What's new
What's new

Tool setting in shops. Dedicated setter vs in machine

huleo

Hot Rolled
Joined
Feb 12, 2014
Location
UT
I'd like to see how people are setting mill tooling? We have done it about every way, but considering an optical setter. However, in my quest in crash mitigation, I have concerns with out of machine setting mostly because I know too many shops where this has turned into very expensive machine damage. Whether that was from fat fingering the inputs, or just flat out wrong data.

We have some machines that do not have auto machine setters, but I tend to prefer that method as the value is recorded for you, and 'maybe' is a bit more fool proof?

Another concern I have with setting out of the machine is reliance of a taper device for the holder to sit in that make differ from how it sits in the actual machine spindle. If a spindle has been ground for instance, that tool will be a touch deeper??

Then you get into the costs of a good setter, all the different holder adapters, etc. Maybe this becomes the standard once you have X many machines?
 

Vancbiker

Diamond
Joined
Jan 5, 2014
Location
Vancouver, WA. USA
Why would one want to finger input the offset data from a presetter? Every presetter I have used could output a file with the offset info. Load the file into the CNC, press cycle start and the values are written to the offset registers.

This can be different depending on control brand. For example on Mitsu controls the offset values can be set just by loading the file.
 

huleo

Hot Rolled
Joined
Feb 12, 2014
Location
UT
In machine. Push a button. Or, tell it to measure a number of tools while you watch, or maybe not watch.
Do you typically just move around to various threads, give vague WTF responses, and role out? Asking for a friend.
 

Scruffy887

Titanium
Joined
Dec 17, 2012
Location
Se Ma USA
The machines prefer that human thinking be removed from the equation. Machine controls will soon become aware. Shops of the future will have only a dog and a man. Man is there to feed the dog. Dog is trained to keep the man from touching the machinery controls.
 

spinninquin

Plastic
Joined
Oct 1, 2015
Location
USA
We have both tool setters in the machine, and tool setters outside that require manual entry. The issue we see across the board are lazy operators. If an endmill is held in a side lock, they don’t run the probing routine because it’s “in the same spot as the last one.” The same thing happens with the setters outside the machine. This doesn’t lead to crashes, but does scrap parts quite often due to features being too deep/shallow.

I personally prefer the setters inside the machine. I program at my desk for a group of machines and can easily write a quick program to check the offset of all the tools in that program for the first run.
 

bryan_machine

Diamond
Joined
Jun 16, 2006
Location
Near Seattle
The offline tool setter can have other advantages:
a. On a machine where access to the in-machine tool setter may be blocked by the setup, the offline tool setter is not affected. (Part of what drove my getting an offline tool setter.)
b. I think it's an easier way to set boring bar initial diameters, to measure out-of-holder lengths when that is critical.
c. The tool setter can be a superior way to examine tools - to measure some aspects of wear, to investigate chipping, etc.
d. Mine is manual data collection - but for mere money you can get fancier units that write data to files, rfid tags, etc.

Issues with tool setters:
1. you DO have to work out wether the gauge length reported by the tool setter matches the gauge length reported by an in machine setter - but but either way you have to sync them up with the actual behavior of spindle probes etc. I wouldn't expect this to be a big deal, but it does require working quietly for a little while, so if the shop is a mad house it may be hard.

2. the offline tool setter is not an effective break, or runtime wear, detector.

3. space and money
 

PROBE

Hot Rolled
Joined
Jan 23, 2003
Location
Tel Aviv, Israel
The “setting tools” definition includes the tool measuring, offset registers update and last but not least - insertion of the new tools into the magazine. These activities are time consuming. This time (exactly as part probing) is “per definition” machine's "not productive (chips making)" time. Some very fancy machines have tool magazine changers, which in matter of seconds insert into machine the complete new measured tools setup. But this is very rare in the real world. In the real world we fight to make this process as short as possible, yet not jeopardizing its reliability. The reliability is mainly based on amount of human based involvement / interference in the process. The less human involvement, the higher reliability.

The process includes following steps:
1. Measuring the tools in remote setter, marking the tools (either just numbers, or dimensions too).
2. Loading the tools to appropriate pockets in the tool changer magazine.
3. If not transferred in file, dialing the tool data into offset registers.
4. If remote preset not executed, measuring the tools on in machine tool setter.

Obviously the minimum of human involvement is achieved while the tools are measured in the machine, using wisely written routine, in which the tools are loaded through machine spindle as a part of the program. Program stops and the message on machine screen instructs the operator which tool number should be loaded to the spindle. This way the steps 1, 2 and 3 in above mentioned list are omitted, and obviously higher process reliability is achieved.

I haven’t the opportunity to compare the time needed to set the machine with 20 new tools using the described method versus external setter. The difference does not seems to me to be substantial, as loading tools to tool magazine is cumbersome.

And being PROBE has not any influence on my conclusions. For breakage detection tasks the setter will be installed in machine anyway.

