How to select

Coordinate

Measuring

Machine

A short buyer’s guide

1. Introduction
Process control and quality assurance in modern manufacturing depends more and
more on Coordinate Measuring Machines (CMM’s). Over the last 20 years CMM’s
have widely replaced the traditional way of inspection with gages and fixtures.
Because of their flexibility they reduce the investment in hard gauges. CMM’s also
drastically reduce the time and manpower required in traditional quality control.

In addition, precision CMM’s not only inspect the standard geometrical dimensions but
also provide special inspection features , such as gear measurement, camshaft
measurement, roundness testing etc. Each of these inspections would require a single
purpose testing machine in a traditional factory.

High product quality does not only depend on the quality of the machine tools used for
manufacturing. High product quality mainly depends on the accuracy and repeatability
of the instruments used for quality control.

A low cost, low performance machining center in combination with a high precision
CMM can still guarantee a top quality product, because only parts within the tolerance
can pass the CMM’s inspection.

On the other hand an expensive, high quality machining center in combination with a
low cost, low accuracy measuring device cannot guarantee a high quality product.
A certain percentage of parts “out of tolerance” will always pass the low accuracy CMM
inspection. Likewise, a certain percentage of “in tolerance” parts will be rejected.

The selection of a suitable CMM is a critical decision in regard to a companie’s quality

control standards. Even more so as CMM’s have a very long life expectancy,
compared to machine tools.

This brief buyer’s guide is kept intentionally short and therefore does not compare the
advantages / disadvantages of different CMM design concepts or different probe head
principles etc.

Its purpose is to enable potential CMM buyers to choose a suitable CMM, which meets
their accuracy and application requirements, by focusing only on the most important
selection criteria.

2
2. The first step: The measuring range

The first important selection criteria is the “Minimum Required Measuring Range”.
It usually depends on the dimensions of the largest part to be measured.

However, if the inspection requires the use of extensions and fixtures, the actual
required measuring range can be considerable larger than the workpiece dimensions.

Example : Machine tool housing

250 650 300

Part length : 650mm

Required clearance (left) : 250mm
Required clearance (right) : 300mm

 Minimum required measuring range : 1200mm

This analysis has to be done for each axis, so that the minimum required measuring
range for the CMM can be determined.

CMM makers can assist in this analysis, if drawings are provided.

The result of your measuring range analysis:

Min. measuring range in X : mm

Min. measuring range in Y : mm
Min. measuring range in Z : mm

3. The Measuring Uncertainty

3
The second selection criteria is the “Minimum Required Uncertainty”.
It is dependent on the drawing tolerances.

The uncertainties and test procedures for CMM’s are described in ISO 10360-2.

Some CMM makers still do not conform to ISO 10 360-2, but use older performance
standards (CMMA, VDI/VDE 2617, B 89, JIS).
In order to compare CMM’s from different manufacturers, the customer has to request
the CMM specifications based on the ISO 10 360-2 standard, which is in force since
1994.

3.1 The ISO 10 360-2 standard

ISO 10 360-2 specifies two uncertainties :

A. The volumetric length measuring error E

 A set of gages has to be measured

several times with one probing at each
end.

 The gages can be aligned in any direction.

 All results (100%) must be in tolerance.

Note: All specifications according ISO 10 360-2 are „volumetric“. The uncertainty along
an axis, which was specified in older standards (U1) is obsolete, as workpieces
on CMM’s are not mechanically aligned, therefore the measurements are
always volumetric.

4
B. The volumetric probing error R

 High precision sphere is

measured with 25 equally
distributed probing points

 The resulting FORM

deviation is R.

 All 25 probings (100%) must

be included in the
calculation of the FORM
deviation.

R is the form error of the CMM.

If a precision ring gauge is measured, the form error shown on the plot is the form error
of the CMM ( =R ).

5
Insist on these two specifications (E + R) according to ISO 10 360-2 from any CMM
maker, in order to be able to compare them. Do not accept any outdated national
standards.

There are further extensions of the ISO 10360 in preparation, such as

ISO 10360-3: Uncertainty of CMM’s with rotary table as a fourth axis

ISO 10360-4: CMM’s used in scanning mode
ISO 10360-5: CMM’s used with multiple styli

They are however in a draft status and therefore will not be considered here.

