Scanning with XYZ data

[roscek]

Is there a way to take a scan using XYZ data without manually inputing all the coordinates? I need to scan an airfoil that follows a certain path and hit specific points. It needs to follow the contour and angle of the airfoil not a straight line through a set cut plane. I've attached an image showing just a few of the points to be taken. There will be 4 sections with 28 points in each. In this image Z is up and down, X is front to back, and Y is left to right.

[rangerboat72]

Have you used the TOPOINTS script before?

[MrComment]

Is there a way to take a scan using XYZ data without manually inputing all the coordinates? I need to scan an airfoil that follows a certain path and hit specific points. It needs to follow the contour and angle of the airfoil not a straight line through a set cut plane. I've attached an image showing just a few of the points to be taken. There will be 4 sections with 28 points in each. In this image Z is up and down, X is front to back, and Y is left to right.

If you have the Target Coordinates in a Text File or Excel as X,Y,Z then yes.

Use a Freeform scan then under path definition tab you can import the file of target points into a scan.

[cmmguy]

Wont you need the IJK info also.

[rangerboat72]

Wont you need the IJK info also.

That's what I was thinking. I wonder if it is pulling that in the code???

[roscek]

Since I am running pcdmis 3.7 mr2 I don't have the freeform option. Is there another way that anyone can think of. My XYZ file does have all of the IJK values as well. Thanks for the help.

[MrComment]

Do you want all of the points to be just be regular Auto Vector Points?

If so I could run your data file through a Freeform scan then convert it to Points and then finally save it down to 3.7MR2.

I think that may work at least.

[roscek]

No, I have an SP25 and want to actually scan across the part instead of taking individual hits, but thanks for the offer.

[MrComment]

Have you tried using a perimeter scan?

When you click on your control points 1,D,2 stay close to tip of the airfoil so you just get the end selected as your boundary first. Then with the CAD offset get your Scan Path to cross over the points you want to pickup.

[Don Ruggieri]

Basic?

Are you familiar with BASIC, scripts of prjoects? If so, there was a routine floating around that served as the predecessor to the Freeform scan. It would read data from a file and create a scan feature in the part program. It would work for 3.7, and most other versions. . .

If it would help you, I would be happy to see if I can find it.

[Don Ruggieri]

Found it

I found the utility, so I might as well post it here in case others benefit from it. It is an executable only (right now) as I am not quite able to make sense of the source code, its been a long time.

I've attached a ZIP file with the executable and 3 test cases. These serve to show the formats and to easily demonstrate its capability. For example, scan data with blank lines between groups provides multiple basic scan segments.

It does require vectors, so you might have to fudge those if they are not available. It was written for v3.2, and I just tested it with v3.7MR3. I suspect it will work with more recent versions as well.

It does require that PC-DMIS already be open as far as I know, with a part program open as well. To create a simple test case, open PC-DMIS and a part program, select a probe, and then switch to DCC. Now switch to another (Explorer) window and start the utility. Select a few simple options and see what happens.

[Cumofo]

XYZ, IJKs work well imported as a .csv file. Build one auto point feature with variables for each XYZ, IJK and read the .csv in a loop.

-K?
GAbriel

[roscek]

Thanks for all of the help. The utility that was posted worked perfectly.

[roscek]

I thought everything was fine, but ran into a problem. The scans I created with the utility worked on the convex side but not on the concave side. When I ran the program it scanned the first section of the concave side without making contact with the part. Then at the end of that same first scan it just stopped. There was no error message or anything. The only thing it said at the bottom was that it was waiting for the scan to end. I created the scans using the defined mode with an increment of 1mm, and find nom tolerance of 1mm. I also tried adjusting the find nom tolerance but still had no luck. Does anyone have any suggestions as to why this happened and how I can fix or work around it?

[Don Ruggieri]

try some things

I thought everything was fine, but ran into a problem. The scans I created with the utility worked on the convex side but not on the concave side. When I ran the program it scanned the first section of the concave side without making contact with the part. it?

That sounds just like the DEFINED scan, as you said you are using. If the data is not on the part, it does not care. It travels through the defined path.

Three things to consider;

1) the data is not on the part. Not trying to sound funny here, but maybe the data is bad? or off the part by a probe radius? I suspect the probe is traveling where you told it to go, but the part is not there.

2) the alignment used for this scan is off the part, causing the data to look bad

3) try using a RELEARN scan. This uses the defined data to start with but still searches for the part.