[wi...] Posted August 19, 2020 Share Posted August 19, 2020 Curious how many of you measure on a micron level? I have tolerances less than .005mm on most parts. Curious to see what uncertainty and variance you've seen with your machines. What strategy's do you use for parts? Stylus systems? (Carbon Fiber?) Scanning vs Probe points Looking to see what other people are using. Link to comment Share on other sites More sharing options...
[SH...] Posted August 19, 2020 Share Posted August 19, 2020 It depends upon what characteristic you check, If it is diameter ,it is possible. True position, concentricity will be difficult. Usually I use scanning for critical dimensions. I use tungsten carbide , all are zeiss accessories. I think temperature plays a critical role here. What field you are working, aerospace ?? To be honest I have not worked with 0.005mm tolerance.. Link to comment Share on other sites More sharing options...
[Ro...] Posted August 20, 2020 Share Posted August 20, 2020 in my last job we had several parts that had a diameter cut to +/- 0.0002" inch (0.005mm), we just went with class XX Carbide gage pins because you could actually feel the diameter as it got bigger & smaller. Also the pins were sent to the machine so the operator could inspect the parts 100% without leaving the cell.fgdjfgdjhfgdjhfgd.JPG Link to comment Share on other sites More sharing options...
[Cl...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. A couple things that I would do: First thing is, I would let the part acclimate to the CMM room environment. I would use a high number of scanning points, filters/outliers. Slow scanning speed. A robust and looped alignment. Use the Zeiss cook-book for your strategies/evaluations. That's a good place to start. Link to comment Share on other sites More sharing options...
[Ow...] Posted August 20, 2020 Share Posted August 20, 2020 Probe material/length/temperature/evaluations/cleanliness/handling and secure yet non-deforming holding fixtures is important, as well as the form/shape of the material being checked. Having a master part with known/certified dimensions available to verify the program accuracy and to throw on when the accuracy is questioned is another good practice. Link to comment Share on other sites More sharing options...
[SH...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. Does stylus extension have a significant impact in measurement?, I have one with thermofit rest of them is aluminium. Link to comment Share on other sites More sharing options...
[Ow...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. Yes, especially with positions. If you can ensure the CMM room temperature doesn't change more than a couple of degrees, it shouldn't be too big of an issues but, the coefficient of thermal expansion on aluminum is much, much higher than carbon and will grow more for every degree in temperature increase. The rigidness of the stylus extension also plays an important role in accuracy and the speed it can be run at, I.E> less rigid, slower speeds needed to reflect bending pressures. Link to comment Share on other sites More sharing options...
[SH...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. Ok Link to comment Share on other sites More sharing options...
[Da...] Posted August 20, 2020 Share Posted August 20, 2020 We work with closer tolerances than the machine (Micura) is supposed to be able to hold. We check size down to +/-.000050". Cylindricity down to .000050", Roundness down to less than .000050". Flatness, Parallel, Perpendicularity, Runout, Concentricity, all of it ! I usually don't mention any of this because I am tired of explaining myself to every one who says that "I can't hold that" based on a machine spec. Calibration numbers tell a different story, but what ever. Almost all of our parts are under about an inch and a half. They are usually ground to under a ten RA. They are consistently round and straight to under .0001". Most of you never see parts in this good of condition. The biggest issue by far, Is trouble holding the parts firm enough to get a reading , and not so firm as to crush or distort the part. we purchased a small lever type load meter to try to push parts around and see if they move under small loads. We had a gear that we were checking just about a week ago. It had a parallel tolerance of .0001" , measured near the tips of the gear. I built a fixture to hold the part by the gear teeth vertically so we could get to both side directly. Even though I have a XTR head , I use a TEE shaped probe to avoid rotating the head . The clamp is essentially a parallel clamp , ground on all surfaces. Its held in a precision vise, so it can't wobble. At first the parts were checking high on par..000159" or more. Looking at the fixture we found that the gear had taper on the gear teeth, so once clamped they still could pivot pretty freely. Clamping harder produced higher roundness on the ID. I removed the jaws and put a relief down the center of the clamping surface on one jaw, and down both sides on the other this gave us three point contact and allowed lower clamping force with almost no pivoting . Parallel on the same pc. went to .000018" . it repeats within .000005". This part is around 3/4" in diameter, and .500" wide. You can't get this kind of repeatability on large parts. We use 100MN at 50% dynamic as standard around here. Everything is cleaned and re-cleaned before measuring. Temp in the lab is about 69 degrees. We still use the temp sensors, either on the part or at least on the fixture. Every diameter we check is also analysed for roundness. We use LSQ most times for the added repeatability, changing to OTE doesn't move things much more than .000010" So about the stated accuracy of the machine , and the calibration numbers... The stated accuracy is expressed as the total error over almost the entire travel of the machine. We're using about an 1 1/2" of that. Do you really think it has the same error over that small distance ? Our calibration number usually run about 50 % of the allowable error. Once again over the entire distances of the machine. They use a trumpet chart for a reason. In short distances the error is quite a bit lower than the stated accuracy. We use an XTR head and a XT in the other machine we get about .000010" or less difference between machines. Link to comment Share on other sites More sharing options...
[Da...] Posted August 20, 2020 Share Posted August 20, 2020 Temperature plays a very limited role if your parts are small. COE is expressed in INCHES per DEGREE per INCH of material so one of our parts is .125 in diameter with a about .07 ID . So for steel that's .0000072" per inch. We're at .125" so It takes an 11 degree difference to change the size by .000010" !!! We also run a few parts around 13" in diameter. So same formula gives us a WHOPPING .001 difference for the same 11 degree difference !!! Not everyone gets this, My boss made me soak the .125 part overnight 😕 Link to comment Share on other sites More sharing options...
[Cl...] Posted August 20, 2020 Share Posted August 20, 2020 Just out of curiosity, are these fuel injector parts? Link to comment Share on other sites More sharing options...
[Da...] Posted August 20, 2020 Share Posted August 20, 2020 Some parts are for fuel injectors but not all . Hydraulic parts take up a sizable chunk of our production. Some of the hydraulic parts make metal to metal seals that hold thousands of pounds of pressure without leaking. We are a precision grinding job shop. We take almost anything that comes thru the door. Link to comment Share on other sites More sharing options...
[Cl...] Posted August 20, 2020 Share Posted August 20, 2020 Thanks, I thought so. I ask because up the hill from us is an automotive manufacturer that does pretty much the same type of small parts with tight tolerances. Believe it or not, they use an OGP with a Renishaw PH9 (I think) and feather probes. In addition they use the Talyrond to check roundness. Link to comment Share on other sites More sharing options...
[Ri...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. Oh no! Not the feather probes! Yikes. Link to comment Share on other sites More sharing options...
[Ch...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. These Micura's are crazy good. All of my metrics are below 50% of allowable error as well. I have to metrics that are under 10% of allowable error. Link to comment Share on other sites More sharing options...
[Ri...] Posted August 20, 2020 Share Posted August 20, 2020 Please sign in to view this quote. Yes, and the price is nice for one as well. They do have a small measuring volume, but it's a lot cheaper than a Prismo Ultra. Link to comment Share on other sites More sharing options...
[SH...] Posted August 21, 2020 Share Posted August 21, 2020 Please sign in to view this quote. I think micura is for small scale industries and prismo is for OEM companys. 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