Movespeeds

[A-machine-insp]

Just as a general question, what movespeeds do you guys (and gals) use and do you use percentages or absolute?

I run absolute speeds on everything and, generally, between 150-250 mm/s depending on the part. I have one family of parts (one part that is actually running now) that I run full speed 500 mm/s but that is a part that I programmed, I'm the only one who has ran it, and it's a well proven program. Vary rarely (and I mean VARY rarely) do I run 500 mm/s.

[louisd]

I inherited percentages. I still have yet to convince my site to convert to absolute, as it would force a re-MSA for ALL validated measurement routines. Which is a ton of red tape.

Everything I write starts with a prudent movespeed %(like 40), for fixture alignments and initial hits.
Once part is located, movespeed is set to 100%. I also inherited some on-site belief to keep jogbox speed at 80%, so this 100% movespeed isn't necessarily WOT (wide-open throttle).

Just a quick reminder, movespeed has nothing to do with touchspeed. so... you can set it to whatever you want it to be set.

We have hardware maintenance agreements, and seem to burn through drive motors at a rather alarming frequency (we've burned at least one motor on each of our three CMM's in the 3 years I've been here). If we didn't have the HMA, I would push to slow them down considerably.

[A-machine-insp]

Doesn't the jogbox speed effect all speeds, even touch speeds? To the eye, it looks like it does to me.

[louisd]

Yes, jogbox will affect touchspeed & movespeed.

[Kp61dude!]

Oh how that must suck! You see something you can improve but you can't, dang those MSA and bless your patience louisd

[bfire85]

I inherited percentages as well.

[Mike Ruff]

So do you guys always run the same parts on the same CMM? Or do you have different copies of the same program with different movespeeds saved for each CMM? Or do you use some variables and ask which CMM it's running on with an Input comment to control movespeed?

[A-machine-insp]

Mike, personally I don't because whatever the movespeed is set at, the machine won't run over its max speed. That being said, the programs that are set with the fast move speeds usually are only run of the machines that can do those speeds. Occasionally I will run a fast program on our big machine that will only do 80 mm/s but it will only run at 80 mm/s.

[louisd]

Mike Ruff, we validate each routine on each machine via MSA (Gage R&R) or comparison study. Typically the routine is unaltered. As long as we have <10% dimensional variation between machines, we are good.

[SingularitY]

We are a high production environment with a lot of 100% sorting so I have my programs cranked to 500mm/s. I do like the idea of lowering the move speed for the initial part pickup then, once the machine locks down all 6 DoF, increasing increasing the move speed. I'm stealing this idea louisd!

[Quality ish]

I inherited percentages. I still have yet to convince my site to convert to absolute, as it would force a re-MSA for ALL validated measurement routines. Which is a ton of red tape.

Everything I write starts with a prudent movespeed %(like 40), for fixture alignments and initial hits.
Once part is located, movespeed is set to 100%. I also inherited some on-site belief to keep jogbox speed at 80%, so this 100% movespeed isn't necessarily WOT (wide-open throttle).

Just a quick reminder, movespeed has nothing to do with touchspeed. so... you can set it to whatever you want it to be set.

We have hardware maintenance agreements, and seem to burn through drive motors at a rather alarming frequency (we've burned at least one motor on each of our three CMM's in the 3 years I've been here). If we didn't have the HMA, I would push to slow them down considerably.

From everything I've read, your touch speed needs to be the same checking the part as it was when the tips were calibrated. Any change, one way or the other could lead to inaccurate measurements. For example, if the probes were calibrated at a speed of 3, you certainly wouldn't want to be touching off your parts at a speed of 5... I can only imagine that even changing from touching the sphere at 2, and then checking parts at 80% of that (1.6) would have some small effect.

In the long run though, the jogbox pretty much just regulates what percentage of the speed the program will be run at. If you don't want a move speed of 100, but rather at 80, why not just program it that way?

Sounds like somebody was milking a program for time, got caught, and convinced "the powers that be" that that was the way it should be done.

[A-machine-insp]

We have a guy here that calibrates with the jog box at 100% then runs the program at 50% cutting the touch speed as well. The boss and I have tried to get him to understand and change but to no avail. I am with your thinking though. The most I do is, if I'm coming up to a questionable area, I'll either cut the movespeed down or turn the jog box down during the move then back to 100% for the hit.

[louisd]

"Just a quick reminder, movespeed has nothing to do with touchspeed. so... you can set it to whatever you want it to be set." "It" being movespeed, not touchspeed. The subject of the post was movespeed. So yes touchspeed is always at 2 in my routines, unless we are using a probe smaller than 1mm, than it's calibrated and running at 1. I don't know the backstory of the 80% jogbox, but it is basically a fixed value when running production parts. Some routines will actually produce significantly different measured values when jogbox is increased. Since they were all validated with jogbox at 80, jogbox stays at 80.

honestly, I think the guy before me was 'faking it till he made it', there's a lot of rookie mistakes all over his code. decimals & hyphens in feature id's, multi-position fixtures, with unique code (theo's not matching cad) for each fixture position, you name it. Some routines rotate the PH10 like 40+ times to measure four small 1" diameter x 6" long parts. I re-create the routine, and probe rotates 6 times (a few of which are inclusive of probe changes). lol.

[Quality ish]

The problem is: the jogbox regulates the entire running of the program, touches included. So if by chance the jogbox was set at 100% when calibrated, and 80% during the running of a program, you've got a disparity.

