Laser Trackers - How do we conduct a Gage R&R?

[lee01]

Regarding Gauge R&R and Laser Trackers.

I have to conduct a Gauge R&R on a Laser Tracker, my problem is: Conducting Gauge R&R with other devices is easy but with a laser tracker is difficult.

The Laser Tracker uses spherical reflectors to bounce the beam back to the Tracker Head. This causes problems because I cannot guarantee that the reflector is the same position throughout the 3 measurement trials (9 ‘hits’).

Imagine measuring the side of your house, the getting someone else to do it, it would be impossible for them to hit the same spot every time. With a CMM the programme would ensure it would hit the same position for each ‘hit’

How do we conduct a Gauge R&R for Laser Trackers, or should it be conducted using a field test?

 

[Al Rosen]

Regarding Gauge R&R and Laser Trackers.

I have to conduct a Gauge R&R on a Laser Tracker, my problem is: Conducting Gauge R&R with other devices is easy but with a laser tracker is difficult.

The Laser Tracker uses spherical reflectors to bounce the beam back to the Tracker Head. This causes problems because I cannot guarantee that the reflector is the same position throughout the 3 measurement trials (9 ‘hits’).

Imagine measuring the side of your house, the getting someone else to do it, it would be impossible for them to hit the same spot every time. With a CMM the programme would ensure it would hit the same position for each ‘hit’

How do we conduct a Gauge R&R for Laser Trackers, or should it be conducted using a field test?

Isn't that the point of doing a GR&R?

 

[Dave W 2005]

Hi Lee.

Long time no speak!

Have a similar issue here. What about creating a plane rather than a point to measure. That way it may not be as important to be precise (or is that accurate?)

What features will you measure in production? Are they holes / surfaces / pins etc. What about using these characteristics as the subject of the GRR (or use the GD&T to define requirements)?
The system should be able to measure them repeatably and with good reproducibility.

 

[Charmed]

What is laser tracker doing?

Regarding Gauge R&R and Laser Trackers.

I have to conduct a Gauge R&R on a Laser Tracker, my problem is: Conducting Gauge R&R with other devices is easy but with a laser tracker is difficult.

With a CMM the programme would ensure it would hit the same position for each ‘hit’

How do we conduct a Gauge R&R for Laser Trackers, or should it be conducted using a field test?

Dear lee01:

How about doing exactly what you would do with a CMM?

To me, the point is not making measurements of the side of the house, but to show that the laser tracker is performing reproducibly. If the side of my house measures 10 meters, can I guarantee that all measurements are being made from the same point A , not just to the same point B , say on the wall of the house? The measurement of of the length AB will no doubt be very accurate but we must be able to guarantee the position of the starting A as well as the end point B.

Secondly, the whole point of the laser tracker, IMHO, is "parallelism". The laser shoots a ray of light, which travels in a perfect straight line. Hence, the the line AB revealed by the laser is perfectly parallel to a second reference line CD, which may be the floor, the ceiling etc. Perhaps, this is an even more important aspect of what you are trying to do with your laser tracker. Hope this helps.

 

[lee01]

Hi Dave, how are things?

Well we’re only using the Laser Trackers for measuring surfaces (IML/ OML/ EOP) as I believe the build up of measurement error is far too great when using the adaptors (both fixed and expanding mandrel type) to measure hole position.

My problem is when measuring a surface (especially those that are not flat) it becomes critical that the reflector is positioned in the very same place every time a measurement is taken during the GRR. Otherwise it will become difficult to establish if the measurement uncertainty is due to the device/ appraiser or the due to positioning of the reflector?

Put this way, if you have a curved surface that over any inch has a deviation of ½” in ‘Y’, the positioning of the reflector becomes critical as any error could be based on simply the error in positioning the reflector and not the ability of the device?

I will have a word with our GD&T expert, although I have done the courses, its something I tend (like) to forget.

Have you done anything with your digital data capture issue yet Dave?

 

[lee01]

Doing a GRR on a CMM is different as the CMM uses the program to ensure the probe is hitting the same point every time during the GRR. The Laser Tracker uses a hand held reflector, the error I’m talking about is this.

When inspecting a surface the ‘inspection program’ calculates the deviation from the nominal surface, the position of the reflector doesn’t matter as long as the reflector is on the surface it will calculate ‘normal’ to the nominal surface. When conducting a GRR the positioning of the reflector is important.

The only way I can think of is to mark a cross on the surface and ensure each user places the reflector on that cross. (How accurate is that though?).

As for your statements on parallelism etc . . . not too sure I understand your thoughts?

 

[Charmed]

Purpose of laser tracker

As for your statements on parallelism etc . . . not too sure I understand your thoughts?

From what I understand (based on commercials that I have seen not from actual use of a laser tracker), this new device helps the user maintain a perfect straight line. Example, when one is installing cabinets, or photographs, on a wall. The straight line defined by the laser light, which defines the position of the cabinets, photographs, etc., must, however, be parallel to the straight line defined by the ceiling and/or the floor of the house. Hope this helps.

 

[lee01]

Okay I think you have misunderstood what a laser tracker is.

A laser tracker is a highly accurate piece of kit that costs, in some instances over £250000! There are three main suppliers of these devices:

API Laser Tracker - www.apisensor.com
Leica Laser Tracker - www.leica-geosystems.com/corporate/en/products/laser_tracker/lgs_814.htm
FARO Laser Tracker - (broken link removed)

They look like little darlicks from Dr Who and omit a laser beam that is reflected back to the laser tracker head by a ‘Reflector’, then software calculates the distance of the beam, adds the diameter of the reflector and that’s your position. The reflector is moved along a surface and the laser beam follows the reflector thus gaining real-time data.

Have a look at the internet sites above.

Out of interest the FARO tracker has moved on and seems to be more advanced then the previously class leader – Leica. I’ve heard very bad things about the API’s, I think they are overheating or something.

The UK distributor cannot get his hands on a demo because they are being delivered as replacements! Not sure if this is true but?

 

[Charmed]

Thanks

Okay I think you have misunderstood what a laser tracker is.

A laser tracker is a highly accurate piece of kit that costs, in some instances over £250000! There are three main suppliers of these devices:

API Laser Tracker
Leica Laser Tracker
FARO Laser Tracker

They look like little darlicks from Dr Who and omit a laser beam that is reflected back to the laser tracker head by a ‘Reflector’, then software calculates the distance of the beam, adds the diameter of the reflector and that’s your position. The reflector is moved along a surface and the laser beam follows the reflector thus gaining real-time data.

Have a look at the internet sites above.

Out of interest the FARO tracker has moved on and seems to be more advanced then the previously class leader – Leica. I’ve heard very bad things about the API’s, I think they are overheating or something.

The UK distributor cannot get his hands on a demo because they are being delivered as replacements! Not sure if this is true but?

Thanks for the clarification. I was wondering why you were concerned about this "cheap" commercially available device!

 

[Dave W 2005]

Hi Lee,

We are thinking about purchasing a laser radar system, so I'm really interested in any solution you come up with.

One thought, can you not simply relate your surfaces back to the CAD model? The process might go something like this:
Operator 1: take 100 points across surface. Calculate mean and std dev for deviation from modelled surface
Operator 2 does the same
Operator 3 etc

Then carry out f and t tests on the data to decide if the results are statistically different.
You could use the deviation data to determine how much of the tolerance is taken up by variation. You may have to do some number crunching to separate out the actual product variation from the measured values because there will be no way of measuring in the same place. You may have to take an averaged surface????

Ref data capture, we have developed a system in France and our trialling it here next month!