Good evening Pcdmis masters, I`m fairly new to Pcdmis, and i would love to get some help , I am trying to measure a Countersink Dia at the intersection of the top surface on this part, the available area is very limited and the thread leads are not clocked to the model to get an auto cone reading accurately, is there any method i can use to first identify the rotation of the the double lead thread and then construct a cone avoiding the thread lead`s in order to get an accurate reading? any tips will be appreciated ,thanks in advance
Really small , the Countersink dia is 3.3 millimeters and the thread hole is 3 millimeter, only about 3 threads, this might not even be doable on the CMM , I tried freeform scanning the small section staying away from the leads in and then constructing a cone and it worked great with a .5 stylus but since the threads are not clocked to the model I will have to move the scans as needed, i was hoping there was a way to locate where the leads are and then build the scans for the cone once I know the orientation , thanks
Using auto-center scan, with 2 different probes (Ø4 mm and Ø5 mm), level and origin Z to the top plane (PT1 is the autocenter point with Ø4 and PT2 with Ø5)
ASSIGN/V1=PROBEDATA("diam","tip4")/2+PROBEDATA("prbrdv","tip4")
ASSIGN/V2=PROBEDATA("diam","tip5")/2+PROBEDATA("prbrdv","tip5")
ASSIGN/V3=RAD2DEG(ATAN(PT2.Z-PT1.Z)/(V2-V1))) is the angle of the chamfer
Its length depends on the radius of the threaded hole.
You can get the diameter, but, the angle etc will probably be obliterated by the threads.
Measure a plane around the feature, then construct an offset plane (into hole), say 0.01mm below. Now, set an alignment. Take a set of points (in line) crossing where you expect the edge to be, construct a curve (you'll have to play around with thinning value), construct pierce point with your offset plane and that curve, now pattern that last bit as needed to make a bunch of pierced points. This works great for spotface diameters and positions as well. Example has only one edge point for simplicity. This actually works, I do this all the time.
I get these often too & if they are on the smallish side I skip wasting time on the CMM & use optical comparator. If the part is too large setting 2 calipers on min & max size works well. For the angle I bought an angle gauge set long time ago & it does the job. There are some things that other tools will do faster & better job than the CMM still.
i have another couple of questions, this is a sample of my code
how can i get the diameter?
and how can i dimension this?, i`m really new to this, and only have Pcdmis 101 training `im not understanding too much,of what i`m doing here,sorry if i sound dumb
thanks in advance
PT1 =FEAT/CONTACT/VECTOR POINT/SELF_CENTERING_POINT,CARTESIAN
THEO/<34.5,0,0.18595>,<0,0,1>
ACTL/<34.57659,-0.01821,2.06588>,<0,0,1>
TARG/<34.5,0,0.18595>,<0,0,1>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE BEFORE=3,DISTANCE AFTER=3
SHOW HITS=NO
MOVE/INCREMENT,<0,0,25>
LOADPROBE/T1_3X20
TIP/T1A0B0, SHANKIJK=-0.00138, -0.00057, 1, ANGLE=-0.19084
PT2 =FEAT/CONTACT/VECTOR POINT/SELF_CENTERING_POINT,CARTESIAN
THEO/<34.5,0,0.38776>,<0,0,1>
ACTL/<34.49686,-0.01152,0.2738>,<0,0,1>
TARG/<34.5,0,0.38776>,<0,0,1>
SNAP=NO
SHOW FEATURE PARAMETERS=NO
SHOW CONTACT PARAMETERS=YES
AVOIDANCE MOVE=BOTH,DISTANCE BEFORE=10,DISTANCE AFTER=10
SHOW HITS=NO
ALIGN_2 =ALIGNMENT/START,RECALL:ALIGN_XY_ORIGIN,LIST=YES
ALIGNMENT/LEVEL,ZPLUS,PT1
ALIGNMENT/TRANS,ZAXIS,PT1
ALIGNMENT/END
ASSIGN/V2=PROBEDATA("3.0","T6")/2+PROBEDATA("prbrvd","T6")
ASSIGN/V4=PROBEDATA("5.0","T1")/2+PROBEDATA("PRBRDV","T1")
ASSIGN/V5=RAD2DEG(ATAN(PT2.Z-PT1.Z)/(V2-V4))
I wish I could do that, the top surface Of this plate is a radius and some part # an spline , a vision system is actually reading this diameter about .040 microns smaller than the actual CMM reading, i also tried reprorubber and it was 003 microns different then the Cmm.
Use vision system if you can. I have vertical comparator & use that instead of CMM (providing parts ain't too large). I had exactly the same problems that originated from the simple fact that the thread is being machined all over the countersink regardless of the CAD model rendering CMM point-to-point methods practically useless.