Believe me, I understand someone trying to figure things out as they're moving forward, with a lack of formal training. I'm much in that same boat. I'm still trying to figure out what I don't know, so that I can improve on what I do know. You know?

[mb0258]

my view:

same movespeed/touchspeed during calibration as max meas. routine speed, jogbox during calibration at 100%, so you have covered all range from 0 to 100%, this way you should be accurate even when than measuring at jogbox or movespeed lower than 100%, but keep touchspeed the same.

[Quality ish]

Again, the jogbox will change the speed of every aspect of the program, touches included. It could affect the measurements, if it were set to anything other than what it was during calibration.

[Matthew D. Hoedeman]

Coding to use either of the following

500/6/1000 (move, touch, accel) ~ Global
150/2.5/n-a ~ Validator

[A-machine-insp]

For me the move speed changes but the touch and accel are the same as I run on the global.

[Matthew D. Hoedeman]

Can't do that with the Validator, a 6mm touch speed would cause the head to un-seat if the touch was normal to the probe. BIG heavy machine that can't stop at that speed in less than 4 or 5mm

[KIRBSTER269]

100% or however MM per second the machine can handle, and Fly Mode.

gif_drift_5[1].gif

[A-machine-insp]

I LOVE me some fly mode. I've been starting to incorporate move/all more and more too.

[louisd]

oooh yeah, flymode is noice! use it all the time

[Quality ish]

I've recently fired up an old Sheffield Discovery 12 with 2017 R2 on it, only to find out that it's not capable of fly mode. Yuck.

[Kp61dude!]

Just as a general question, what movespeeds do you guys (and gals) use and do you use percentages or absolute?

I run absolute speeds on everything and, generally, between 150-250 mm/s depending on the part. I have one family of parts (one part that is actually running now) that I run full speed 500 mm/s but that is a part that I programmed, I'm the only one who has ran it, and it's a well proven program. Vary rarely (and I mean VARY rarely) do I run 500 mm/s.

How much speed? All of it!!!

I run absolute speeds, I inherited a fairly new setup that is automated and locked away from everyone except for myself and a robot, so I got to Bob Ross the entire thing from basically scratch. Everything and anything lives in Sub routine library including speeds (although adjusted to pick up fixture ie. large pre/retract distance, slower touch/movespeed, etc.).

I add the following after initial pick up:

            MOVESPEED/ 500
            FASTPROBEMODE/ON
            FLY/ON,0.1188
            CHECK/0.15,1

[A-machine-insp]

Can you explain the fast probe mode? I know it works on some setups and not on others. I run all Renishaw, 1 PH10MQ/SP25-M, and the others are PH10MQ/TP20 and PH10T/TP20 setups.

Again, love me some fly mode.

[SingularitY]

Fastprobemode allows a non-touch trigger machine, can't recall the proper terminology, act like a touch trigger machine. Touch trigger takes the hit when the probe hits the part at a certain pressure. Non-touch trigger takes the hit after it hits the part, and as it moves away from the part, slowly, once it reaches a certain pressure. This allows the machine to more accurately determine when the point should be taken.
I use fastprobemode for my initial "find part" alignment and features with open tolerances. I turn it off for probing my datums and tight tolerance features.

[Kp61dude!]

A-machine-insp


From the help menu...

This probing mode applies to analog scanning probes when taking individual hits rather than scanning. When it is ON, it provides a significant reduction in the time required to take individual hits. The biggest reduction in time occurs when the probe retracts from the part.

Note: Some probe types (such as the SP25) do not support Fast Probe Mode. Some machines do not support Fast Probe Mode regardless of the probe type. If you have a measurement routine with the FASTPROBEMODE/ON command in it, but your probe type or machine does not support it, the command will not do anything.

End help menu note.

I use an lspx sensor. It is much slower at taking points than the alternative sensors, fastprobemode makes it act like a 'normal' sensor.

[A-machine-insp]

Cool. I'll try it on the TP20's and see if it quickens it up a bit. Obviously it won't do anything for the SP25-M.

[Matthew D. Hoedeman]

Won't help, the TP20 is NOT an analog probe, it only effects analog probes.

[measurestuff413]

Is there a way to variable-ize move speeds?

After reading this thread, I made the little code below. Something like this seems like it would be useful. I just tried it and it doesnt work =/

C1         =COMMENT/INPUT,NO,FULL SCREEN=NO,
            What machine?
            IF/C1.INPUT=="1"
              ASSIGN/V1=500*0.9
            END_IF/
            IF/C1.INPUT=="2"
              ASSIGN/V2=350*0.9
            END_IF/
            ASSIGN/MOVESPEED=V1
            MOVESPEED/ MOVESPEED

[dph51]

MOVESPEED is probably a reserved name.

[louisd]

and V2 will never work as it's not referenced. Keep input 2 assignment labeled V1 and it should work, as long as the movespeed assignment is renamed (if that's a valid contributor to why it no-workey).

[Tinyfxds]

We used percentages but switched to absolute. Half the time the guys would slow down the jog box because they wanted to slow the machine down but then the touchspeed would also be affected by that. That created really long run times. So we switched to absolute so we could set our movespeed to an actual number without slowing down the touchspeed. This has been working well for us for quite some time.

[dph51]

Try this, note changes.

I

C1 =COMMENT/INPUT,NO,FULL SCREEN=NO,
What machine?

IF/C1.INPUT=="1"

ASSIGN/V1=500*0.9

END_IF/

IF/C1.INPUT=="2"

ASSIGN/V1=350*0.9

END_IF/

MOVESPEED/ V1