Pulled the Trigger on a PM 45 CNC...

[Ray C]

Howdy Gang,

I've been looking at this machine for a while now and still vascillating on whether to pull the trigger on one or go the more arduous conversion route on a machine with the larger 9x39 table? Something along the line of an IH model 12Z or a ZX45L. Price of an IH turnkey CNC is not in the cards

I like the idea of just bucking up and having an operational machine dropped at my doorstep, but I do have a few parts in mind that could take advantage of the longer table and X travel of a conversion.

What kind of backlash are some of you seeing on the PM machines?

waddell

My backlash adjustment in X is 0.0002 (2 ten-thousandths) and Y is 0.00015. I haven't checked Z as I'm first now starting to make 3D parts. Most of the parts I've made so far are coming-out within a half-thou tolerance and the main reason for that error seems to be related to the "many times sharpened" carbide bits I use. When I critically measure the bits and re-gen the tool paths with up-to-date numbers, things are coming-out about dead-on. This has far exceeded my expectations but am fully aware that ALL machines will loosen-up a bit over time.

Keep in mind, I'm still in the learning phase and I'm making fairly simple parts. I assure you though, I'm quite thorough in my procedures and methods...


Ray

 

[waddell]

Ray,
Are those numbers raw or after some compensation? I thought I read some ways back that you had some runout in your circular interpolation?

What kind of numbers are other guys getting?

Is the cast iron base a valuable option or would one be better off building a tubular stand and fill the tubes with sand or lead shot?

How about the coolant option........or stay with Koolmist?


Thanks in advance,
waddell

 

[Ray C]

Ray,
Are those numbers raw or after some compensation? I thought I read some ways back that you had some runout in your circular interpolation?

What kind of numbers are other guys getting?

Is the cast iron base a valuable option or would one be better off building a tubular stand and fill the tubes with sand or lead shot?

How about the coolant option........or stay with Koolmist?


Thanks in advance,
waddell

Initially, I was getting a consistent measured error (i.e consistently incorrect) in X of just over a half thou and in Y, it was about a half thou. I initially plugged in those corresponding numbers into Mach 3 and it way over-corrected. I cut the initial entered values in half and bingo, measurable error in parts and measurable backlash is next to nothing. In a sampling of several small (ranging in size from 1/2" to about 2.5") aluminum and steel parts all measured within about +/- 0.0005.

If you look at some of my other posts about this machine, I commented about experimenting with different CAM toolpaths to achieve better finishes. My prior knowledge of manual machining combined with analyzing toolpaths also helps set the part up properly to get good results. I've also paid nearly $1000 in additional CAM training to learn how to get the most out of the software.

Anyhow, on the realistic side, I fully expect that in-time, some additional compensation may be needed -this ain't a $275,000 Mori Seike.

Coolant: I hate flood coolant with a passion.

Bases: Cast iron is always preferable but, I have the normal steel base under my non-CNC PM45 and I have no complaints. I think making a tube steel frame and filling with cement and lead shot will be time consuming, costly -and (LOL) in the worst case, create an EPA nightmare on the day you decide to retire it...



Ray

 

[Boswell]

I have not done any work requiring critical tolerances and I am not a trained machinist so I won't comment in that area but as for the cast Iron base vs making a base and turnkey CNC vs conversion. My only question is
Q1: would you rather be making parts or making tools.
For me, I have much less interest in making my tools than using them so I went with the PM-45M CNC and have been very happy. OK. I don't mind making some of my tools when there is not an equivalent tool that can be purchased.

Howdy Gang,

I've been looking at this machine for a while now and still vascillating on whether to pull the trigger on one or go the more arduous conversion route on a machine with the larger 9x39 table? Something along the line of an IH model 12Z or a ZX45L. Price of an IH turnkey CNC is not in the cards

I like the idea of just bucking up and having an operational machine dropped at my doorstep, but I do have a few parts in mind that could take advantage of the longer table and X travel of a conversion.

What kind of backlash are some of you seeing on the PM machines?

waddell

 

[Thomas Paine]

1/2 thou, hey... That's way better than i thought on one of those machines. Now you just need a 50 pot toolchanger and your good to go.

