[Ch...] Posted May 28, 2019 Share Posted May 28, 2019 I need to write a good justification as to why touch points are more accurate when measuring a short radius vs scanning. I have data to support the hypothesis; However, how can I articulate it into words? So far, I've got that 'I know its better because the data says its better'. Doesn't sound very convincing if you ask me. anyone got any ideas? Link to comment Share on other sites More sharing options...
[Me...] Posted May 28, 2019 Share Posted May 28, 2019 Why are touch points more accurate than scanning? How does your data support that hypothesis? I'm skeptical. Link to comment Share on other sites More sharing options...
[Ro...] Posted May 29, 2019 Share Posted May 29, 2019 no point masking at start and end of scan, so if you're measuring say 15º of arc the scan may filter out a degree or more at the start & end giving you 10º to 13º arc. whereas the same segment with touch points would be the full 15º. Plus i have a Vast XT and my probe qual program is 13 points (x2 forces) on 180º of sphere, so if its good enough to qualify the machine then it should be good enough to measure a feature, and it is. Finally I love using touch points just because it blows the mind of the calypso purists who think that anything & everything must be scanned. its the teenage me (Gen X) just wanting to buck the system, and prove people wrong. Link to comment Share on other sites More sharing options...
[Is...] Posted May 29, 2019 Share Posted May 29, 2019 Measure short radius segments is always a nightmare (points or scanning); normally I try to convince the designers to draw in a different way: Distances, Profile or Position. There are many examples of uncertainty influences due to evaluation range (this is from ZEISS) https://world.probes.zeiss.com/en/Train ... _4084.html Link to comment Share on other sites More sharing options...
[Aa...] Posted May 29, 2019 Share Posted May 29, 2019 In my experience, the masked points at start issue can be easily overcome by scanning the arc in both directions, which is still generally quicker than touch points. Israel, I agree that generally there is a better way control them than radius size. Roberto, I get bucking the system. Link to comment Share on other sites More sharing options...
[Br...] Posted May 29, 2019 Share Posted May 29, 2019 The biggest issue I see with the measurement of a radius isn't so much the strategy, as the interpretation of what a radius tolerance actually looks like. Most CMM software, when outputting a radius tolerance, simply take the diameter of a circle or cylinder feature, and divide it by 2. This introduces all kinds of issues, as all of you know, with how the missing part of a complete circle gets created, often making a center point or axis that is way off. A radius tolerance zone actually controls the surface directly, more like a profile tolerance zone does. The tolerance zone is bounded by two arcs(the minimum and maximum radii) that the part surface must fall within. If the center of the radius is located via dimensions, the arcs are concentric. When the center of the radius is not located by dimensions(like a fillet), the arcs are tangent to the adjacent surfaces and create a crescent shaped tolerance zone. Using constraints with a radius tolerance in Calypso comes really close to this definition. You'll get more repeatable results, and avoid rejecting parts that should be conforming. You can even use formulas within your circle strategy to define the location of the radius to its respective tangent edges. Link to comment Share on other sites More sharing options...
[To...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. But you're qualifying 180° on the sphere. The larger segment on any radius will help the result. Have you ever seen Dynamic Tensor qualification? It scans around and over at 2 speeds after it finishes the Tensor. Regarding being called Zeiss purist, I feel scanning is not necessary for everything. I believe you match the strategy to the requirement. Then, I balance accuracy requirement vs productivity requirement. Clearly, scanning allows one to collect more data which typically creates a more stable data set. I can't argue the issue regarding masking, but Aaron has valid solution though the size of the radius could affect scanning time vs several touch points. Though a larger radius would make me think about taking more points. I think the reality of this subject requires a study.... I'm thinking the actual radius with a 15° arc segment isn't doing much more than resting on something or clearing something and so profile solves many of these types of issues. I invite anyone to enlighten me otherwise. Link to comment Share on other sites More sharing options...
[Is...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. agree with almost everything but if measure radius is a nightmare; Using constraints is Hell (basically value is fake when form deviations are significantly high)... Link to comment Share on other sites More sharing options...
