is CAM Software Necessary?

[camoto91]

Our shop is very new to CNC, and I'm the only who's involved in running our Milltronics VMC 3018 so I've had to do a lot of independent research. I know for a fact CAM is something we have to utilize if we want to progress further with our CNC Machining, but trying to convince old school machinist and plant managers that's something we need has been challenging. They believe that I can do everything they've seen in online videos by strictly conversational programming. They don't realize all that it takes to write out something that complicated, all the math that's involved, all the double checking and triple checking your work. Can anyone give me a list of Pros and Cons of having CAM assisting with CNC programming? I'd like them to see/hear it from experienced programmers and machinist out there.

Sorry about the lift story, but I have been frustrated with this for a long time and needed to vent. Cheers...

Similar Threads:

• Software for 1.6 mW Laser Engrave Machine Software for filled logos ?
• Problem- Will the SteelTailor Nesting software support a local CAD software ?
• Just IN- MIE Solutions - Free Quoting Software Trial Software available
• First post in the software dept. Free cnc software?

[awerby]

One missing decimal point can send a tool crashing into a fixture, or blow an expensive ball nut apart with a runaway axis. Maybe your old-school machinists never make any mistakes, but looking at the posts here, there are few of them without mistakes in spelling, grammar and/or punctuation. That's in English, where it's relatively easy to spot them. Proof-reading a G-code program is much harder.

It's also a lot faster to generate code with a CAM program, or to change it if necessary. And you can tackle much more complex parts, which you'd never be able to code by hand.

[SteveWill]

I am thinking that they bought the CNC and hired you, so they have spent the money this year, So shut up and do the job. Yes, you should have a CAM program. If it was me I would ask if it would be OK to put together a written proposal for next year budget. Then I would list some different software to make the job more cost effective and what each would cost, save or increased production. If I did a good enough job they mite buy sooner. Forget the old school machinists, it is the boss and plant manager that make the decision and they are looking at bottom line.

[109jb]

Does the shop make more than $100k per year? If not Fusion 360 is free and has CAD and CAM integrated.

[louieatienza]

Well if all you have to do are pockets and bolt hole circles, then maybe you don't need CAM. If you're main parts are 2.5D - pockets, profiling, slots, drilling, even bosses - you can do this with $150 CAM. But obviously that won't tap into the potential your machine has.

Have someone design a mildly complex part for you, and program the toolpaths. Then show your plant manager the 350,000 lines of g-code (it would take about 10 reams of copy paper single-spaced) needed to run the part, and ask them, "Do you want to pay me for a month to type this all in, or get the software that will generate this g-code in about 15 seconds?" If you don't win this argument, it's time to move on...

[javanree]

Download a trial of CAMBAM, which is about the cheapest package I know of at around $120 and allows 40 sessions of full use before dropping to a 500 line limit, and do a side by side comparison of a few standard parts. This will quickly show your coworkers and management the improvement it makes.

[GeoffW1]

Download a trial of CAMBAM, which is about the cheapest package I know of at around $120 and allows 40 sessions of full use before dropping to a 500 line limit, and do a side by side comparison of a few standard parts. This will quickly show your coworkers and management the improvement it makes.

Hi,

I agree with this. I use CAMBAM to do modest 2.5 D geometry, and it is very capable with everything short of splines and full 3D. The geometry drawing (simple CAD) is good to use, and it has good capability with profiles, pockets, point lists. It has a selection of common post-processors, or you can write your own. Then if you use a good software package to actually control the machine (or it has one on board) you can inspect the tool paths offline and make minor edits. I use Mach3 which is old and creaky now, but our needs are simple. You can obtain better for sure.

If you need more power to create your geometry there is a wide range of choice.

That said, I would hate to have to go back to longhand G-Code programming. It just isn't a productive or economic thing to do with all the software that is available now.

Cheers

[RCaffin]

It depends on what you are machining, and whether you know how to program.
I make some fairly complex parts using 3 and 4 axes on the mill and on a lathe as well, and I do not own any CAM SW.
For the parts I am making, I would challenge any CAM package to be effective. It does not know what I am thinking or trying to do - and that includes machining up to 18 identical parts in parallel.
On the other hand, if you are carving faces in wood or something like that - buy CAM.