 

[Ray C]

1/2 thou, hey... That's way better than i thought on one of those machines. Now you just need a 50 pot toolchanger and your good to go.

Would love a tool changer... Need to pull in more machining jobs to pay for it though...

It took about 4-5 tries on a particular test piece to get it dialed in. Also, as mentioned, there's a lot you can (and need to) do to optimize the CAM tool paths to make it cut clean. The CAM program suggests various feeds and speeds and has many types of entry/exit techniques -and many other configurable parameters and cutting paths. Without some background in manual machining and knowing how those factors impact the outcome, it would be difficult to get that kind of precision right out of the box. Must emphasize though, the issue is not the machine -it is very capable... It's how it's programmed that makes the difference. I probably ran the simulation 40 times and watched it very closely to make sure the machine was doing exactly what I knew would produce decent results.

-And for the record, I broke my first tool bit the other day... I was manually moving the head to zero the tool and instead of hitting Z-, I hit X- and in a split second, a $50 carbide tool was worth $0.

Ray

 

[jpfabricator]

Ouch. I dont have cnc, but I know those costly mistakes well.

Jake Parker

 

[JimDawson]

Finally found some time to actually fiddle with the CNC machine today. Most of the time, I dream-up parts, draw them in CAD then, setup toolpaths in CAM and just run the simulation. The simulation is very good and all it needs are some sound effects and maybe throw some chips around and there'd be no need to actually mill anything -just make virtual parts all day long .
Ray

If you don't mind my asking, what CAM program are you using?

 

[Ray C]

If you don't mind my asking, what CAM program are you using?

Short answer: No, I don't mind at all. BobCAD/CAM V26, 3 Axis Pro with 4th Axis toolpath extension option.

Long answer: Was undecided about MecSoft vs BobCAD after very careful feature/performance/training/price comparison and long-term demo analysis. MecSoft has a very convenient plugin that works with my CAD program of choice (Alibre/Geomagic Professional with motion and stress analysis add-ons). With BobCAD, solid files must be exported by Geomagic and imported into BobCAD. Initially this seemed fine but at the onset, I was having difficulties learning/understanding the pros/cons of various solid input file types. It took a couple of "all-nighters" (quite literally) to figure it out. BobCAD's 6-DVD training series is excellent but, it's hard to find specific answers by re-watching 70 hours of video. -Was initially annoyed -but nothing is allowed to get the best of me. Finally made headway and saw the light. After become more self-proficient, I signed-up for one-on-one online training (very expensive) but, was armed with very specific and advanced questions. I took the last of the 4 hour long training sessions this morning. I see the light now.

In retrospect, either solution is probably good but, cannot fully comment on MecSoft as I don't know it as well as I now know BobCAD. BobCAD sales offered some very good discounts. -No dice with MecSoft. FYI, both companies purchase the intellectual rights of the toolpaths from the same company. My initial feeling now is that BoBCAD is better on the pocketbook and offers an advantage if someone hands you a solids file and says "cut it for me". Big difference between the two is that BobCAD has an AutoCAD-like built-in CAD functionality whereas Geomagic is a parametric-based CAD; thus, I had to learn some aspects of Auto-CAD style CAD editing.

BobCAD: Very stable so far. IMO, much better workflow for CAM setup. A little bit less expensive -not as convenient to use with Geomagic. I won't get rid of Geomagic -outstanding CAD, motion analysis, stress analysis, BOM, sheetmetal ops, 2D Mech layout drawings.... BobCAD doesn't even offer those things.

I'd like to learn MecSoft for the heck of it....

Ray

FWIW, when I finished the training today, the instructor had me speak to the director of training. He asked me if I'd be interested in training to become a certified instructor... LOL, I told him I've only been using the program for 6 weeks. -He didn't believe me. Who knows, maybe I'll look into that opportunity....

 

[JimDawson]

Thank you very much Ray.

I am currently using CamBam, a pretty good CAM program for the price but a bit limited in scope. By fudging you can create a sudo-3D tool path. On the other hand it's only $150, you get what you pay for.

I am looking for a true 3D CAM program for when I get my Z-axis conversion finished. I have been playing around with CAM360 by AutoDesk, but it is still in beta. AutoDesk did provide a post processor to my specs that works with my software, but I'm not married to them.