• Ron went to the right school of mathematics.In the auto industry, at least, incorrect calculation of Cp/Cpk is rampant. Most are calculating Cp asCp = (USL – LSL)/6sigmaand Cpk asCpk = min(USL – mean, mean – LSL)/3sigmaThat sigma is the population, or RMS standard deviation. These calculations are for Pp and Ppk – the LONG TERM capability…[Read more]

  • Yep.Issue a format statement, show 5 decimal places, then BELIEVE EVERY ONE because there is a non-zero digit there.I have actually won quite a bit of money from my supply base making side bets if I measured a part on their CMM then recycled the machine without touching it – how much would it repeat ….Scary how many of them give me answers like…[Read more]

  • Yeah. The key is to record the distributions of the original full blown stats on SOME of the fixtures.Then, bring on the other equipment as needed ahead of the ramp up schedule with enough time to correct. No problems doing the full studies on these when production volumes indicate. Once the measurements are there, the analysis is easy.Straight…[Read more]

  • I wholeheartedly agree.My point is: at the launch, I don’t NEED all the fixtures and machines. It is capacity that will come on line as required.So if I show that with a limited capability study the amount of variation fixture to fixture, and machine to machine I have proof of concept.If I show through well designed control plans and SPC that…[Read more]

  • On the financial side ….If I run all these samples (due diligence work) and the design changes, the OEMs will look at me and say “Why did you run all those parts, stupid…” and I will own them.I like the DOE suggestion. The problem is the acceptance of submissions among them is VERY SQE specific. They are not consistent with what they want…[Read more]

  • The multiple fixtures are to handle volume. I am not saying in any way that I would do this for different product on the same machines. Each part is different. This is carbon copy stuff.And the comment about cutting corners on the modeling only MAKES my point. That DOES happen – I have had one launch out of maybe 20 that has NOT had changes to…[Read more]

  • It is an expensive problem. It costs money to do this. Look at it this way. I have designed a fixture for a machine. Each machine holds 8 fixtures and I have 3 machines. I don’t want to bring in all the fixtures – the design might be bad. If I bring them all in and they are bad, I own them. So I buy my 3 machines and bring in ONE fixture.…[Read more]

  • Right. But 32 parts per fixture is 32 * 3 * 8 = 768 parts. Over twice the PPAP order. I like the customer, but not enough to “freely float” 468 * $30 = $14,040 for him. That could be XMAS bonus.So if I am only going to run 300 pieces and measure 100 – and meet his order and requirements – I can:a) divide that up among all the combinations and…[Read more]

  • Maybe I was not clear.This is a new process, that has not been in production yet. Everything is IDENTICAL. There are no different machine types, fixture types, etc. There are no old machines vs. new machines. The annual volume and the estimated pph said that for this process we needed 3 machines, 8 fixtures each.Of course as the process…[Read more]

  • Sorry. I am at the starting line …Answers:1. We don’t know this. I am talking copies of the same machine. Parallel processes to make required volumes. They “should” be the same, but we all know what that means.2. I don’t understand this one. We would select a center at random. There is no issue because there is no data. I want to minimize…[Read more]

  • ncwalker replied to the topic CMM Rant in the forum General 12 years, 10 months ago

    Well – there is a way. “Better” CMM operators can take the data right into Excel. It is just that most I encounter give you the printouts. And no help with interpretation. (I don’t need it, but others do).What I mean is, engineering is typically the customer of the metrology lab. And data is the product.We all make parts and in the making of…[Read more]

  • ncwalker replied to the topic What is Tpk index? in the forum General 13 years ago

    Well that would be “Theoretical Capability Index” -A number typically used by sales guys when they make promises about what you can actually do. :)Seriously – never heard of it.

  • ncwalker replied to the topic Correlating Gages in the forum General 13 years, 3 months ago

    Correction:Something funky happened with my greater thans and less thans,Here is the decision in words:If T is greater than or equal to t, it means the measurements are different and my CMMs do not correlate.If T is less than t it means the measurements are the same and my CMMs do correlate.

  • ncwalker replied to the topic Correlating Gages in the forum General 13 years, 3 months ago

    If I group CMM methods into the following two categories:1) Measured Datums: The part is supported by unimportant features and the primary datums are measured on each part, then the balance of features are measured back to these individualized datums.2) Fixtured Datums: The part is fixtured on the CMM BY the datums. The datums are measured FIRST…[Read more]

  • ncwalker replied to the topic Correlating Gages in the forum General 13 years, 3 months ago

    hacl,Let me see if I have this right….The paired t-test is for small sample sizes (because if I had a lot of them, I could use the tails of the normal distribution). I am going to measure parts on one CMM and measure the same parts on another CMM (Once?) maintaining traceability of sample numbers.Then, I line these up and calculate the…[Read more]

  • ncwalker replied to the topic Correlating Gages in the forum General 13 years, 3 months ago

    Krishnam,I don’t have any data yet. I am trying to define the test so I don’t waste time taking data I cannot use.ncw

  • ncwalker replied to the topic Lean Project ROI in the forum General 13 years, 5 months ago

    Well, that is a simple method. What it lacks is the time value of money.Do you have Mircrosoft Excel?Search for help on ROI. You can even get templates that have sample problems set up in them. If you don’t want to use the templates, read the help on the IRR function. This function takes a cash flow and converts it to an equivalent interest…[Read more]

  • Woops.Forgot to mention I am tier one to the big 3 :-).Sure you want to suggest pre-control again?(Personally, I love it and have used it successfully in a “less lethargic” industry).Sorry. Result has to smell like SPC.

  • ncwalker replied to the topic Gage Studies in the forum General 13 years, 7 months ago

    Jonathan,And you have stumbled upon the inherent weakness of the MSA manual. You get your result and … what next? The results are in terms like NDC and %GRR, but your gage is measuring things like mm or ft-lbs.If you do not have long term process variation to look at, you are correct in using the study variation. Because it is all you have.…[Read more]

  • Well, I got called out of town on a supplier issue so I have been away for a bit.Here is a re-cap.1) What I have done is make a form for Gage R&R that mimics the 3rd Edition MSA in Microsoft Excel. I know there are a lot of them out there, but most of them are not protected. It is very easy to break the calculations with a simple cut and…[Read more]

  • Load More