Travel in XZ or YZ plane Arc filter to .0001-.0003 (need to experiment) Filter ratio 3:1ish. TURN SMOOTHING ON and slide way over- that's the key to get arcs in G18 or 19

I'm supposed to tour the NC facility tomorrow (feb 26th), however all flights have just been cancelled. Maybe Friday I'll make it. Chris, will you be there? Our tour guide is a Gosiger AE. thx

Sorry I can't remember which forum member(s) are there. Please PM me at your convenience.

Rant: What bugs me big-time is the data entry boxes on some newer functions are no longer yellow, which historically has been a visual cue that it is calculable field. eg the entire tool manager doesn't have it. This UI inconsistency is yet another factor that shows how fragmented and poorly integreated mcam has become. Sure it's a little thing, but it is adding up. (Why in simulator are g-view icons red, while for the past 20 years been green in main environment?) Has anyone else seen a software package so inconsistent and lacking in UI management?

The gained productivity of max rpm compared to the low price of a haas spindle too.

Good point. I program low-load/dynamic cuts with fast feeds, with low spindle loading, and the long run times are finish with minimal spindle load. So theoretically if lube system is clean and flowing, air is dry, balance is good, no-load wide open running spindle "should" have a very long life. If it's good from the get-go. Shade-tree theory there, you read it on the interweb

I've run the living crap out of haas spindles. 15000k spindle, ran 48hrs straight on a mold. Still running 8 years later. 6k spindle, 24hrs at least. Still running 10yrs later. 10k's? Same about thing. 12k's? somewhere up there too. 7500 at least 48hrs. A vf-1, the literally Very First One, from the early 90's, has been run for at least 24hrs many times. The spindles that were bad seem to go right away. If they last a year, it seems they will last 10. Thru-spindle coolant I would be hesitant about. That said, every spindle has it's "sweet spot". Years ago a tech left a vibration analysis oscilloscope type device at our shop for a week. I found each machine had an optimal rpm where vibration was lowest. eg, a 12k machine that ran optimally at 9500. a 6k machine at 5xxx. It was interesting to really experiment with. A few japanese that we had were MUCH more consistent in the vibration profiles. Haas was not. m2c

Between the middle boss and the wall? Backplot it and watch very carefully. It can be deceiving looking at the path only.

Been busy all day- missing out on the fun here... Yes the first column not locking really makes it a hassle to view the information. Chrome browser will lock the first column, and firefox might now. (It behaves a bit different for me because of full edit permission). Now that I think of it, I need to go in and organize some recent entries. (The google auto-entry form will only output a horizontal format spreadsheet, which is nearly unreadable. I have a shell that re-formats it into the legible vertical format and fancy colors). Been wanting to try that triangular mesh for a while, you've been really talking highly of it, thanks for the example file Colin. Interesting if you change the "Type" from adaptive to parallel, you get a completely different toolpath. Lots of options in there. I really need to set aside some time this Xmas break and learn it. Oh, this is back in X5 when I learned about the over engagement "issue"....with a $200 Sandvik Plura Mill.

If you backplot and save as geo, you can analyze and see the user-entered stepover value is violated. Period. Not by a bit, but sometimes 40-50%. After a few passes and the sharps are knocked off and the morphing gets going, and stepeover is NOT violated, and the toolpath becomes "correct". Inside corners it stays at the programmed stepover, and rounding radius.

The first pass offset and feed reduction are clever patches, but just that, patches. They don't address the over engagement. I'm curious what volumill and other software packages look like in this situation.

Perfect timing. Last night we were testing the limit of what a 1/2 cutter, .875 deep would do in 1018. Tool broke at 750 ipm at 6000rpm. 10% stepover. Imco pow-er-mill tool. HIGHLY recommended use x8, because it has first pass offset and feed reduction. Those are two must-use features for rectangular shape stock. Otherwise over engagement in corners = tool breakage. I'm assuming that's what you were referring to in your picture? The increased corner engagement? A few of my latest findings: Only use micro-lift if the floor matters. Even the fastest machine will decel coming in and out of a micro-lift. Keep filter tolerance at .001 regardless of stock to leave. Any value larger and tangency between moves goes away, thus machine decells a to minimize overrun. I'm working with an Okuma Geno OSP 300 with super-nurbs, so I've got the control and machine to digest these rates. My goal is to cut at 1000ipm.