[To...] Posted May 29, 2019 Share Posted May 29, 2019 I think if you're going to use constraints, you're better off to use Radius Measurement. If you constrain the circle, you're still getting an average or LSQ radius result. Radius Measurement gives you the radius value of each point in the circle constrained to its' center. Link to comment Share on other sites More sharing options...
[Br...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. I completely agree. That characteristic is a much better reflection of a radius tolerance. It just reports every point which can make your report look messy. I'm sure you could somehow report the maximum deviation of that Characteristic and report that. Link to comment Share on other sites More sharing options...
[Br...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. Not any worse than letting the software determine the centerpoint with high form error. If this is a concern, I think Tom's recommendation would be better, because then you will see where each point ended up individually instead of an LSQ fit off all the data. Link to comment Share on other sites More sharing options...
[Is...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. agree, then we measure some kind of distances instead of a single radius value Link to comment Share on other sites More sharing options...
[An...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. Right! Link to comment Share on other sites More sharing options...
[An...] Posted May 29, 2019 Share Posted May 29, 2019 Here some thoughts about constraining the center of a radius.Contribution_29_05_2019.pdf Link to comment Share on other sites More sharing options...
[An...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. What you are describing is the drawing in the standard. The tolerance zone is always made by: TWO CONCENTRIC CIRCLES Try to understand!!! Link to comment Share on other sites More sharing options...
[Br...] Posted May 29, 2019 Share Posted May 29, 2019 Please sign in to view this quote. Actually that's pretty much copy/pasted from the standard, so make sure to let them know they wrote it wrong. Hopefully they will try to understand. Here's the full text as it is in ASME Y14.5-2018 Please sign in to view this quote. Link to comment Share on other sites More sharing options...
[An...] Posted May 30, 2019 Share Posted May 30, 2019 Please sign in to view this quote. If you talk about surface points: CORRECT THIS NONSENSE!!! See attached!Contribution_30_05_2019.pdf Link to comment Share on other sites More sharing options...
[Me...] Posted May 30, 2019 Share Posted May 30, 2019 The only point I was trying to make is that touch points are no better or worse than scanning as the OP stated. It's all in how you interpret and analyze the data. Link to comment Share on other sites More sharing options...
[Da...] Posted May 31, 2019 Share Posted May 31, 2019 I deal with VERY close tolerance partial radii quite often, and I have found that a scanned PROFILE is really the only way to accurately determine whether a radius is in the right position and of correct form. By the way, I LOVE how heated and passionate we all are about partial radii 😃 😃 😃 Link to comment Share on other sites More sharing options...
[An...] Posted June 1, 2019 Share Posted June 1, 2019 Please sign in to view this quote. Advice to the designers: Watch the ratio when it comes to PROFILE. See attached.Contribution_01_06_2019.pdf Link to comment Share on other sites More sharing options...
[To...] Posted June 1, 2019 Share Posted June 1, 2019 Andreas, I'm guessing many designers would make the radius BASIC so it wouldn't be subject to a size tolerance. Link to comment Share on other sites More sharing options...
[An...] Posted June 3, 2019 Share Posted June 3, 2019 Thoughts about transforming the Plus-Minus-Tolerance into a Profile-Tolerance. See attached.Contribution_03_06_2019_.pdf Link to comment Share on other sites More sharing options...
[Is...] Posted June 6, 2019 Share Posted June 6, 2019 Please sign in to view this quote. Your calculation is not correct ... the gap (0,0752) is smaller (around 0.014): Link to comment Share on other sites More sharing options...
[Is...] Posted June 6, 2019 Share Posted June 6, 2019 Please sign in to view this quote. I would stop here, plot as it is in the sketch and Report a non-uniform tolerance: "Maybe increasing tolerances in the Corners and add segments of the adjacent geometries" Link to comment Share on other sites More sharing options...
[Ri...] Posted June 6, 2019 Share Posted June 6, 2019 What in the world are you doing where a corner radii is moving ±10mm!!!!???? 🤣 🤣 🤣 🤣 🤣 🤣 🤣 Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in