Stefan

Cogito Ergo Sum
 

DMF_TomB

Diamond
Joined
Dec 13, 2008
Location
Rochester, NY, USA
I'd like to see how people are setting mill tooling? We have done it about every way, but considering an optical setter. However, in my quest in crash mitigation, I have concerns with out of machine setting mostly because I know too many shops where this has turned into very expensive machine damage. Whether that was from fat fingering the inputs, or just flat out wrong data.

We have some machines that do not have auto machine setters, but I tend to prefer that method as the value is recorded for you, and 'maybe' is a bit more fool proof?

Another concern I have with setting out of the machine is reliance of a taper device for the holder to sit in that make differ from how it sits in the actual machine spindle. If a spindle has been ground for instance, that tool will be a touch deeper??

Then you get into the costs of a good setter, all the different holder adapters, etc. Maybe this becomes the standard once you have X many machines?
.
optical tool setter advantages
.
1) in theory you can be doing tooling while the cnc is running, its one thing only a few tools but
if doing 10 to 40 tools it can be time consuming obviously
.
2) optical tool setter usually has a way to calibrate whether a test bar tool or it looks at base of
the optical tool setter measuring calibration balls on side made for calibration. many tool setters
want calibration checked every hour it only takes maybe 10 seconds usually
.
3) most optical tool setters are connected to cnc network where tool data is updated automatically.
yes where this auto network feature not working and manually entering data is not fool proof. A excel
spreadsheet with tool data and history if printed out you can see tool length / dia or radius offset
previous values. on carbide insert tooling and usually taps you can see if there is a big change
from previous values (date column part of the row.) also excel can record max length and max diameter
and do the math automatically this can catch math mistakes that were done manually. the tooling data
excel spreadsheet can have date and time at the top header every page printed out daily you can see
easy if you got old spreadsheet. this is done if you got say a 120 tool magazine. obviously if this done
automatically as part of tool setter software it makes things easier
.
4) optical tool setter can catch runout problems and dirt or dust on tool, fully automatic optical setters
can read dust or dirt rather than tool edge. this can be a problem with laser setters as optical you
can see the dirt and wipe off or use sticky clay to remove it. usually optical tool setter is semi automatic,
operator has to check for dirt dust and or cutting edge damage
.
5) optical tool setter usually do not trust less than +/-.003" yes the cnc spindle can have runout issues. also
spindle taper and spindle face might be off. that is dual contact tool holders might stick out farther than
tools just in the taper especially if taper reground but the spindle face was not reground. grid shift checks
with a taper only test bar and a dual contact test bar usually can measure the difference. if its a boring bar
you usually always take a test cut unless you got a big tolerance
 

huleo

Hot Rolled
Joined
Feb 12, 2014
Location
UT
I think one of our realities is not all our machines have a setter. It's easy enough to manually set tools in a VMC, but not nearly as much fun in an HMC.

I am just juggling the possible problems that can arise, and ones I have seen in other shops such as somehow the wrong length data ends up in the machine, sometimes due to the confusion of tool numbers being in random pots.

Even though it is more labor intense, at times I prefer to call for a specific tool in MDI, then load the spindle with that tool. I also typically request an M01 after every tool change so during a first run, someone is forced to visually look at the tool and verify it is the right general geometry, then hit go. Once things are proved out, turn off Ostop.

I am more or less trying to formulated a standardized procedure that can work for 'most' machines. I say that because some things are just special, like lathes and boring tools.

I think most bigger shops try to stick with a standardized tooling line to help in programming and setup. While I agree with it, we seem to just keep running into required tools that are not loaded and if the mag if full, something has to come out. Then we have tools out of the machine.
 

Orange Vise

Stainless
Joined
Feb 10, 2012
Location
California
We have a dedicated optical presetter as a backup for the entire shop. Its primary use these days is to check runout on special tooling like critical finishing tools.

By and large though, tools are set in the machine with an operator watching it. We have programs written to automatically call tools, force the operator to swap a spent tool one-for-one with a redundant, and watch it touch off and reset the tool life counter. On average it takes less than a minute per tool. The 5-15 minutes of downtime is negligible compared to the 12-22 hours the machines run on a daily basis. The peace of mind knowing that a tool has a correct offset is invaluable.

Now for a rant: I'm pretty disappointed that RFID tagged toolholders never caught on. This is the fault of the machine tool builders charging way too much to add the option. These were available when I first started machining 20 years ago. The promise was that you can set a tool offline and then feed the holders back into the magazine in any order, and the machine would figure out the rest. I would've bet money back then that RFID would be mainstream by now, but sadly they're still a rare, expensive option, which makes no sense at all.
 

huleo

Hot Rolled
Joined
Feb 12, 2014
Location
UT
Orange, I think you and I are on the same wavelength. however, without a setter in the machine, the manual process can be a pain. But there is a level of safety where you are visually looking at the tool in the spindle, you set it, and you have that sense of assurance. As I look to adopt a better process, I also know that any error can cause way more than a few wasted minutes. Like KO the whole machine and she is down for weeks.

