See other templatesSee other templates

While I'm training people on CMM, one thing that I've noticed is that from time to time, people seem to mix/mis-use between the term of "alignment" and "co-ordinate system". I think it's very important for anyone who uses a co-ordinate based measuring device (e.g. cmm, portable arm, scanner, etc.) to understand the difference between "alignment" and "co-ordinate system" to create dimensional results. If you're quite familiar with GD&T, then a more "correct" term will be DRF (Datum Reference Frame), because once a DRF is established it covers both alignment and co-ordinate system. Without too much digging into theoretical part, I would simply say alignment should be established with physical objects, but co-ordinate system is established with virtual elements. And then let's study a practical case so you can clearly understand the difference between them.


Have a look at above drawing and ask yourself 2 simple questions:

1. How many alignments do we need to measure the part?

2. How many co-ordinate systems do we need to report all dimensions?


If you can answer them confidently then you pretty much have a good idea about the difference between these 2 terms. To measure this part, we really need only 1 alignment because we only have one set of datum reference frame, established from -ABC-. And this alignment should be established by measuring all datum features and processing them accordingly (primary/secondary/tertiary). For co-ordinate system, we have at least 2: one is a cartesian coordinate system with origin at the lower left corner and the other is a polar coordinate system with origin at the centre of a bolt hole pattern. But this is important: both still share a same alignment system established from -ABC-! This means:


1. Once your part alignment is established from -ABC-, you should NOT construct another alignment to measure or report any of the dimensions in this drawing;

2. To report true position for those bolt holes, you only need to establish a new co-ordinate system by translating your origin with nominal (basic) dimensions;

3. If you set origin at the centre of a circle constructed from the six measured holes, then you're changing your alignment, which is wrong;

4. If you want to report actual position in each axis for those small holes, to match the drawing basic dimensions, then you only need to translate your origin accordingly with basic dimension. Again it'll be wrong if you set your origin to each measured hole to establish zero for next one.


So you can see the co-ordinate system is used to create basic dimensions (linear, angular, etc.) in a more meaningful way to describe the relationship between features theoretically. But alignment reflects the true physical functionality. Therefore it's possible:


1. The entire drawing has only one alignment, but there're many various co-ordinate systems to dimension features at different angles, etc.

2. The entire drawing has only one co-ordinate system, but multiple alignments based on various datum reference frames (e.g. -ABC-, -DEF-, -ADF-, etc.) because of functionality;

3. There're multiple alignments and multiple co-ordinate systems on a same drawing;

4. No co-ordinate system at all. This is possible when digital cad model is used, everything is from cad, no basic dimension needed to define features. But you still need an alignment to align the part, based on functionality, as minimum.


When a cmm programmer create a cmm program in any software, he can freely choose to establish or change an alignment or a co-ordinate system, thus generates different dimensional results. You won't be able to tell which one he does unless you check the program. So it's extremely important that the cmm program is created correctly according to drawing requirement but not any assumption. I have supported many customers by providing offline programs in various cmm software. One thing I've done is to make sure my programs comply with drawing GD&T to the best of their cmm software capability. If you need help or training to create accurate and reliable cmm programs for your cmm system, don't hesitate to contact me at:  This email address is being protected from spambots. You need JavaScript enabled to view it.


Currently I can offline program on following software:

Open Dmis, Camio, Modus, Mcosmos, CMM Manager, PC-Dmis




Ray Xing                                                                 

GDTP S-0605

CGP Certified (ITAR equivalent)

CMM Application Specialist

RX Metrology Solutions Inc.              

One stop solution for CMM and GDT training


JSN Megazine template designed by