Page 3 of 4 FirstFirst 1234 LastLast
Results 21 to 30 of 32

Thread: standardized work instructions for writing cmm programs.

  1. #21
    Member
    • Additional Information
      • Years CMM Experience:
      • 26
      • Software Version:
      • 2012
      • Machine Type:
      • BnS Global Advantage( brand spanking new machine!)
      • Location:
      • Fort Collins, CO(work)
      • Job Function:
      • Layout inspection, cmm programmer. I was using my real name on this board, Robert Squire. I forgot my password and user name, go figure, so I just went with my email name. Orinally from Michigan, Ann Arbor area was home for 20+ years. Colorado now since 2005. If you know me drop me a note!

    Join Date
    Oct 2011
    Location
    Fort Collins, CO(work)
    Posts
    31

    Default

    We tried this at a previous employer of mine. It just doesn't work that way. You cannot standardize how a programmer writes his programs. You can do very basic standardization. But not across the board standardization. Too many variables as someone said above. It's a nightmare, but it gets worse if the management doesn't understand this and they INSIST on standardization. Then you are fighting your boss or your bosses boss and you will lose the battle. This is one reason of many I not employed there anymore. The best you can do is to have one programmer writing all the programs. Then allowing the (trusted)others to edit as they see fit. It does all vary because of the experience of the programmer(s)

  2. #22
    Senior Member
    • Additional Information
      • Years CMM Experience:
      • 3 years with PCDMIS
      • Software Version:
      • CAD 2010 MR3
      • Machine Type:
      • Hexagon
      • Location:
      • Somerset, UK

    Join Date
    Apr 2010
    Location
    Somerset, UK
    Posts
    211

    Default

    Quote Originally Posted by jdpower1 View Post
    Just a thought...

    You'll want to have about an 8 page list of "What if's". Such as, "If you are asked to update commands due to an alignment change..." or "If you are asked to carry nominals back to the feature..."

    I can't count how many hours of fixing time or the number of irrepairable programs I created while learning to use this beast. The depth of knowledge needed can only come from time, trial, and tears.
    It's not just CMM programming knowledge that your manual would have to cover, machine accuracy, fixturing, vectors, trig, an understanding of manufacturing defects i.e roundness, how to inspect a job manually, how to interprate drawings correctly (GD&T), cartesian coordinate programming.

    A time served CNC maching centre setter is an ideal person to train to be a CMM programmer as they will have covered an awful lot of the underlying knowledge that's required for the job.

    Fair enough to set out some guidelines for naming features, and whether you decide to put your dimensions throughout the program or all at the end, but there's very little else that can be nailed down other than vagaries such as 'the tighter the tolerance of the hole the more hits are required'.

    Best of luck, I'd love to see a copy if you're forced to do it.
    Mnfctr: Hexagon,
    Model: Global B Perf,
    Hardware: Control FBII,
    Software: PcDmis CAD 2010MR3
    PH10MQ, TP20, Leitz equipped

  3. #23
    Senior Member
    • Additional Information
      • Years CMM Experience:
      • 12
      • Software Version:
      • 2013 MR1
      • Machine Type:
      • BS One
      • Location:
      • Tulsa, Oklahoma
      • Job Function:
      • Quality Inspector and Engineer Babysitter
    REDWINGJ's Avatar
    Join Date
    Sep 2008
    Location
    Tulsa, Oklahoma
    Posts
    102

    Default

    I still think your best bet is to write a program using the Hexagon block as a baseline for how you want certain features programmed. A previous employer of mine had standards related to the flow of a program. For example they wanted all the profiles, surface hits etc.. to be at the beginning and all the holes at the end. In a machine shop environment it made sense because there were many times when holes were omitted and then put in at a later time so your program had to be flexible and easy to adjust if needed. Having all of one certain type of feature grouped together made it very easy to mark the program accordingly if needed. Honestly though I think if there were a "Standard" method of programming it would have already been written and it would be for sale somewhere. There are guidlines that the Level 1-3 classes emphasize, like how to create an alignment, but a standard method to programming is going to have you chasing your tail. I hope you get something figured out that works for you. Best of luck!
    PC-DMIS 2011 MR1

