CAM Data

[[To...]]

Is there a simple solution for measuring a cam profile at 5 degree increments? In the past, I've created 3d lines at the 5 deg increments and asked for the intersection of the 3d line and the curve but the processing time was a bit longer than I wanted. I'm guessing there is something in Calypso that makes this more automatic.

[[Cl...]]

Tom, can we see the drawing/screen shot?

[[To...]]

A and B are the bearing diameters at each end. I need to see point deviation every 5 degrees around cam.

cam1.jpg

[[Cl...]]

What about CAD - creating features - section? Slice a thin nominal plane through the model and keep rotating it by 5 deg increments. sectioning and creating a small curve each time. I know it's not the individual point you are looking for, but it should work?

[[To...]]

Clark, I'm using a curve to go around the entire cam shape, but need to get the radius measurement of the cam in 5 degree increments.

[[Ow...]]

I've never used it and you may not have it but, if you go to size, more, cam, there are some characteristics that supposedly allows you to outputs the curve measuring results together with angle, nominal value, actual value and deviation for each point in a text file.

[[Cl...]]

Going to have to think a little deeper about this one. There are a lot
of sharp programmers out there, I sure someone has the solution.

[[To...]]

Cool. I don't when they put that in there but it's been awhile since I had to work with a CAM, I may have just overlooked it. The only thing I don't understand is the Cam Shape property. It appears that I can get the lift data as though there was a roller or barrel tappet which means I can figure out what it's going to do when it's "camming". Currently need just the radius value from the centerline.

[[Ri...]]

I am not sure this will work but it might get you close. Measure the Cam with curve. Go to Size/Curve Dimensions/Curve Length. This will deliver the circumference of the curve. Take that value, divide by 72. Now go to your Cam Curve and open up the Strategy. Set your step width to the circumference divided by 72, and the number of points to 72 and put a check mark in the field for single points.

[[To...]]

Single points? Richard, I'm trying to sell a Zeiss, not a Hexagon...LOL

[[To...]]

Just found a new option under Modify Nominals in the curve feature. It allows you to set the angle, rotating around an axis.
so I figure I will use the 72 points with the Zeiss suggested 3 to 1 number of scanning points. Then, output the deviation through the curves points output in Results to File.
.
.

[[Cl...]]

I used the thin plane slice and created a cad face point. Then I went into the points options and created a polar pattern every 5 deg. This created all the points around the od. Only problem is they all have the same vector. But that may be due to the type of point I used? A different point selection may produce a point that has an editable vector? I'll play around with this tomorrow.

Pattern.PNG

[[Ri...]]

Clark, instead of using Polar Pattern, use Rotational Pattern. This will rotate the vector as well as the location of the point.

[[To...]]

I think I'm onto something with the new Modify Nominals option.
.
.

[[Da...]]

Not sure this will work, but can you create one line at 0 degrees and ask for an intersect, then ask for simple distance from the intersect and a center point, (3d distance) and then scallop (rotational) pattern the line and intersect ?

[[Da...]]

Please sign in to view this quote.

Won't that have a vector (Slight) error ? IMHO, I think the intersection would be more correct.

[[Da...]]

How's that for a sketch !?!? I think you want to use an intersect.

[[To...]]

Please sign in to view this quote.

Hi Dave,

I agree. Like I said in my original post, I've used this technique before on a scroll pump housing and it really dragged the system down to a crawl. Every time you'd open an intersection or related result, it would literally take 5 to 10 seconds to process the results. Of course I was measuring a whole bunch more points. I was hoping to find something easier. By doing what I stated in one of my last posts, by spreading my curve points out in 5 degree increments and setting the curve to "Radial Deviation" and turning on "curve points" in Results To File, I'm hoping I can see the deviation I need.
.
.

[[Cl...]]

Could you start with a single point and pattern it all the way around at 5 deg?

[[Ow...]]

Tom, Did "spreading my curve points out in 5 degree increments and setting the curve to "Radial Deviation" and turning on "curve points" in Results To File" work as you expected?

I'm not really a fan of southpark (to each their own) but, if it did, I learned something today.

[[To...]]

I haven't measured the part yet. Waiting on some tooling. I'll keep you posted.

[[hu...]]

Please sign in to view this quote.

hello Tom, if memory servies, there is a way for calypso to load points in txt , right?
if yes, maybe a lazy solution , asking designer to create the points with 3D software and then output the points with x,y,z,i,j,k into txt.
it's easy for designer to create the points.

[[Kl...]]

Switch from "NormalDirection" into "RadialDeviations"

The CAD visualisation will be OK, PiWeb curve.FD Shows something
I make a ticket at ZEISS Software Service

OutPut with "old" DrawEdit

radial.PNG