Old question, still need answer. :)

[thenettrainer]

Hi group.

I asked this question a few weeks ago. Originally I posted it in the "Blade" section, but it seems no one spends any time there answering questions. About 1000 views to a few questions, and only a few answers. Or maybe very few use Blade software.

So . . .

OH! I also posted it here. I did get one reply from a lady who was able to help me with part of it. But I still need some help, or an idea.

Below is my original question. Then an update as to where I am with this idea . . .

~

We have 8 B&S CMMs using PCDMIS (3.7 & 4.0) Plus two 25yr old CMMS (Not B&S) We have about 4,000 programs on the old machines, so we're fazing them out slowly.

What I ultimately would like to do is run Blade on data from the old CMMs.

My ideas were something like this . . .

1). Old CMMs can output data in ascii format. We have a translator that puts the actuals into an XYZ/CSV formated ascii file.

Perhaps I can input this XYZ data into Blade from the ascii file in some way. ???

OR . . .

2). Fool PCDMIS into thinking it ran the part!?! OFFLINE: If I could import that ascii XYZ data as Nominals into a PCDMIS program, then run the program, being offline, there would be zero deviation. Therefore the actuals would be the actuals from the old machine.

THEN I could use Blade.

Comments welcome, and much appreciated.

~

Update: (offline) As suggested, I made a csv text file with the XYZ values. Saved it as a .xyz file. Imported that file into PCDMIS. Used AutoVec pt and now have those points as though they were actually run.

I then made feature sets of each blade section.

Now I need to get that data into Blade.

I also went into CAD and pulled nominal points for the sections. However, the Noms won't nearly match the input actuals because they were not taken as discrete points, but was a scan. So the "points" won't match the actuals, but the spline should be the same.

Admittedly, Blade is totally new to me, and there's no one here who knows anything about Blade. But we have customers now demanding we use it for our analysis.

Once again, comments welcome, and much appreciated.

[Wes Cisco]

Well you may not think this is much help: but I recommend you contact the B&S rep who sold you the blade package. You may need/want to get some inhouse training, or at the very least be put in touch with one or two individuals @ B&S who are knowledgable about blade. This is a highly specialized add-on package to Pc-Dmis that obviously very few use. Given that you have posted in several areas of this forum and not gotten much of a response, I think you have to look to B&S for support. There are a couple of other boards you might try. Cmm talk is one. Pc-Dmis.com is another. A last chance is your cmm.com. I doubt you will have any better luck at any of those, but maybe. HTH

[Virgil UK]

Blade.

In order for Blade to work you will need to create 5 files.
*.nom, *.tol, *.flav, *.op, *.mth, and store them all in one directory.
There is some help about creating these files depending on which version of blade you are using.

You can use Blade to process your data, but to do this you will have to create the Blade spline, (nom file), so Blade will know what to compare the actuals against.
This can be done by importing the xyz values of the nominal points.

I have ran Blade to process my results but I have first ran the program in DMIS to get the correct formats, below are the result formats that Blade will accept:-

DIM PROF4= PROFILE OF SURFACE OF SET CV OR FORMONLY USE FIT UNITS=MM
AX NOMINAL +TOL -TOL MEAS DEV MAX MIN OUTTOL
M 0.000 0.010 0.000 0.000 0.000 0.000 0.000 0.000 #--------

POINTDATA HIT# MEAS X Y Z VEC I J K DEVIATION
CV OR
1 -20.489 24.160 16.518 -0.678 0.735 0.000 -0.000
2 -20.426 24.214 16.523 -0.643 0.766 0.000 -0.000
3 -20.319 24.292 16.502 -0.541 0.841 0.000 0.000
4 -20.200 24.337 16.493 0.052 0.999 0.000 -0.000
5 -20.194 24.337 16.490 -0.412 0.911 0.000 0.000
6 -20.190 24.342 16.489 -1.000 -0.006 0.000 0.000
7 -20.191 24.345 16.488 -0.557 -0.831 0.000 0.000
8 -20.205 24.340 16.499 0.892 0.451 -0.000 -0.000
9 -20.201 24.339 16.505 -0.034 0.999 0.000 -0.000
10 -20.200 24.341 16.508 0.057 0.998 0.000 0.000
11 -20.198 24.340 16.509 0.323 0.946 0.000 0.000 .

I would suspect that you can get away with omitting the IJK, especially if it is a 2D section. These results are needed to go into a Blade.rtf file which blade will process.

If running from PCDMIS,

When you create your feature sets make sure you have "text reporting" toggled "on" to get the above format if you are running it from PCDMIS:-

DIM PROF4= PROFILE OF SURFACE OF SET CV OR FORMONLY USE FIT UNITS=MM ,$
GRAPH=OFF TEXT=ON [/COLOR]MULT=10.00 OUTPUT=BOTH
AX NOMINAL +TOL -TOL MEAS DEV MAX MIN OUTTOL
M 0.000 0.010 0.000 0.000 0.000 0.000 0.000 0.000 #--------
PRINT/REPORT,EXEC MODE=END,$
TO_FILE=ON,OVERWRITE=C:\BladeRunner\BLADE.RTF,$
TO_PRINTER=OFF,DRAFTMODE=OFF,$
TO_DMIS_REPORT=OFF,FILE_OPTION=OVERWRITE,FILENAME= ,$
REPORT_THEORETICALS=NONE,REPORT_FEATURE_WITH_DIMEN SIONS=NO,$
PREVIOUS_RUNS=KEEP_INSTANCES

The last lines create the Blade.rtf file from PC DMIS, this is where your results need to be in order for Blade to work.

All you need to do now is run a DMIS prog from within blade, put a jump in the program that will take it to the feature sets at the end of the program.

As Wes has stated Blade is specialized and it is also very temperamental, all the "I's" have to be dotted correctly and even then you get problems. I have had two teaching courses with it and still get problems. Let me know how you get on, but what you want to do is possible.

Darren.

[thenettrainer]

hey Wes & Virgil, THANKS!

Just for the record, training is right around the corner. I plan to have the training very soon. I just hired in here, having worked on CMMs since '74.
Problem was, they needed a solution pronto, and I told them if I could get an ascii file with actuals, I was pretty sure I could "trick" PCDMIS into thinking it ran the parts that actually ran on the old CMMs. I was able to figure that out very easily. But now, to figure out how to get Blade to work!

So . . . I jumped right in, and started working on the solution. LOL

So, yeah, training is in the mix. FYI: I've been a CMM Specialist and GD&T Professional Consultant for over 12 years. Back in the late '80s I wrote a program similar to Blade. But my hopes were . . . Git'r dun! Make the company happy. I become a star. (LOL) Then head over to B&S for training on this new software.

Virgil, thanks tons. You took far more time than I would have expected someone to do.

I see from your helpful notes, I was just missing a few pieces of the puzzle. I'd done most of this already, but couldn't quite get the pieces to all fit!

1). In some instances, I was putting the pieces in the wrong place.

2). With a few missing pieces, none of it wooda worked anyway!!!

Thanks TONS! I'll work on this today, and let you know what I come up with.

If I can ever help either of you with a GD&T or CMM metrology issue (or you wanna learn how to play harp or guitar! LOL) just let me know!

Much thanks, Wayne