Understanding Probe correlation in PC-DMIS

In all but the most basic programs it is likely that you will use more than one probe-tip in your PC-DMIS measurement routine.

​

Probe correlation is about ensuring that measurements taken with one probe-tip relate to measurements taken with any other probe-tip.

​

You will notice the use of 'probe-tip' rather that 'probe' or 'tip-angle' as this covers all scenarios; here we could be referring to using either more than one tip angle from the same probe, a different tip from a star probe, an entirely different probe build, or indeed another sensor entirely, be that a Laser scanner or Vision sensor.

​

Whether you have a single straight probe with multiple angles, or a Multi-Sensor System (MSS) we can agree it is imperative that these probe-tips all relate correctly to each another.

​

One of the major issues surrounding this topic is that every company is different and has different requirements - some will have no tool rack with only a couple of modules, which need to be reconfigured depending on the job at hand, whilst some will have probe racks with a range of probes which can be used on all parts. Some companies will set the CMM up for a job and it will run just that part for months, others will be switching programs with every part that hits the CMM. Some people will be working in the confines of a small CMM where the reference sphere needs removing in order to measure that part, and some will have huge machines where the reference sphere is never removed. Some inspection departments will have one guy running the CMM, and other will have multiple people across multiple shifts using the CMM and potentially calibrating probes. All these factors can have in impact on exactly how you decide to manage your probe calibration, and to that end, understanding how this works within PC-DMIS will allow you to make the right decision.

​

There are three commonly used methods to ensure this happens which are discussed below:

• Calibrate on a per-job basis

• Periodic calibration of all probes

• Use a 'Master' probe

Calibrate on a per-job basis

​

Calibrating all probe-tip combinations used in a program before running the program is often seen as the safe option and indeed it can be. It is however time consuming if you're regularly switching jobs.

Pros

• Safe - especially if managed through Auto Calibrate commands (which can also be managed via flow control to either give the option to calibrate or not, or even auto-detect if it needs to be run).

• Necessary when you have to constantly reconfigure probes to suit the job (i.e. if you only have a couple of modules).

Cons

• Time consuming and inefficient.

• Prone to human error if not using auto calibrate.

• Especially inefficient on Multi-sensor systems.

When to use

Typically this system can be employed on basic systems where probes are routinely taken apart and reconfigured, often when no probe rack is used. This can also be a good option when the CMM runs just one job for long periods.

Periodic calibration

Periodic calibration refers to routinely calibrating all probe-tips, this can be on a weekly, daily or even per-shift basis.

Pros

• Safe - especially in managed through an Auto Calibrate program (which can also be managed via flow control to force the calibration at the specified interval).

Cons

• Can be very time consuming and inefficient.

• Prone to human error if not using auto calibrate.

• Especially inefficient on Multi-sensor systems.

• In the event of a broken stylus - the full calibration may need to be run again.

When to use

We might use this system in companies where different jobs are being run frequently, and where a probe rack of multiple probes are used across the range of parts - but only if there are a fairly small number of probe-tip combinations. Often this might be in an environment where many people (sometimes less-skilled) use the CMM on a regular basis.

Master probe

The Master probe method relates to having one probe-tip (which we refer to as the Master-probe) which is used to guarantee that all other probe-tips and sensors correlate to each other.

Pros

• Flexible - allows maximum flexibility of managing all probe calibrations.

• Efficient - used properly it can massively reduce the time spent recalibrating probes. We would often use a probe check program to give confidence that the calibration is still good.

• Multi-sensor systems - as both Vision and Laser probes are non-contact, they usually don't require periodic recalibration in the way that touch-trigger and scanning probes do.

Cons

• Requires a sound understanding of PC-DMIS probe calibration.

• Requires that either systems are in place and adhered to, or that all personnel are well versed in the usage.

When to use

We might use this system in companies where different jobs are being run frequently, and where a probe rack of multiple probes are used across the range of parts. This option is better if there are a lot of probe-tip combinations. This is also the most sensible option for Multi-sensor systems.

Probe-tip relationships in PC-DMIS

No matter which method we use, it is useful to understand the mechanism within PC-DMIS which governs if different probe-tips will correlate. For simplicity reasons we are only going to focus on contact probes (touch trigger or scanning) initially.

​

Probe offsets

​

Every probe-tip within PC-DMIS has both theoretical (THEO) and measured (MEAS) probe offsets - these are the offsets (in XYZ) from the centre of the probe-tip, to the end of the CMM arm (typically for a bridge CMM the end of the Z axis where the probe head is mounted).

​

When we define a new probe in PC-DMIS, all that is known is the nominal sizes of the components used in the probe build (i.e. the probe head, the sensor, any extensions and the stylus length). In effect, PC-DMIS 'knows' where the theoretical tip centre is.

​

We will come back to this shortly!

​

Qualification Tool Moved

​

When we go to calibrate a probe we are asked one of the most important questions:

Has the qualification tool been moved, or has the Machine zero point changed?

​

Lets deal with the second part of this question first: the Machine zero point refers to the home position of the CMM. You might think that your CMM homes to the same point each time, but you are most likely mistaken. It can vary, but depending on many factors the CMM will home to either a slightly , or significantly different location each time.

​

The first part of this question seems more obvious - has the Qualification tool been moved? If the Qualification tool has been removed and put back in a different location entirely, then we should of course assume it to have been moved. However, if the CMM has rehomed then we should also assume the Qualification tool to have been moved (whether it physically has or not). If the Qualification tool has been removed, but replaced in the same location we should still assume the tool to have been moved, as this location won't be repeatable enough to guarantee good probe correlation.

​

Why is this important?

​

What your answer to this question actually does, is decide whether we are defining a new Qualification tool location, or calculating tip offsets relative to the old Qualification tool location.

Perhaps better wording to this question would be:

Do you want to define a new Qualification tool position with the selected tip?

​

In order to better understand this, lets look at exactly what happens here.

​

Example

​

If we take the example of a brand new CMM with no probes defined, we can define out first probe with one tip T1A0B0 (and define our reference sphere) and go to calibrate. When asked the 'Qualification Tool Moved' question we must say 'Yes (Manual hit to locate tool)'