3.2 The ratio „Uncertainty to Tolerance“

Depending on your drawing tolerances, the required ISO specifications ‘E’ and ‘R’ can
be determined. However the measurements for the verification of a CMM’s ‘E’ and ‘R‘
specifications are quite different from real workpiece measurements.

You have to keep in mind that a CMM’s uncertainty under real working conditions can
be considerably larger than stated on the CMM’s datasheet because of the use of
extensions, long or thin probe pins, rotary tables, revolving probe heads, temperature
changes, dirt, dust etc.

E and R as specified: E and R (working condition):

- 1 probe pin, - combination of several probe pins
- fixed directly in probe head - use of extensions
- no extensions - rotation of probe head (Renishaw)
- no rotation of probe head - probe change

If a part can be inspected with this Most workpieces require complex probe
probe pin, then a „uncertainty to configurations, for which E and R are not

6
tolerance“ ratio of 1 : 5 can be applied. specified. Then a „uncertainty to tolerance“
ratio of 1 : 10 should be applied.

In order to maintain a 1:5 ratio (uncertainty of CMM : part tolerance) under any
working conditions, the CMM datasheet specs should be 10 times better than the
tolerances to be inspected.

4. Analysis of the required CMM uncertainty

On almost all workpieces, CMM’s have to inspect three groups of features:

A. Diameters and Distances

B. Position Tolerances
C. Form Tolerances

For each group an analysis about the required CMM uncertainty has to be made.

A. Diameters and Distances

Study your part drawings and search for the distances with the tightest tolerances.
Note, that a larger tolerance on a very long dimension might be a tougher requirement
than a smaller tolerance on a very short dimension.
Find the toughest requirement and calculate the required length measurement
uncertainty ‘E’ by applying the 10% rule.
See the example on page 9 and then fill in the form on page 10.

B. Position Tolerances

Study your part drawings and find the tightest position tolerances. Please note that the
distance to the datum is also important. Therefore, a larger position tolerance with a
long distance to one of the datums may represent the tougher requirement.

Position tolerances usually define a tolerance diameter. Therefore you can only
consider the radius as the allowable deviation from the nominal center.

With the help of the example on page 11 you can analyze the volumetric length
measuring uncertainty E, which would be required to inspect your position tolerances.

C. Form tolerances

7
Study your part drawings and find the smallest form tolerances, like roundness or
profile tolerances on a screw-compressor or scroll. 10% of this form tolerance gives you
the required volumetric probing uncertainty R.
See the example on page 13 and then fill in the form on page 14.

4.1 Uncertainty charts

Calculated for a theoretical ratio „Uncertainty : Tolerance“ = 1 : 10

 As shown on the previous page, the actual ratio achieved in the lab or workshop
is usually considerably lower because data sheet conditions are not met !

A. Diameters and distances

Distance or diameter [mm]
Tolerance 10 100 300 600 1000
0.003mm 0.3+L/1000
0.005mm 0.5+L/900 0.4+L/1000
0.007mm 0.7+L/700 0.6+L/900 0.4+L/1000
0.010mm 1.0+L/400 0.8+L/500 0.6+L/750 0.4+L/1000
0.015mm 1.5+L/300 1.2+L/300 0.8+L/450 0.5+L/600 0.5+L/1000
0.020mm 2.0+L/250 1.6+L/250 1.3+L/450 0.8+L/500 0.8+L/800
0.030mm 3.0+L/250 2.6+L/250 2.0+L/300 1.5+L/400 1.0+L/500
0.050mm 5.0+L/150 4.3+L/150 3.5+L/200 2.6+L/250 1.7+L/300
0.070mm 7.0+L/100 6.0+L/100 5.0+L/150 4.0+L/200 2.0+L/200
0.100mm 10+L/100 9.0+L/100 7.0+L/100 6.0+L/150 4.0+L/150
Example: A 300mm diameter has a tolerance of 0.020mm.
For the inspection of this feature a CMM with a volumetric length uncertainty of
E = 1.3 + L / 450 [m] would be required.