  4. #24
    PC-DMIS demi-Guru
    • Additional Information
      • Years CMM Experience:
      • 20
      • Software Version:
      • The Best 3.25MR3 and 3.7MR4/3
      • Machine Type:
      • All kinds
      • Location:
      • Belle River, Ontario
      • Job Function:
      • President of Metrology Solutions
    jjewell's Avatar
    Join Date
    Mar 2006
    Location
    Belle River, Ontario
    Posts
    1,047

    Default

    Quote Originally Posted by robertcolorado View Post
    We tried this at a previous employer of mine. It just doesn't work that way. You cannot standardize how a programmer writes his programs. You can do very basic standardization. But not across the board standardization. Too many variables as someone said above. It's a nightmare, but it gets worse if the management doesn't understand this and they INSIST on standardization. Then you are fighting your boss or your bosses boss and you will lose the battle. This is one reason of many I not employed there anymore. The best you can do is to have one programmer writing all the programs. Then allowing the (trusted)others to edit as they see fit. It does all vary because of the experience of the programmer(s)
    +1

    Maybe after they have been trained to write a guideline....standardizing alignments and naming conventions, whether to put dims at end of program or not...those are about the only things you could put in.
    Jim Jewell

  5. #25
    Junior Member
    • Additional Information
      • Years CMM Experience:
      • 0+
      • Software Version:
      • PC-DMIS CAD++ 2011 MR1
      • Machine Type:
      • GLOBAL ADVANTAGE
      • Location:
      • tenderEGOville, weAreExpertsAtEverythingWeDo
      • Job Function:
      • Idiot
    IQ_drop's Avatar
    Join Date
    Feb 2013
    Location
    tenderEGOville, weAreExpertsAtEverythingWeDo
    Posts
    4

    Default

    Quote Originally Posted by robertcolorado View Post
    We tried this at a previous employer of mine. It just doesn't work that way. You cannot standardize how a programmer writes his programs. You can do very basic standardization. But not across the board standardization. Too many variables as someone said above. It's a nightmare, but it gets worse if the management doesn't understand this and they INSIST on standardization. Then you are fighting your boss or your bosses boss and you will lose the battle. This is one reason of many I not employed there anymore. The best you can do is to have one programmer writing all the programs. Then allowing the (trusted)others to edit as they see fit. It does all vary because of the experience of the programmer(s)
    At my place of work, they do not want any new or advanced techniques used because they want everyone to be able to understand and edit the programs. I disagree with that and think the programs should be controlled and locked out so that a person can not accidentally change it. I don't think we are ISO compliant in our program control. We have had crashes and we can't be absolutely sure from one time to the next that the program will run exactly the same or different because we have so many operators that think their way is the best and will change the original program because that's the one they use. We all only know what we know and some are more dangerous because they think they know all when they really know very little. It's sad.
    Mulitple suggestions, one decision maker. Sorry guys, I should be the one. We'd be so much farther ahead now if it were so.

  6. #26
    PC-DMIS demi-Guru
    • Additional Information
      • Years CMM Experience:
      • Since 1986
      • Software Version:
      • V3.7MR4-V2011MR1 CAD++ and Calypso 5.2
      • Machine Type:
      • B&S PFx 4-5-4, Global 5.7.5 & Zeiss Contura HTG
      • Location:
      • Columbia City, IN
      • Job Function:
      • CMM Programmer
    dwade's Avatar
    Join Date
    Feb 2006
    Location
    Columbia City, IN
    Posts
    1,786