Cheers
Roger

[Jim Dawson]

Download a trial of CAMBAM, which is about the cheapest package I know of at around $120 and allows 40 sessions of full use before dropping to a 500 line limit, and do a side by side comparison of a few standard parts. This will quickly show your coworkers and management the improvement it makes.

I agree. CamBam is easy to learn and does a fine job of doing even complex parts. With some limitations it will do 3D profiles. Beats the heck out of hand coding, saves hours of conversational programming. You would see a massive productivity increase, not to mention the ability to machine more complex parts.

[rdnck68]

It depends on what you are machining, and whether you know how to program.
I make some fairly complex parts using 3 and 4 axes on the mill and on a lathe as well, and I do not own any CAM SW.
For the parts I am making, I would challenge any CAM package to be effective. It does not know what I am thinking or trying to do - and that includes machining up to 18 identical parts in parallel.
On the other hand, if you are carving faces in wood or something like that - buy CAM.

Cheers
Roger

I completely agree.... I work for a 70 million dollar per year company and I am the only programmer for our okuma lb3000exII MYW been working with our cad team to try to take some of the programming time out of the equasion and it's a huge hassle.... sure he gets me a great looking program complete with dynamic milling with about a million lines of code that I have to manually delete all of the feedrates because it thinks C axis can feed at 6000 degrees per minute.... every line of code that changes x or y axis MASTER cam gives it another feedrate.... so 4 hours later after deleting the feedrates (cursor over backspace backspace backspace backspace cursor down cursor over backspace backspace backspace backspace..... you get the point) standing at the machine I then have to delete all of the tool offset cancels and make sure my g50 isn't above 2400rpm. Then change all the nat11 nat12 call codes to something that makes sence for an aporator nturn , npart, npick.... Then after I do that I have to watch it run in animation to make sure it all looks good. And that is only for a main spindle part don't ask him to write a part and pick to transfer to subspindle it won't happen.... it's alot faster utilizing hand written macro programs for your parts. Luckily i do many parts that are very similar and use only four different macro programs that allow me to change 27 different variables to make each part. If that part doesn't need that part of the program put a zero in that variable and a strategically placed IF V25=0 GOTO NRUF2 skips it and continues.... as for your multiple parts a well placed vzshz=vzshz-v22-v23-.01 then a /goto..... master cam is a pain in the ass for me because it's only as smart as the keyboard operator who has no idea how to program an okuma multi axis lathe... for those of us that can write gcode on the fly kudos to you and this buds for you so grab your extra tall stool and plant your butt in it for an hour and get that last line of code written before your arm goes numb for the 10th time.... keep up the great work and remember you are a rare breed with a skill set that the next generation will never know.... sorry got a little carried away... can you tell I enjoy my career choice lol

Sent from my SM-G930T using Tapatalk

[RCaffin]

with about a million lines of code that I have to manually delete all of the feedrates because it thinks C axis can feed at 6000 degrees per minute....
I should not laugh - I apologise.

Since all 4 sides and the ends of these parts are machined, it takes about 6 separate programs (and 6 separate jigs too) to hold the parts. This is one of the more complex programs for these parts, machining 20 separate parts in one long hit. The code is about 9 kBytes long. Other programs for this part are smaller.

Same story for many other parts. I will skip the photos.

CAM could not do this sort of planning.

Cheers
Roger

[louieatienza]

with about a million lines of code that I have to manually delete all of the feedrates because it thinks C axis can feed at 6000 degrees per minute....
I should not laugh - I apologise.

Since all 4 sides and the ends of these parts are machined, it takes about 6 separate programs (and 6 separate jigs too) to hold the parts. This is one of the more complex programs for these parts, machining 20 separate parts in one long hit. The code is about 9 kBytes long. Other programs for this part are smaller.

Same story for many other parts. I will skip the photos.

CAM could not do this sort of planning.

Cheers
Roger

Depends on what CAM...

[RCaffin]

Depends on what CAM...
Educate me - what CAM?
And at what price?

To be sure, there is always Boris the Spider, but that cost $100k for the CNC and $100k for the CAM SW.
Even so, I have reservations.

