What's new
What's new

Renishaw probing accuracy

tcncj

Cast Iron
Joined
Dec 15, 2016
Hi all,

I've got a question regarding Renishaw probing.

We've bought a new Hurco VM10i since a month with the complete probing package (TS27R and OMP60)
There we're allot of issues with installation and some issue are still present, due to a lack of knowledge from the service guy. Total disaster.
I shall only tell the issues regarding probing, since that's where this topic is about. And otherwise the list will become too long.

- renishaw workprobe stylus not adjusted so it had 0,03mm (0.118")runout.
- toolprobe was not aligned on the table

This was fixed today by a different service guy.
But there are still issues.

When checking diameter with a pin gauge in a collet the measurements are not consistent.
When repeating the diameter probing the maximum difference between measurements was 0,08mm (0.315")
That's huge!
When adjusting the probing speed he could reduce the difference to within 0,02mm (0.08)

And sometimes it gives an error about the max spread is exceeded.

According to the Renishaw manual it should repeat to 0,001mm (micron) -

Above measurements done with the pin gauge not rotating.
When using an endmill or facemill and rotating it, the numbers are way! not usable.
So all this was done with the tools not rotating.

I highly doubt this is correct. And the service guy is trying to let me think this is normal.

He says non of their customers use it the way how I want to use it (rotating a cutter and measuring the tool by length and diameter)
And they only use it to check for length and broken tool detection. He called Hurco Germany and they said it was normal. Wtf? why spend 10.000 Euro on a probing package. Even my digital caliper is giving more consistent results.

He also says the readings vary because of the machine is hot.
It's been sitting idle all day and did only 20 probing cycles...total bs

Then he said it was due to machine accuracy.
That the toolsetter could repeat in a micron doesn't mean the machine can.

True...but I'm sure the machine is accurate enough to repeat within 0,02mm / 0.078"
And that wouldn't explain the variations between readings. And the slight decrease in variations when using a lower probe feed.


So I'm checking if I'm going crazy or have a mechanic that doesn't know how it works?
Why would you even probe diameter of the readings are way off and vary each time? Why would you want to update the tool wear like this?

We didn't checked the tool length yet.
But before the aligment of the TS27R there where also big variations in readings when the tool was rotating.
 
Has it been wired up correctly? There are two ‘fast’ probing inputs at TB4-30 & 31 that if they are not present or transposed, will cause issues. If the control parameters are incorrect, it may cause problems.
As for the part probe, it doesn’t matter if the stylus runs out as long as the ring gauge deflection offsets have been done.
 
Tool probe
on the tool rotation and not repeating with endmill, I am assuming you are rotating it in m4 IE reverse. if not thats one of your problems.

I Should have made it more specific
you dont rotate the tool when checking the height only the dia.
 
Has it been wired up correctly? There are two ‘fast’ probing inputs at TB4-30 & 31 that if they are not present or transposed, will cause issues. If the control parameters are incorrect, it may cause problems.
As for the part probe, it doesn’t matter if the stylus runs out as long as the ring gauge deflection offsets have been done.

I don't know if the inputs are correct. I didn't wire up the machine haha

Part probe should be fixed now.
Tool probe is an issue

Tool probe
on the tool rotation and not repeating with endmill, I am assuming you are rotating it in m4 IE reverse. if not thats one of your problems.

I Should have made it more specific
you dont rotate the tool when checking the height only the dia.
Yes it's turning the correct way.
Otherwise it would eat up the TS27R ;)

I'm told you do rotate it on bigger endmills or insert bodies. Because some insert may be a bit lower compared to the others.

The difference in measurements is still present when not rotating.
But once rotating it's waaaay off.
 
.03 mm is equal to .00118”. .08mm is equal to .00314. You may be closer then you think. Remember 1mm is equal to .03937. Which means to get to inch from mm you divide by 25.4 NOT BY .254 From what I am understanding from your post is that the runout of your probe is slightly off.

You can fix that by taking a test indicator and putting it in an indicator stand or mag base of some sort and check the run out of the ruby,just like you would check the runout of an endmill/drill. If it is off(which it most likely is regarding your post) you need to tweak the set screws on the probe body until you get an acceptable runout on the tip of the ruby (.0002 or less). Once this is achieved you need to run the calibration program to re-calibrate the diameter of the ruby until you get the correct readings in your macro numbers for “size”...and “Run-out”. When I say size i mean that the probe must output the correct size of the ring gauge you are using to check within your acceptable tolerances. The runout is the difference between the contact points in “X” and “Y” when the probe rotates 180 degrees to check the ring gauge again for the second time.

I hope this helps a little.
Ive struggled with probes for awhile, but once you understand the entire concept of how they work mechanically and all the human error involved in getting the probe accurate to your needs, you wont trust it.

When you do finally get your “AH HAH” moment and realize that the probe is accurate and consistent you wont be disappointed!!!

Happy Fourth of July!!!!
 
When you say probing do you actually mean tool setting using a non contact tool setter?? Probing means using the actual tool probe to measure part geometry after it has been cut and also setting work offsets to the part. I just replied to your post thinking “tool probe” was the issue, bit it actually sounds like you ate having trouble measuring your tool diameter and length????? This is due to your laser not being aligned correctly and incorrect calibration after running the alignment program.
 