B. Position tolerances
Position Distance to the datum [mm]
Tolerance 10 100 300 600 1000
0.005mm 0.25+L/1000
0.010mm 0.5+L/900 0.4+L/1000
0.015mm 0.7+L/500 0.6+L/600 0.4+L/1000
0.020mm 1.0+L/400 0.8+L/500 0.6+L/750 0.4+L/1000
0.030mm 1.5+L/250 1.2+L/350 0.8+L/450 0.6+L/700 0.5+L/1000
0.050mm 2.5+L/200 2.2+L/300 1.6+L/350 1.2+L/450 0.8+L/600
0.070mm 3.5+L/200 3.0+L/200 2.5+L/300 1.5+L/300 1.0+L/400
0.100mm 5.0+L/150 4.3+L/150 3.5+L/200 2.6+L/250 1.7+L/300
0.200mm 10+L/100 9.0+L/100 7.0+L/100 6.0+L/150 4.0+L/150
Example: A hole has a position tolerance of 0.05mm. The distance to the datum is 100mm.
For the inspection of this feature a CMM with a volumetric length uncertainty of
E = 2.2 + L / 300 [m] would be required.

8
C. Form tolerances

Form tolerance [m]

5 7 10 15 20 30 50
R [m] 0.5 0.7 1.0 1.5 2.0 3.0 5.0
Example: A 20mm diameter has a roundness tolerance of 0.007mm.
For the inspection of this feature a CMM with a volumetric probing uncertainty of
R = 0.7m is required.
Please note that R is specified for relatively small areas (up to 30mm).

4.2 Analysis of Measuring requirements

for: Example Factory Ltd.

A. Diameters and Distances

300 ± 0.010 500 ± 0.020

Tightest distance tolerance: ± 10 m

Required measuring uncertainty: 0.1 x ± 10 m = 1 m

Distance or diameter: 300 mm

 E (required) = ± 0.6 + L / 750 [m]

(see 4.1, chart A)

Suitable Measuring Machines:

9
 Manufacturer Vol. Length Uncertainty Uncertainty Tolerance Ratio
E over 300 mm
CMM Type
B & S PMM 866 MG ± 0.6 + L / 600 [m] ± 1.1 m ± 10 m 1:9

B & S PMM 866 ± 0.8 + L / 400 [m] ± 1.6 m ± 10 m 1 : 6.5

.

± +L/ [m] ± m ± m :

4.2 Analysis of Measuring requirements

for: Example Factory Ltd.

B. Position Tolerances

0.03 A 0.02 A B
200

30 0

A 150

0.02 = +- 0.01
Attention: A Position tolerance of 20m (circular) is
equivalent to a tolerance of ± 10m for the
measurement !

Tightest Position tolerance: 20 m

Required measuring uncertainty: 0.1 x ± 0.5 x 20 m = ± 1 m

Longest distance to datum: 200 mm

10
  E (required) = ± 0.7 + L / 650 [m]
(see 4.1, chart B)

Suitable Measuring Machines:

Manufacturer Vol. Length Uncertainty Uncertainty Tolerance Ratio

E over 200 mm
CMM Type
B & S PMM 866 MG ± 0.6 + L / 600 [m] ± 0.9 m ± 10 m 1 : 11

B & S PMM 866 ± 0.8 + L / 400 [m] ± 1.3 m ± 10 m 1:8

± +L/ [m] ± m ± m :
4.2 Analysis of Measuring requirements

for: Example Factory Ltd.

C. Form Tolerances

Roundness Flatness Cylinder form Straightness Profile form

0.01 0.007

0.02

0.005 0.01
Drawing tolerances:

11
 Tightest Form tolerance: 5 m

Required CMM probing uncertainty: 0.1 x 5 m = 0.5 m

 R (required) = 0.5 m

Suitable Measuring Machines:

Manufacturer Probing Uncertainty Form Ratio

R Tolerance
CMM Type
B & S PMM 866 MG 0.6 m 5 m 1:8

B & S PMM 866 0.8 m 5 m 1:6

m m :

4.3 Calculation of Ratio “Measuring Uncertainty : Part Tolerance”

Based on the part drawing and the CMM type the ratio “Measuring uncertainty : Tolerance”
must be calculated for each feature. Only then can be seen if a CMM is capable of
inspecting a workpiece.