Cheers
Roger

[dharmic]

Depends on what CAM...

And on what post.
And on whether the monkey driving the CAM is just a CAD guy who hit "print" - ie used some kinda best guess CAM wizard - or actually knows his arse from his elbow when it comes to optimal toolpath generation.

[louieatienza]

And on what post.
And on whether the monkey driving the CAM is just a CAD guy who hit "print" - ie used some kinda best guess CAM wizard - or actually knows his arse from his elbow when it comes to optimal toolpath generation.

Totally agree - it's the driver, not the car...

[louieatienza]

Depends on what CAM...
Educate me - what CAM?
And at what price?

To be sure, there is always Boris the Spider, but that cost $100k for the CNC and $100k for the CAM SW.
Even so, I have reservations.

Cheers
Roger

If you're a hobbyist, you can do Boris The Spider (which is an almost decade-old example, by the way) with DeskProto for $295. The Hermle 5-axis machine however, probably would set you back about three to four times what you estimated. You can machine it with most any 5-axis machine. But you've grossly over-estimated the software cost. You can start at around $15-20K for 5-axis continuous. You can even buy software that takes 3-axis code and converts it to 5-axis toolpaths. If I had to choose between that, and waiting for someone for a few years to hand-code Boris The Spider while the machine sits idle, just so he could impress himself, I think I'd choose the former.

Your example shows 20 identical parts... you could just program it once and use nesting to create the other 19 toolpaths; even some control software can do that. You could have all 20 oriented that way in CAM, with all the fixtures and the machine rendered as well.

I completely agree.... I work for a 70 million dollar per year company and I am the only programmer for our okuma lb3000exII MYW been working with our cad team to try to take some of the programming time out of the equasion and it's a huge hassle.... sure he gets me a great looking program complete with dynamic milling with about a million lines of code that I have to manually delete all of the feedrates because it thinks C axis can feed at 6000 degrees per minute.... every line of code that changes x or y axis MASTER cam gives it another feedrate.... so 4 hours later after deleting the feedrates (cursor over backspace backspace backspace backspace cursor down cursor over backspace backspace backspace backspace..... you get the point) standing at the machine I then have to delete all of the tool offset cancels and make sure my g50 isn't above 2400rpm. Then change all the nat11 nat12 call codes to something that makes sence for an aporator nturn , npart, npick.... Then after I do that I have to watch it run in animation to make sure it all looks good. And that is only for a main spindle part don't ask him to write a part and pick to transfer to subspindle it won't happen.... it's alot faster utilizing hand written macro programs for your parts. Luckily i do many parts that are very similar and use only four different macro programs that allow me to change 27 different variables to make each part. If that part doesn't need that part of the program put a zero in that variable and a strategically placed IF V25=0 GOTO NRUF2 skips it and continues.... as for your multiple parts a well placed vzshz=vzshz-v22-v23-.01 then a /goto..... master cam is a pain in the ass for me because it's only as smart as the keyboard operator who has no idea how to program an okuma multi axis lathe... for those of us that can write gcode on the fly kudos to you and this buds for you so grab your extra tall stool and plant your butt in it for an hour and get that last line of code written before your arm goes numb for the 10th time.... keep up the great work and remember you are a rare breed with a skill set that the next generation will never know.... sorry got a little carried away... can you tell I enjoy my career choice lol

Hey, on the bright side you have work, which is far better than the one in 6 in the US that are dependent on some form of aid. Then again, if you took the time to master MasterCam, with your hand-coding knowledge, you should be in the office programming these machines, instead of on the shop floor!

-----------------------------------------------------------

As an aside, a lot of folks program these 2.5D shapes using 3D toolpaths. Even the "high speed" 3D toolpaths produce a lot of point-to-point moves which can glut up a program. Some CAM have Z-level arc-optimization which will fit arcs in the X-Y plane during Z-level roughing and finishing operations. This greatly reduces lines of code immensely, and makes for smoother movement and finish to boot. Some CAM can do this in 3 axes, though some older controls may not handle it. There are optimization software that will take 3-axis code, and fit arcs even NURBS if the control can handle them. The starting price is a lot cheaper than I would have thought.