Point Evaluation

[[Da...]]

Point Evaluation

If you use point measurement make sure you have the right point type for what you are measuring.
This can and has create problems for inspection results.

Thanks Dan

Capture.JPG

[[Ke...]]

Truth!

But, also, does anyone have a cheat sheet, for best application of each? Personally, I have never used Net or CAD Face points...

[[SH...]]

Is there is any special rules for what points should be used at what condition?. Generally I use the default option touch point.

[[Da...]]

Please sign in to view this quote.

If you use only touch and it is a compound probing your data is bad.

[[SH...]]

Please sign in to view this quote.

Please explain how can I correct it, that online manual is only a flash in the pan. It too brief.....

[[Da...]]

If it is a point with I J K values not set to 1
you need to use space point.

Capture.JPGCapture1.JPGCapture2.JPGCapture3.JPG

[[SH...]]

-

[[Da...]]

As for Net Point and Cad Face Point

I don't have an example at this time. I can see it mathematically, but don't know when you would need them to check a part.

[[SH...]]

I think cad face point only used for free form surface, but net point l have no idea.....

[[Er...]]

Please sign in to view this quote.

Are you sure? Let's say you are measuring a "precision" cast 3d-surface with points specified on the drawing, X Y Z i j k. And based on your results, a CAM-program is generated. To machine the part. Do you still stand with your statement, to use Space Points with radius compensation in the nominal vector and projection to the nominal location?

[[Da...]]

Yes
LOL

[[Er...]]

Care to explain why?

Dont you think a plane point with radius compensation in actual vector is more suitable?

[[Da...]]

You have the Point definitions

Make your own call.
Make sure you verify your results either way.

[[Ia...]]

Please sign in to view this quote.

Earlier you answered "Yes LOL" as if it was the dumbest question ever asked. And now after he asks you to explain, this is how you respond?

LOL

[[Da...]]

Please sign in to view this quote.

Not dumb question just not getting in to it with him.

[[Da...]]

Let me just put it this way

Do you own testing and use different Stylus Systems(not just straight down) and you will understand why I say use Space Point.

But if you get better results with plane point use it.

Thanks
Dan

[[Er...]]

The reason Im asking is because you said that if none of the vector isnt parallell to any of the coordinate systems axis. Use space points. While it eliminates to cosine error otherwise introduce by a touch point, (that compensate in the vector of the closest axis). It does so in the nominal vector by default. It also create a projection back to the nominal location. How ever, on the 3d surface there is no guarantee, that the nominal vector is a good option, Its more likely to be more accurate at a flat surface.

Lets say a milling head is to scrub this surface 0.01mm, based on the value of your space point. Dont you see any risk of machining in the air, or to deep?

A general statement like yours is in my opinion just to general. But I tought maybe you hade some solid research behind your claim instead of a LOL.

Space poins is handy to have, but not a general solution.

And if you dont like me, sure, atleast dont give others false hope of a solution for everything.

[[Ri...]]

No one seems to give my boy Midpoint some love. 🙁 *cries as Gear guy*

[[Ri...]]

To add input to the topic though - I tend to use Space Point a lot. I've gotten great results out of it. Plane Point is very useful too though, especially when working with RPS/3d Best Fit Alignments.

I cannot say that I've compared a thickness/distance measurement between using Space Point/Plane Point. Mostly because I tend to not use Points a lot, and also majority of the Points that I would use for these type of measurements would be in the vectored in a single axis only.

[[Er...]]

Please sign in to view this quote.

Midpoins are great, I use them when measuring some of our turbine blade roots. Point corrections is more trouble then help at that specific task.
Even when scanning sometimes.

[[Ch...]]

Eric,

Would you like to elaborate one how you measure turbine blade roots?

[[Er...]]

Sure, you have 9 days of nothing else to do? 😃

Anything specific you have in mind? Blades are like a whole category in it self.

[[Ch...]]

I get it and was just fishing for some good brain food. I measure on turbine blades for automotive turbos. I've not had the chance to pick the brain of someone who measures aerospace turbine blades.

[[Er...]]

Theres just to much, and a brief description isnt of any help. But alignment is crucial, (as always) and we tend to use lsq only for everything that has with it to do. Pressure flanks control pretty much every DoF. Some times a standard alignment is good enough. Some times we need Geometrical best-fit. Some times we devide pressure and suctions side separatly. Rotating by actual flank angle, so one CS-axis is prep, then best fit with translation, only along that axis. Then rotate it back, repeat at other side. Calculate average and translate base alignment.

Theres just so many variants. And hard to give a quick how to. I see a lot of our suppliers when visiting (quality assurance and some times trouble shooting or brain storming (stupid corona, no paid travles any more)), in a big picture, the whole blade world does the same thing, with small tweaks.

Its easier to say something of value if you are specific 🙂

Oh, Just to be clear. I dont measure aero space blades. I measure blades for power generation. Both compressor (fan) and turbine blades. Range from 25mm to ~1000mm in height. Our blades must withstand force that would get an air plane to cry 🙂
But to make it confusing we are manufacturing rb211 and trent engines (Rolls Royce) But they will never fly, land based. But customers try to force us to stick to Aero space standard on those two...

[[Ch...]]

Yeah, I realized after I posted that it was too general. Just looking for some gold. I've found out that sometimes people will give away the keys to the vault if you just ask.