Measuring machine: Brown & Sharpe PMM-C 700

Machine specifications according ISO 10 360-2:
Volumetric length measuring error E = 1.0 + L / 400 [µm]
Volumetric probing error R = 1.0 µm

1. Evaluation of distances and diameters

Part Drawing-No.: Sheet Loca Feature Nominal Tolerance CMM Ratio Remarks
tion Uncertainty
mm mm
Cylinderhead JR8-7-1999 1 d8 Diameter 100 +- 0,010 +- 0,0013 1 : 8,0
JR8-7-1999 3 k9 Distance 600 +- 0,007 +- 0,0025 1 : 2,8
JR8-7-1999 5 b10 Distance 300 +- 0,020 +- 0,0018 1 : 11,4
+- 0,0010 1 : 0,0
+- 0,0010 1 : 0,0
+- 0,0010 1 : 0,0
+- 0,0010 1 : 0,0
+- 0,0010 1 : 0,0

2. Evaluation of true position

Part Drawing-No.: Sheet Locati Feature Distance Tolerance equivalent to CMM Ratio
on to datum Uncertainty
mm mm +- Tolerance
Cyl-head JR8-7-1999 1 d8 Position 100 0,007 +- 0,0035 +- 0,0013 1 : 2,8
JR8-7-1999 3 k9 Position 600 0,030 +- 0,015 +- 0,0025 1 : 6,0
JR8-7-1999 5 b10 Position 300 0,035 +- 0,0175 +- 0,0018 1 : 10,0
Position +- 0,000 +- 0,0010 1 : 0,0

12
 Position +- 0,000 +- 0,0010 1 : 0,0
Position +- 0,000 +- 0,0010 1 : 0,0
Position +- 0,000 +- 0,0010 1 : 0,0
Position +- 0,000 +- 0,0010 1 : 0,0

3. Evaluation of form tolerances

Part Drawing-No.: Sheet Location Feature Tolerance CMM Ratio Remarks

mm Uncertainty
Cylinderhead JR8-7-1999 1 d8 Flatness 0,010 0,001 1 : 10,0
JR8-7-1999 3 k9 Roundness 0,003 0,001 1 : 3,0
JR8-7-1999 5 b10 Cylinderform 0,005 0,001 1 : 5,0
0,001 1 : 0,0
0,001 1 : 0,0
0,001 1 : 0,0
0,001 1 : 0,0

Above charts are available as MS Excel sheets. By entering various CMM’s and their E and
R specs, the ratios of each CMM is shown immediately.
5. Other Requirements

5.1. Environment

The uncertainty of every CMM depends very much on the environmental conditions.

Therefore the manufacturers usually specify:

the Temperature range

the Temperature variation per hour (C / h)
the Temperature variation per day (C / d)
and the Temperature variation per meter (C / m)

within which their CMM will achieve the specifications.

You have to check if your measuring room can meet the CMM’s specified
requirements.

5.2. Special Applications / Software

All CMM makers provide software packages for basic measurements.

Some makers also provide software for parts with more complex geometries, such as
Bevel gears, Impellers, Screw compressors, Hob cutters, etc.

You have to check, if your measuring problem can be solved by the CMM’s offered.
The best way to check it is to ask for a demonstration.

5.3. Throughput

13
The more parts a CMM can measure per day, the lower are the inspection costs per
part.
Moving speed, acceleration and number of probing points/min are the factors which
determine the overall throughput.
The throughput can also be increased by features like the capability of pallet
measurement of parts.

5.4 Service

Check if the CMM maker/dealer has factory trained service and application engineers
locally available. Yearly maintenance contracts should also be available.

6. Evaluation Sheet (Example)

Requirements CMM A CMM B CMM C

1. Measuring range
X =
Y =
Z =
m³ ( X x Y x Z )

2. Volumetric uncertainty
E:
R:

Resolution:

3. Environment
Temp. range :
Variation: K/h, K/d, K/m

4. Throughput
Travel speed
Acceleration
Probing points / min
Pallet measurement

5. Special Applications
(i.e. Gear)

14
6. Options
6.1 Autom. Probe changer
6.2 Optical probe head

7. Computer

8. Software options

9. Price including

Installation + Training

15