Lathe- I was making 3" dia wafer thin parts: Turn, drill, part off. Incrementally move in Z -.500, turn, drill, part off. Incrementally move in Z -.500, turn drill part off....parts are running great and I'm all proud of myself......of course the turret is getting closer to the chuck with each part....until a tool change caught the chuck at about 3,000 rpm. Spun the turret, sheard all sorts of stuff, exploded 8" chuck. 12,000lb machine jumped. Took it apart myself and fixed for about 2k. . Learned what a curvic coupling is and what a cool piece of engineering it is. Ever wonder how a turret withstands cutting forces, yet indexes with perfect repeatability? Curvic coupling.

1) In the Settings button, turn off "Switch to Manipulate geo after inital gnomen placement" . (Fyi, this was also on by default in X7) 2) After your done adjusting gnomen, there is a small translucent box to the lower left of the gnomen. That switches to move part. (#4 below) 3) After you've hit aforementioned move geo button, then the old Cplane/WCS etc options appear. 4) My MAJOR gripe is the axis labels are gone. Thus dynamic xform is 100% NOT usable to the colorblind.

Almar Tool right here in Portland is phenomenal. Used them several times, with one company ongoing, and they are great.

Count me in on that tour I haven't seen your place Bob since you've made the Makino conversion. (Actually you just got the A51 installed when I was there).

No problem Bob. I'll bet you'll be able to crank up that feed a bit. I recall running an uncoated garr 1/4" .500" loc at about those speeds in A2. I believe about 8-10% stepover. hth!

I wouldn't think you'd be able to beat peel mill performance, here is a link to the old topic and the mori that had problems 2009 thread on Mori smoked bearings Also false brinelling. Mayye Makino could do some calculations if your ossilation distance would be enough for a ballscrew bearing to circulate. Tell them you'll make a video on how bad-xxxx it will be

I've had this in an older Oi control. Set control to "don't break arcs" and "allow 360 deg arcs".

+ 1000 on the ability to modify colors to both levels and viewsheet tabs. (I think I sent this in as a request a while ago actually). Most of us use weird level #'s in the attempt to provide visual clues as to groups of levels eg. 1111, 222222, 33333333,44444444. This is our hackneyed method to see them as a "group" or cluster related to an operation set. Colors would be a BIG help. I'm not criticizing mcam developers or anyone, but they may not work on the same file for weeks or months like users do..and have to open it up years later and try and to discern things. Nor provide a decent trail of bread-crumbs for co-workers to follow, or experience the pain of rev changes. They may just not use mcam in this way, and therefore havn't directly experienced the helpfulness these suggested features could provide. ? I certainly could be wrong. Constructive criticism I hope however

Viewsheets goes a long way towards this.. Activates levels as a "set", and also planes and views, which is hugely helpful.

Lathe is waaay more advanced that mill. Always has been. LIVE stock updating, IF stock is defined. Intelligent retracts and clearance moves based on remaining stock, IF stock is defined. Perfect stock flip IF stock is defined. No need to ever change coordinate, IF stock is defined, and WCS T/C plane is left at top, and part is located correctly. Just use it how it's designed at it works great. I wish I replied first to this thread and saved everyone else the hassle. Lathe training www.streamingteacher.com. Also Mill/Turn main/sub and everything else you want to do with lathe.

What control is it? In the control def arc page set arcs to "don't break arcs" and "allow 360 deg arcs". That has fixed several problems for me in the past. The important question is can you re-run the program with changed settings to test it?

The absolute core of it all is Just in Time. Customers buy cars one at a time.. Therefor they should be made one at a time. The holy grail is for an assembly line to make 1 blue camry, 1 red suv, 1 silver pickup, 1 prius, 1 white suv, etc etc. one after another. Yes it can drive a shop to insanity.

Years ago I had a really sweet gang-tool type setup. Turning the heads of titanium SHCS. In an ER holding a threaded stub. Then made a vice jaw with a lathe tool holder, drill, and scotchbrite pad. At the end of the program I had an M00 dwell, with an M04 to unscrew it. If I recall, have your C-plane set to top, and T-plane set to front (or the other way around). Worked just fine with regular mill post.