What's new
What's new

5-Axis Controls (Looking for a comparison of common controls)

mutiny

Cast Iron
Joined
Jan 27, 2019
Location
Raleigh
Hi friends,

I'm considering an eventual 5-axis machine tool purchase and would love to hear people's impressions of commonly (or uncommonly) available controls. Primarily interested in full simultaneous 5-axis machining, as opposed to 3+2, plus robot/automation integration. What's important, what should you look for, what should you avoid? Very briefly, here are the impressions I currently hold having never actually used any of these (I'm using Brother C00):

Fanuc
- Pros: Familiar to a lot of people with prior experience
- Cons: Old hardware/limited lookahead, clunky interface

Heidenheim
- Pros: More modern hardware/interface
- Cons: More expensive, potentially higher learning curve

Siemens
- Somewhere between Fanuc and HH, don't have a clear

Haas NGC
- Pros: Easy to learn/use
- Cons: Relatively low "power ceiling"

OSP-P300
- Some people love it, some people hate it
- Super-NURBS seems to be good, if expensive (like $15k???)
- Potentially difficult robot integration

Looking forward to hearing what folks have to say.
 
Are you planning to do the robot integration yourself or part of the machine package?

Makes a huge difference.

How much part variation are you planning for?
 
Usually these threads just turn up everybody chiming in to let you know that what they are using is the best.

For a slightly different take - I would probably start with the builder and then work backwards. For instance, I vastly prefer a Siemens controller to Fanuc, but Makino doesn't use anything other than Fanuc (AFAIK). So, if the Makino is the right fit for your application, all that really matters is - does the Makino/Fanuc do everything you need?

There are a few builders that will give you a choice, but then you really should let the builder help steer your decision. If a company does 50 HH integrations for every Siemens build, it would probably be foolish to choose the Siemens. In the year 2020 most major controllers can handle most applications with similar performance. Probably 90% of the time it just comes down to personal preference.
 
Boosted's point can be expanded on with consideration about how well any particular control is integrated to the machine. Fanuc on a Makino is going to be much better than Fanuc on many other brands.

Another factor is what you estimate the useful life of the machine to be. If it is 10 years then any of the controls will likely have decent support. If it is 15 years then support is probably going to be iffy for a couple of those manufacturers. If it's 20+ years then Fanuc will be the standout in the group.
 
Usually these threads just turn up everybody chiming in to let you know that what they are using is the best.

For a slightly different take - I would probably start with the builder and then work backwards. For instance, I vastly prefer a Siemens controller to Fanuc, but Makino doesn't use anything other than Fanuc (AFAIK). So, if the Makino is the right fit for your application, all that really matters is - does the Makino/Fanuc do everything you need?

There are a few builders that will give you a choice, but then you really should let the builder help steer your decision. If a company does 50 HH integrations for every Siemens build, it would probably be foolish to choose the Siemens. In the year 2020 most major controllers can handle most applications with similar performance. Probably 90% of the time it just comes down to personal preference.

I really agree with these sentiments ^^^ .

Although some shops don't want the "Fourth" too much, too many different controls on a shop floor vs. training.

So, if the Makino is the right fit for your application, all that really matters is - does the Makino/Fanuc do everything you need?


^^^ So for me that's a major point. I almost don't care what isn't "Great" about a Fanuc flavored control (interface wise) as it seems any hoops to jump through seem totally worth it.

Also if you get a handle on the Fanuc "Environment" then that gives you access to a lot of other good older iron / second hand machines. + the ability to fix stuff especially in the USA. (kinda dovetails with what @vancbiker said/ is saying.).

HAAS control for 5 axis kinda matches what their machines are mechanically capable of but seem to actually lag behind a bit. (low bandwidth). BUT on the the other hand if their controls were blisteringly fast + multi channel high bandwidth + very fast response times in any and all control / servo loop (with or without scales) then that might open up a whole Pandora's box of new things that they might have to address on the hardware side/ mechatronics + additional systems and even re-design of the "Iron" in some case .

Maybe in a few days I pull together some numbers for one thread or another HAAS vs. Heidenhain.

_____________________

@Mutiny no MAZAK Smooth X ? :-)

Or Mitsubishi ?
 
@Mutiny no MAZAK Smooth X ? :-)

Or Mitsubishi ?

The various Mazak flavors definitely seem interesting! I honestly was just listing what came to mind, didn't mean to be exhaustive. If not for the pandemic I'd likely have made a number of showroom trips by now to see these in person.

(The new C 600 / SmoothAI looks very interesting...)
 
Hi friends,


Fanuc
- Pros: Familiar to a lot of people with prior experience
- Cons: Old hardware/limited lookahead, clunky interface

Heidenheim
- Pros: More modern hardware/interface
- Cons: More expensive, potentially higher learning curve

Siemens
- Somewhere between Fanuc and HH, don't have a clear

Haas NGC
- Pros: Easy to learn/use
- Cons: Relatively low "power ceiling"

OSP-P300
- Some people love it, some people hate it
- Super-NURBS seems to be good, if expensive (like $15k???)
- Potentially difficult robot integration

Looking forward to hearing what folks have to say.

Fanuc is the standard that everyone compares to. Current 30i series controls are pretty fast, but the old Fanuc memory bugaboos are still there. There's a data server option, DNC, and reading from the card, but sooner or later, they're gonna have to jump memory. WAY up.

Haidenhain - REALLY fast. Unlimited number of work offsets and tool offsets. Hard drive for storage. Great 5 axis interpolation capabilities. Programming language isn't too tough, takes a little getting used to. GREAT documentation.

Siemens - VERY powerful. Lots of speed.

Okuma - IMO, and only my opinion, the best. Cycles, capability, speed, memory, just the best as far as I'm concerned.

Mazak - I saw the latest and it was unreal - a real speed monster! Haven't had my hands on one in a long time but looks like a control that can really boogey.

Mitsubishi - For some reason not as prevalent as years past, but the M700 looks like it could really chug data. You used to see a lot of these on multi axis machines in the past.

Just my 2c...
 
Hi friends,

I'm considering an eventual 5-axis machine tool purchase and would love to hear people's impressions of commonly (or uncommonly) available controls. Primarily interested in full simultaneous 5-axis machining, as opposed to 3+2, plus robot/automation integration. What's important, what should you look for, what should you avoid? Very briefly, here are the impressions I currently hold having never actually used any of these (I'm using Brother C00):

Fanuc
- Pros: Familiar to a lot of people with prior experience
- Cons: Old hardware/limited lookahead, clunky interface

Heidenheim
- Pros: More modern hardware/interface
- Cons: More expensive, potentially higher learning curve

Siemens
- Somewhere between Fanuc and HH, don't have a clear

Haas NGC
- Pros: Easy to learn/use
- Cons: Relatively low "power ceiling"

OSP-P300
- Some people love it, some people hate it
- Super-NURBS seems to be good, if expensive (like $15k???)
- Potentially difficult robot integration

Looking forward to hearing what folks have to say.

I have experience with Fanuc, Haas and just got into heidenhain a few weeks ago.

as of right now i would say HH 100000%, the learning curve was piece of cake, granted i dont know EVERYTHING, but enough to do all i need to run the machine, took 2-3 days to get familiar with it. stay away from haas.
we ran some parts on our fanuc machine (matsuura LX160) and had some weird surface marks. same program on mikron with heidenhain control and it looks perfect, i've seen this from other people too, seems fanuc smoothing is not quite as good as HH.
 
seems Fanuc smoothing is not quite as good as HH.

I'm totally with you, give me the German controller any day. When I worked for a builder, I used to tell folks that if they are making the leap into their first 5 axis, just ditch the FANUC. It's often easier to learn the new controller and 5 axis than it is to learn 5 axis on a FANUC.

However, I don't want to crap on FANUC too much. In regards to the quality of smoothing - there are a bunch of different smoothing options (for both controllers), and a lot of it is largely dependent on the MTB too. Plenty of folks making beautiful parts on high end FANUC controlled machines.
 
Okuma - IMO, and only my opinion, the best. Cycles, capability, speed, memory, just the best as far as I'm concerned.
Just my 2c...

Ughh... Still learning to hate mine again every time I get on it. Have fun editing large programs on the controller. Was just working on a 20mb file... The OSP software (at least on our OSP300) is too dumb to handle editing it, so you have to go to Windows and open it up in notepad.

Also, (and this isn't really Okuma's fault) I am SO sick of posting issues. With both Esprit and now hyperMILL the posts are garbage out of the box because there are very few Okuma power users out in the wild doing field testing.

The tool management is very "meh", and unnecessarily cumbersome if used as intended. Parametric programming is a little better than FANUC but has exponentially less support and terrible documentation.

I really want to like it. The OSP was bad ass 20 years ago... Too bad it never got better.
 
I'm totally with you, give me the German controller any day. When I worked for a builder, I used to tell folks that if they are making the leap into their first 5 axis, just ditch the FANUC. It's often easier to learn the new controller and 5 axis than it is to learn 5 axis on a FANUC.

However, I don't want to crap on FANUC too much. In regards to the quality of smoothing - there are a bunch of different smoothing options (for both controllers), and a lot of it is largely dependent on the MTB too. Plenty of folks making beautiful parts on high end FANUC controlled machines.

its not that the finish is bad, for most people it would probably look perfect. our boss is very particular on those things though, and the imperfections are just visual, probably couldnt even measure them with anything, just weird ghost marks. geometry wise it produces very accurate parts, if that makes sense.
 
Ughh... Still learning to hate mine again every time I get on it. Have fun editing large programs on the controller. Was just working on a 20mb file... The OSP software (at least on our OSP300) is too dumb to handle editing it, so you have to go to Windows and open it up in notepad.

Also, (and this isn't really Okuma's fault) I am SO sick of posting issues. With both Esprit and now hyperMILL the posts are garbage out of the box because there are very few Okuma power users out in the wild doing field testing.

The tool management is very "meh", and unnecessarily cumbersome if used as intended. Parametric programming is a little better than FANUC but has exponentially less support and terrible documentation.

I really want to like it. The OSP was bad ass 20 years ago... Too bad it never got better.

thats very interesting to hear as we have been looking into okumas...
 
thats very interesting to hear as we have been looking into okumas...

Just my opinion, so take it with a big ol' grain of salt. Plenty of folks on here like the Okuma controllers, despite what I would describe as an archaic design.

I still think they are certainly a quality machine tool builder. IMHO the controller is a big handicap, but not necessarily a deal breaker.

*edit* Also, I've now totally derailed my contribution into a discussion about "which flavor is best". It's hard to resist!
 
Ughh... Still learning to hate mine again every time I get on it. Have fun editing large programs on the controller. Was just working on a 20mb file... The OSP software (at least on our OSP300) is too dumb to handle editing it, so you have to go to Windows and open it up in notepad.

Also, (and this isn't really Okuma's fault) I am SO sick of posting issues. With both Esprit and now hyperMILL the posts are garbage out of the box because there are very few Okuma power users out in the wild doing field testing.

The tool management is very "meh", and unnecessarily cumbersome if used as intended. Parametric programming is a little better than FANUC but has exponentially less support and terrible documentation.

I really want to like it. The OSP was bad ass 20 years ago... Too bad it never got better.

I've never had a program too large to edit but I can't say exactly what size I have tried. There is a Mode B or something of the sort that is meant for larger program capacity that takes away the ability to run the quick edit function, have you tried that?
 
I've never had a program too large to edit but I can't say exactly what size I have tried. There is a Mode B or something of the sort that is meant for larger program capacity that takes away the ability to run the quick edit function, have you tried that?

Yes. We have to run in Mode B about 50% of the time, which takes away the editing capability. I think it's only about 2mb's file size limit for Mode A.

Switching back-and-forth probably wasn't a big deal 20 years ago when the ability to run huge-ass inefficient files was novel, but in the year 2020 it's fairly common to have surfacing toolpaths that chew up 500kb+..
 
Hi friends,

I'm considering an eventual 5-axis machine tool purchase and would love to hear people's impressions of commonly (or uncommonly) available controls. Primarily interested in full simultaneous 5-axis machining, as opposed to 3+2, plus robot/automation integration.

<snip>

<snip>

- Potentially difficult robot integration

Looking forward to hearing what folks have to say.

And HURCO ?

They give the appearance that integrating all manner of automation systems, pallets, robots etc. is not much fuss at all (Their Erowa integrations look pretty good). I don't know if that's true or not. + other things one might be able to do on the control ~ Even if you are not "big" on conversational controls seems good for other stuff+ prototyping i.e. good for transition from prototyping to production + a lot of editing capability on the control + motion control / dynamic / adaptive toolpaths - (trochoidal roughing) on the control.
 
Yes. We have to run in Mode B about 50% of the time, which takes away the editing capability. I think it's only about 2mb's file size limit for Mode A.

Switching back-and-forth probably wasn't a big deal 20 years ago when the ability to run huge-ass inefficient files was novel, but in the year 2020 it's fairly common to have surfacing toolpaths that chew up 500kb+..

We run a couple machines in mode B and can still edit in the program file screen, just not quick edit. We don't have to pop back out to notepad for anything.
 
Observations:

0. Collision detection can be Very Useful on 5-axis machines. HH has it (and has for a long time, my 2007 DMU60 has it) - I don't know what others do or don't have. If you plan on using something like vericut (or that general class) may matter less. Some systems will prevent collisions between tool and machine (no holes drilled in the trunnion) but not between tool and fixture. I would check this out before buying a new machine.

1. If the controller has a good "file system" AND a good link to your PC (via app, windows net share, web page, especially efficient carrier pigeon, whatever) - THEN - why not always edit using a real editor and real human friendly keyboard? If the controller has a file system that wouldn't have passed muster on a 1978 calculator and you transfer files over a telegraph wire, things are more difficult. So look VERY CAREFULLY at the DELIEVERED ability to transfer, name, and store files.

2. The heidenhain can be booted to speak "ISO" which will be some flavor of g-code, but I've never heard of anybody doing that. Note that what they call "conversational" (which is not Hurco style conversational) and which I call ".h language" is semantically the same as ISO g-code just with a nicer syntax. Meaning you write "L X+20.5 F400" instead of "G1 X20.5 F400". I wouldn't let worrying about the learning curve bother you.

3. I agree with the above about how quality and consistency of integration are a big deal - you probably don't want to own a unique configuration.
 
Heidenhain really is so well geared up for 5 axis, You can do 3+2 with a multitude of different ways to set working planes, TCPM with tool vectors etc which I know the fanuc, mitsubishi seimens etc can do too(not haas)
one of the best if not the best block processing time and insanely good "Dynamic precision".

If you are concerned about the learning curve, it really is so simple to use. it only took me a few days to get going.
for certain types of work probing with HH and I believe Siemens is awesome as its native to the control, not a renishaw gui bolton
another thing I find massively handy is that MDI mode is basically a program that runs in single block only, that may sound odd but i just leave all my commonly used positioning and probing codes in MDI and run them as needed.

Ive looked at a few MAZAK machines and the new Smoothx control looks to be really easy to use and is very powerful.(I belive they are Mistsubishi M80 backend)
 








 
Back
Top