Jump to content

Percentage of diameter's circumference within tolerance?


---
 Share

Recommended Posts

Currently having a dispute with customer over some undersized turned OD's that we shipped them. Their findings are valid but they've also acknowledged that our scanning Contura is the better method, they just "need more data regarding the percentage of the circumference under tolerance."

Tolerance is +0/-.0005", so not much, but the nominal is big enough that I don't question the CMM's data at all. Throwing a 2pt min/max at it, getting as low as -.001". The roundness is a big factor and we screwed up there. They just want to know HOW MUCH of each piece's diameter is out of tolerance and maybe some are usable as-is.

I'm guessing I would need access to ALL point sets the 2pt min/max characteristic uses and analyze that somehow. I also made a freeform on the exact same circle path but not sure how I can use this data other than the profiling and CAD evaluation visuals that aren't much different than just a simple roundness plot. "How many points are inside/outside of this tolerance zone?" I also remembered that circles have some manual point masking functions but not exactly sure how this would factor in reasonably. Maybe iterate the feature multiple times in "opposing segments" with the point masking and process my own percentage breakdown from that? Maybe that puts me down the PCM/Special Program route? I have a license of piweb plus so maybe there's something there I don't know about?

It would be easier to just PDF out some roundness/profile plots and say "green good purple bad" but the customer is meeting us half-way here so I would prefer to give them exactly what they want, if it's possible.

Usually I can figure stuff like this out on my own but this one has me scratching my head. Unless I'm missing something really obvious (highly likely).
Link to comment
Share on other sites

This is not too different to what I am trying to accomplish! I am trying to find the available stock on a part that also has to account for runout and roundness.

Are you working from a model? If not, can your customer or your engineering dept supply you with a model?
Link to comment
Share on other sites

Just thinking out loud here. Could you scan a high point density curve around the
OD, switch it to polar coordinates and export the points? Wouldn't that give you
exact point numbers polar values?
Link to comment
Share on other sites

Please sign in to view this quote.

Do i need curve for this? Like i said in the OP, i have freeform and have used it for a lot of basic profiling but i think it holds back data/analysis on purpose to compel you to pay up for full curve in cases like this

Makes sense though, i'm guessing the polar coordinate result would just be the radius since the OD i'm measuring is also the XY in my base align
Link to comment
Share on other sites

You have free form but not curve? I've tried exporting 2d curve points in polar on
my planner seat and so far, I haven't figured out how to do it, if you even can do it?
Link to comment
Share on other sites

I just leaned this today! Try checking the circle the asking for radius measurement. Constrain to the alignment and it will give you a readout of the radius for each point of the circle. (go to a relatively lower point density, or you'll get a lot of points to look at!) Apply tolerances and each point that lacks stock (U/S) will show up red. Not sure how to get a count, but maybe someone here knows how.
Link to comment
Share on other sites

Please sign in to view this quote.

I am not a PiWeb wizard, so I am just thinking aloud, but I believe you can query #of characteristics red, # Green, and total #of characteristics when making a PiWeb report. If the program were to only contain the radii, you could query those and use a formula to report the percentage on the report, perhaps
Link to comment
Share on other sites

Please sign in to view this quote.

Brilliant! This is at least raw data that I can export to an excel file and analyze there, huge thanks! Hope my idea in your other runout thread helped out
Link to comment
Share on other sites

Please sign in to view this quote.

yeah i was lucky to get freeform licenses for both machines...the first machine came with it (about 6 years before i started in this shop), none of the guys programming at that time ever used it, then when they RFQ'd a second identical machine around 2019 (again, before my time) someone at zeiss conveniently left off the free license and no one really noticed until i got curious and figured out how to use it then the question came up of why only one machine had it

ironically i use it a LOT on zeiss parts that we make for some of their ophthalmology machines, their newer prints are "modernized" where it's one sheet with a handful of critical dims called out and everything else defaults to a 0.4mm profile to -ABC- instead of fully defining their very skeletal parts on 5-6 sheets, saves everyone a ton of time
Link to comment
Share on other sites

 Share

×
×
  • Create New...