    Default

    Quote Originally Posted by IQ_drop View Post
    At my place of work, they do not want any new or advanced techniques used because they want everyone to be able to understand and edit the programs. I disagree with that and think the programs should be controlled and locked out so that a person can not accidentally change it. I don't think we are ISO compliant in our program control. We have had crashes and we can't be absolutely sure from one time to the next that the program will run exactly the same or different because we have so many operators that think their way is the best and will change the original program because that's the one they use. We all only know what we know and some are more dangerous because they think they know all when they really know very little. It's sad.
    We have gone with a controlled master program location, the ops download the programs then run them in operator mode, if something happens to the program, (a DEMON crash or someone playing where they shouldn't be playing) and the program is fubarred, they just download the master again and rock-n-roll again. Only three people in our facility have access to the master programs.
    Xcel & MicroVal Pfx & Global 37mr4 thru 2012mr1sp3
    Contura Calypso 5.2

    Lord, keep Your arm around my shoulder and Your hand over my mouth. Amen.

  7. #27

    Default

    Quote Originally Posted by SYBURR View Post
    So I've been tasked with coming up with creating a template or standardized work instructions for writing cmm programs.
    With six inspectors with varying degrees of experience writing and/or desire to write cmm programs it has become necessary to get everyone on the same page. Has anyone wrote some sort of guidelines for writing programs? The guide would have to be tailored to the lowest common denominator so that the least experienced would be able to edit the programs written by the more experienced and as they become more experienced more advanced concepts could be introduced. So what I'm looking for is a guide or template or standard procedure for writing a cmm program that uses the basic functions and stays away from advanced functions. For example drive moves are a basic function and clearance planes are advanced functions.
    Thanks in advance.
    I don't think the work instructions are for writing the CMM program but more of the entire process of writing a CMM program. This sounds like ISO stuff and it is a requirement.

    For example, the following points would likely be addressed:

    - Who initiates the need for the program.
    - Is this a new program or a revision of an existing program (revision will require its own steps).
    - Some sort of negotiation of what the program will do (some items may be checked by means other than a CMM)
    - What information is provided for writing the program (cad model, print, other)
    - In the program naming convention of features and how the data is output (output file name for example).
    - How the program is stored and what is it called (network drive or local folder, name of program based on part identification, etc).
    - Backup and method of identifying revision (revision based on print revision?).


    If you use a template for the initial program and you want to include this then it should be described as well along with everything relevant. This means the template must be revision controlled, backed up, accessible by everyone who will use it, and documented. If it is something that you use to simply make writing the first part of your program easier it might be a good idea to not mention it.

    If you write it in too much detail then it will be a problem. You can write whatever you want but you *MUST* follow whatever you have written. For example if you write something like "must scratch head three times while pondering the print" then you need to do this.

  8. #28
    PC-DMIS demi-Guru
    • Additional Information
      • Years CMM Experience:
      • GEOPAK a long time ago, then I forgot all....
      • Software Version:
      • 2010MR1, 2012MR1
      • Machine Type:
      • Advantage 7-10-7, Image 7-7-7
      • Location:
      • eastern Pacific ocean
      • Job Function:
      • scapegoat
    sealevel's Avatar
    Join Date
    Aug 2008
    Location
    eastern Pacific ocean
    Posts
    1,541

    Default

    Quote Originally Posted by IQ_drop View Post
    At my place of work, they do not want any new or advanced techniques used because they want everyone to be able to understand and edit the programs.
    sounds like a recipe for disaster.

    Quote Originally Posted by dwade View Post
    We have gone with a controlled master program location, the ops download the programs then run them in operator mode, if something happens to the program, (a DEMON crash or someone playing where they shouldn't be playing) and the program is fubarred, they just download the master again and rock-n-roll again. Only three people in our facility have access to the master programs.
    +1
    an infinitely more sane approach. Everyone can't be armed with the Colonel's 11 secret herbs and spices. It would ruin the chicken, and the finger pointing at who was responsible for over-seasoning would be a circus unto itself.
    ^^^^^^^^

  9. #29
    Senior Member
    • Additional Information
      • Years CMM Experience:
      • 2
      • Software Version:
      • pcdmis 2011 mr1
      • Machine Type:
      • brown and sharpe
      • Location:
      • Da UP, eh?