I used several TS27's on lathes(and mills) , and they were always repeatable within a few tenth's(.0001-.0003'ish). The only times there was variances was when someone hamfisted the toolsetter arm when lowering them and it would slightly move the base(cheap Haas setups). If done carefully and no slamming when bottoming out or on newer automatic arms they stayed true. In the mills , endmills were turned for setting length and diameter(run backwards) except for the indexable facemills which were manually touched and I routinely touched every insert so I could shim or change inserts to get consistent lengths but again touching the same insert 5 times in a row the setter probe would read +/-.0001-.0002"max variation. Infeed speed always had some influence on repeatable but it was very small.

BUT>>> the place I'm at now the probe doesn't repeat nearly as good , common to vari +/-.0005-.0008" but I'm thinking it has more to do with the machine/control inaccuracy than the probe. That probe spins to orient so the same side of the ruby always touches so any tip run-out becomes irrelevant(supposedly)

.
 
1. If the results of measurments with OMP60 are OK, then your machine repeatability is OK and you can disregard any clause about it.
2. TS27R is a fine and reliable instrument and normally works satisfactory. Nevertheless, the specifications claim of 0.001 mm repeatability is for results achieved in the laboratory, while micrometer of 0.0001 mm repeatability is moved against the probe. It has nothing to do with the real world. But normally repeatability of 0.01 mm is seen.
3. I am not familiar with Hurco programming. In order to check if your TS27R is performing correctly I, rather then using measuring routines existing in your control, would write small program, which in Macro B looks like this:

%
G91
M** (SWITCH TO TOOL SETTER)
#1=1
#120=#5021 (REGISTER CURRENT Z POSITION IN MACHINE COORDINATES)
WHILE [#1LT10]DO1
G31Z-5.F30
G53
#[100+#1]=#5063
G1Z[#120-#5021]F1000
#1=#1+1
END1
M30
%

Place tool (best would be 3 mm drill) roughly 1 mm above the surface of TS27R stylus and run the program. The tool will touch the probe 10 times, coordinates of each touch position will be stored in variables 101-110. The scatter should be in the range of 0.01 mm. If this is OK, you probably have problems with the measuring routines or with settings.
 
....
When repeating the diameter probing the maximum difference between measurements was 0,08mm (0.315")
That's huge!
When adjusting the probing speed he could reduce the difference to within 0,02mm (0.08)
.

Why did reducing the speed help by a factor of four and how slow did he/you try?
Did the error curve still have fliers?
Bob
 
1. If the results of measurments with OMP60 are OK, then your machine repeatability is OK and you can disregard any clause about it.
2. TS27R is a fine and reliable instrument and normally works satisfactory. Nevertheless, the specifications claim of 0.001 mm repeatability is for results achieved in the laboratory, while micrometer of 0.0001 mm repeatability is moved against the probe. It has nothing to do with the real world. But normally repeatability of 0.01 mm is seen.
3. I am not familiar with Hurco programming. In order to check if your TS27R is performing correctly I, rather then using measuring routines existing in your control, would write small program, which in Macro B looks like this:

%
G91
M** (SWITCH TO TOOL SETTER)
#1=1
#120=#5021 (REGISTER CURRENT Z POSITION IN MACHINE COORDINATES)
WHILE [#1LT10]DO1
G31Z-5.F30
G53
#[100+#1]=#5063
G1Z[#120-#5021]F1000
#1=#1+1
END1
M30
%

Place tool (best would be 3 mm drill) roughly 1 mm above the surface of TS27R stylus and run the program. The tool will touch the probe 10 times, coordinates of each touch position will be stored in variables 101-110. The scatter should be in the range of 0.01 mm. If this is OK, you probably have problems with the measuring routines or with settings.

1) exactly! that's what I said to the mechanic. The OMP measures perfectly! tested/calibrated on a ring gauge. So it's not a machine issue.
Besides I did measure backlash on all axis before the mechanic came in. To make sure it was not the machine. And it indeed had backlash...
He then tested it with ball thing and a program to let the machine interpolate. He got some readings on his notebook and adjusted the backlash in the software.

2) Yep 0,001mm is in best case scenario. But the variations I get are way too big. It was +/- )0,02mm at the end of the day. Sometimes a negative value and sometimes a positive value. Measured with a gauge pin.
3) I will try that next week

Why did reducing the speed help by a factor of four and how slow did he/you try?
Did the error curve still have fliers?
Bob
He did run it at 25mm/min.
And then at 4mm/min.
Yes it still had big variations between measurements
 
1)
He did run it at 25mm/min.
And then at 4mm/min.
Yes it still had big variations between measurements

Hello tcncj,
Both of those feed rates are classified as slow; one very slow and the other painfully slow. It would be extraordinary, for either to make any difference over the other.

Regards,

Bill
 
What's a decent touch off feedrate for the ts27r?
The 25mm/min is the Hurco standard setting
 
What's a decent touch off feedrate for the ts27r?
The 25mm/min is the Hurco standard setting
Hello tcncj,
The 25mm/min is OK. I was more saying that a slower feed rate, either slightly, or profoundly as in the case of 4mm/min feed rate, with the system working normally, will make no difference.

Regards,

Bill
 
The 30 mm federate is used for not rotating tools. The 4 mm federate is used for rotating tools, using 800 RPM, giving 0.005 mm/rev. If the end mill is 2-flute, then in worth case in which the first contact was just missed, the next will appear after 2.5 micron of Z movement. This is the inherent scatter of the measuring process of rotary tools.
 
Yes I do use wear offsets.
I bought the probing kit to stop having to manually set tool length. But if there is so much variation it's useless.
 
Did some test today on probing endmill length (non rotating), variation of 0,03mm in 10 probing cycles.
 








 
Back
Top