Probing 4 points

[APT2000]

I thought I would share a common question we have been receiving about probing 4 points on a surface and finding the average. Most have been adding the points and dividing by 4 to get the average. The problem with this method is that it can give you undesirable results even if you don’t have negative values. A better way to do it is to find the mid point, that is the point that is half way between the lowest reading and the highest reading. You could code that like this on a Fanuc with Renishaw macros:

G65P9811Zz
#101=#142(FIRST PT Z ERROR)
G65P9811Zz
#102=#142(SECOND PT Z ERROR)
G65P9811Zz
#103=#142(THIRD PT Z ERROR)
G65P9811Zz
#104=#142(FOURTH PT Z ERROR)

(SET VARIABLE TO LOOP)
#10=1
(LOOP CODE TO FIND HIGH AND LOW)
WHILE[#10LE4]DO1
IF[#10GT1]GOTO10
(GET FIRST POINT AND SET)
(IT TO HIGH AND LOW )
#12=#101(HIGH)
#14=#101(LOW)
GOTO30
N10
IF[#[100+#10]GT#12]GOTO20
IF[#[100+#10]LT#14]GOTO25
GOTO30
N20
#12=#[100+#10](SET NEW HIGH POINT)
GOTO30
N25
#14=#[100+#10](SET NEW LOW POINT)
N30
#10=#10+1.(INCREMENT COUNTER)
END1
(NOW FIND THE MIDPOINT AND SET THE ORIGIN)
#16=#12+#14/2.(MIDPOINT)
(UPDATE THE G54 Z ORIGIN)
#5223=#5223+#16

Note that the system variable for G54 Z was used, 5223
So, the expression wrote the new Z value to the existing Z value plus the mid point error. If you’re more comfortable with the parameter write statement you can do this:

G91
G10L2P1Z#16
G90

Because it’s incremental when in G91 it just adds the value in #16 to the existing Z value. Just don’t forget to put it back in G90 absolute mode.

Also, keep in mind that this method just updates the G54 Z by the error found measuring a target surface. It doesn’t change the G54 Z to some other feature. If you want to do that, write the same code but use the actual Z value from the measure output, #137 instead of #142

I hope this helps and find it useful.

APT2000

Advanced Probe Technologies
www.Advprobetech.com