You mention runout in the optical though. Does this mean they have an AC or air bearing in the fixture to rotate the tool? I have realized that I am one of the few that drags an indicator into a machine to check runout. Almost no one checks it, and it is very important, especially when trying to chase deviations in feature size, tool life, etc.

But can you really get all the runout confirmed out of the machine? I mean, the spindle taper could contribute, or other variables I suppose. But I think some of our issues in runout have stemmed from collet systems and it has been quite a time waster to stick a tool in, test it, remove it, change something, retest, etc. It's also a PITA trying to test this.
 

bryan_machine

Diamond
Joined
Jun 16, 2006
Location
Near Seattle
The “setting tools” definition includes the tool measuring, offset registers update and last but not least - insertion of the new tools into the magazine. These activities are time consuming. This time (exactly as part probing) is “per definition” machine's "not productive (chips making)" time. Some very fancy machines have tool magazine changers, which in matter of seconds insert into machine the complete new measured tools setup. But this is very rare in the real world. In the real world we fight to make this process as short as possible, yet not jeopardizing its reliability. The reliability is mainly based on amount of human based involvement / interference in the process. The less human involvement, the higher reliability.

The process includes following steps:
1. Measuring the tools in remote setter, marking the tools (either just numbers, or dimensions too).
2. Loading the tools to appropriate pockets in the tool changer magazine.
3. If not transferred in file, dialing the tool data into offset registers.
4. If remote preset not executed, measuring the tools on in machine tool setter.

Obviously the minimum of human involvement is achieved while the tools are measured in the machine, using wisely written routine, in which the tools are loaded through machine spindle as a part of the program. Program stops and the message on machine screen instructs the operator which tool number should be loaded to the spindle. This way the steps 1, 2 and 3 in above mentioned list are omitted, and obviously higher process reliability is achieved.

I haven’t the opportunity to compare the time needed to set the machine with 20 new tools using the described method versus external setter. The difference does not seems to me to be substantial, as loading tools to tool magazine is cumbersome.

And being PROBE has not any influence on my conclusions. For breakage detection tasks the setter will be installed in machine anyway.

Stefan

Cogito Ergo Sum
Actually there's one counter argument:
If you measure the tool offline, and the machine is like some 5-axis machines where you can load a tool through a side door while the machine is running, and the controller let's you update the offset entries while running, you could build/measure/load tools in parallel with a program running.
I don't know how often this works out in practice....
 

huleo

Hot Rolled
Joined
Feb 12, 2014
Location
UT
Actually there's one counter argument:
If you measure the tool offline, and the machine is like some 5-axis machines where you can load a tool through a side door while the machine is running, and the controller let's you update the offset entries while running, you could build/measure/load tools in parallel with a program running.
I don't know how often this works out in practice....
We have this, and I think pretty common among real production HMCs. However, for obvious safety reasons, we cannot access the door when the machine is in motion, but we can otherwise job the mag and insert tools at will. It then becomes critical to update the offset data. Gets a little risky for my blood. I think it would have been a smart approach to allow us to simply flag tools as "new" meaning they need checked before they run.
 

CAMasochism

Hot Rolled
Joined
Jun 6, 2019
Location
DFW, Texas
Actually there's one counter argument:
If you measure the tool offline, and the machine is like some 5-axis machines where you can load a tool through a side door while the machine is running, and the controller let's you update the offset entries while running, you could build/measure/load tools in parallel with a program running.
I don't know how often this works out in practice....
This is how we do it where I am at. While one job is running the tool room guy is gathering the tools for the next job, setting them up and setting the tool lengths. When it comes time for the next job it's just a matter of unloading the tools from the previous job and loading in the new ones.
 

LockNut

Stainless
Joined
Jan 6, 2007
Location
Bergen County
I use both on a daily basis. Yeah, I work for a MTB. Our machines have both a GUI method of setting tools and also in process methods, which can be more important in some cases. But I can set tools in an off line tool setter faster than I can in the machine. I can also set all tools for the next job while the current job is running. And yes, you must be vigilant in entering the offsets manually. But one of my tricks is to calibrate the off line tool setter with the same calibration master that I use to calibrate the machine tool setter. Both machine and off line tool setters match within a few tenths of an inch.
 

CAMasochism

Hot Rolled
Joined
Jun 6, 2019
Location
DFW, Texas
But one of my tricks is to calibrate the off line tool setter with the same calibration master that I use to calibrate the machine tool setter. Both machine and off line tool setters match within a few tenths of an inch.

This is what we did with our new Mazak, while they were setting it up and calibrating it with their test bar we asked them if we could borrow it and used it to set our tool setter so it's numbers would match the Mazak's.
 








 
Top