within acceptable tolerances.

I've used them before as well, but always follow that cardinal rule of assuming that the load could fall at any point... so stay the #### out from under it, as you would with any load, regardless of the lifting device. Another thing I've seen people overlook that can lead to a dropped load, is not cleaning any chips off the face of the magnet that may be there from previous uses. IMO, no worse than using a vacuum to lift a load. (Assuming all the proper check valves are installed to make sure vacuum is maintained, at least temporarily, if there is a loss of power to the pump). They all have their uses.

I've had pretty good luck using TiB2 coatings in aluminum. Reduces chip adhesion, good wear resistance, and has good lubricity.

Can seriously hose you if you're not careful, but dynamic xform is a quick and easy way to move geometry. I use it often enough it has a hotkey.

Post debugger is your best friend when you're learning. Makes it much easier to trace exactly where each command in the NC output is coming from. Like they both said, unless there is something else going on, all you should have to do is change the lines in the coolant table as long as you have X style coolant enabled. I've only made 6 posts, so I'm no expert, but cannot express enough how useful debugger is for newbies.

I've made posts that work for most of the mills in the shop (all 3+2), but I'm having issues with a post for an old BMC 63 (tosnuc 800 control)... someday, I'll be knowledgable enough with this stuff that I can give back to the forum, but so far you guys have beat me to every newbie question that I could answer. Anyway, here's what I'm having an issue with... B axis needs to be posted in signed absolute mode, but when using a canned drill cycle, it posts out a B90 when it's already at 90 (on that control, it will make the pallet rotate 360 back to where it already was, which will rip the cables out or worse, depending on where it is). It does not do this when it's in not in signed absolute mode though. What am I overlooking? I know it has to be something stupid that I'm just not seeing. (Edit - Does not post the redundant B moves for contours or any other toolpath I tried, only with drill cycles.)

I think dollars is an appropriate substitution here.

Yeah, I have the most recent version, but thanks for the pointers! I'm still learning, and definitely taking the time to double check what my posts are spitting out, and test a few different toolpaths with each change I make. Was feeling pretty overwhelmed at first, but you guys have been extremely helpful. I guess for the last 10 years they ran mastercam here, nobody ever bothered to set it up to post a axis moves, just used a generic 4 axis post and hand edited everything after post to add in A moves and match each specific machine. Still trying to find a way to lock the Y axis for some toolpaths and just cut with x, z and a, but not a pressing need right now.

Works well, also got it to post a G325 for the tool breakage check on all tools less than 1/2" dia. This post editing stuff isn't too hard if you have some experience with other programming languages.

Figured it out, just took a bit of screwing around. I did a little bit of post editing in college, but just barely scratched the surface back then. Got it working from the generic fanuc 5 axis post. We're just using them as 3+2, so don't need them perfect, just need them to spit out the right codes. So far have 4 machines set up with their own uniqe posts - Kafo vertical with a axis, kafo horizontal with a+b, 3+2 mori seiki with breakage probe, and an OLD enshu horizontal 5 axis. They said they wouldn't buy posts except for the new machine they're getting in a month, and everyone they had hired in the past just hand edited everything from a generic post to match each machine. If they're willing to pay me to learn, why not take advantage of it? Haven't had a crash yet, but I also very carefully check all the code before sending it to the machine.

I doubt I'll get them to pay for any post help... so, back to studying for me I guess. On the plus side, within a year here I'll probably be able to figure it out on my own. Knew next to nothing about writing posts a month ago, and I've managed to figure out quite a bit. (With some help from everyone here, of course)

Are selected axis combinations disabled when using a 5 axis post? We have many parts where we plunge cut similar features along the diameter of a round part in z at B0 (machine is set up strictly for 3+2), and I'm trying to figure out how to lock the y axis so it will cut these features using only x, z and a. I tried creating an axis combination in the machine definition that omitted Y, but it did not change the way it posted at all. Several of the machines I post for have very little memory on the control, so the code for plunging along an arc is a lot more compact using G18s, and it is easier for an operator to make program adjustments to comp for tool deflection when they are just changing x coordinates, rather than trigging out the x+y's. That, and I keep getting the "well....... that's just how we've always done it" line.

Yeah, pretty much. Could just add in string literals for the b movement using incremental like that... if it weren't in the encrypted portion of the post. So much of the 5 axis post that I can't see, pick apart, reverse engineer, and learn from!

Looks to be in the encrypted portion of the post file (Generic Fanuc 5X Mill from mcam was what I started with). Guess I'll stick with hand editing for now. Thanks anyway, everyone!

Maintenance said they've already tried adjusting the backlash parameter but the control can't compensate for it. From what I hear, it was a pretty spectacular crash, messed up the ring gear bad.