    Join Date
    May 2011
    Location
    Da UP, eh?
    Posts
    167

    Default

    I think the idea that management types have behind standardizing the writing of CMM programs (and this seems to pop up all over the place in all types of industry) is that they want to assure quality through controlling the method. That just isn't going to work, PEOPLE are the part that can be controlled and standardized. What I mean by that is people can all be given/taught the SAME KNOWLEDGE BASE. That is the solution to preventing fubar'd programs and crashes and false results and all that ugly stuff, not some long boring document that will make their eyes glaze over after a few paragraphs. Training classes are great - experience is better. What is frequently the most valuable learning experience? MAKING MISTAKES. One of life's more frustrating catch 22's.

    SO - I think what you should propose to your boss who wants this programming process standardized is to forget standardizing the writing and execution, naming conventions, structure, etc. and to INSTEAD STANDARDIZE TRAINING and put processes in place to ensure that the people you have doing the CMM programming and operation are all as equally knowledgeable and skilled as possible. Assuming that you don't keep all your programmers locked in individual cages (like I am), they will hash out amongst themselves a more-or-less standard method and program structure that works best for themselves and the company.

  10. #30
    Senior Member
    • Additional Information
      • Years CMM Experience:
      • 13
      • Software Version:
      • 2013
      • Machine Type:
      • Global Performance, Image
      • Location:
      • Texas
      • Job Function:
      • Q.A. CMM Programmer
    CMM Guru's Avatar
    Join Date
    Apr 2010
    Location
    Texas
    Posts
    203

    Default

    Quote Originally Posted by frazz View Post
    I think the idea that management types have behind standardizing the writing of CMM programs (and this seems to pop up all over the place in all types of industry) is that they want to assure quality through controlling the method. That just isn't going to work, PEOPLE are the part that can be controlled and standardized. What I mean by that is people can all be given/taught the SAME KNOWLEDGE BASE. That is the solution to preventing fubar'd programs and crashes and false results and all that ugly stuff, not some long boring document that will make their eyes glaze over after a few paragraphs. Training classes are great - experience is better. What is frequently the most valuable learning experience? MAKING MISTAKES. One of life's more frustrating catch 22's.

    SO - I think what you should propose to your boss who wants this programming process standardized is to forget standardizing the writing and execution, naming conventions, structure, etc. and to INSTEAD STANDARDIZE TRAINING and put processes in place to ensure that the people you have doing the CMM programming and operation are all as equally knowledgeable and skilled as possible. Assuming that you don't keep all your programmers locked in individual cages (like I am), they will hash out amongst themselves a more-or-less standard method and program structure that works best for themselves and the company.
    +1 on all that...
    Keeping in mind that the inspector is the one who accepts/rejects product, not the CMM. Meaning the program is subject to the inspector, not the other way around... There is no substitute for the wisdom it takes to run a program and sort things out as needed. Only experience teaches this.
    PC-DMIS CAD++ 2013
    Brown & Sharpe global performance

Similar Threads

  1. Writing programs on a Sheffield for a Brown & Sharpe CMM
    By Beachman in forum PC-DMIS for CMMs
    Replies: 3
    Last Post: 01-24-2012, 02:13 PM
  2. Replies: 3
    Last Post: 12-21-2011, 10:51 AM
  3. Operator Instructions
    By mldavies in forum PC-DMIS for CMMs
    Replies: 5
    Last Post: 01-28-2010, 08:59 AM
  4. Probe Rack Cal Instructions
    By Gary Fitzgerald in forum PC-DMIS for CMMs
    Replies: 3
    Last Post: 05-15-2009, 04:56 PM
  5. Installation instructions 4.3
    By Ragin1 in forum PC-DMIS for CMMs
    Replies: 3
    Last Post: 07-29-2008, 09:21 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •