detect restart sequence ? - Page 2


Page 2 of 2 FirstFirst 12
Results 13 to 22 of 22

Thread: detect restart sequence ?

  1. #13
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    there is no ODD behavior involved... Why is that strange?
    there are discussed issues about restarting : " scroll + IL + green (method_1)" vs " search + restart button + green (method_2)"

    in the attached movie, there is restarted same location, with method_1, and after with method_2; as you can see, method_1 failes, while method_2 goes nice

    so far method_1 skips things like :
    ......M867 from program begining
    ......tailstock retreat from program begining
    ......so there is something else

    Whether you "Skip" anything in a restart mode is totally up to you and your programming method (what ever the hell that might be)
    there are behaviour differences between these 2 methods; as you can see, in post 10, i had to modify the program for method_1 to work

    so skiping depends on restart method, and not on code

    programming approaches may vary; you can obtain same part differently ; whatever you do, can't avoid M55 / M56 for tailstock ...

    WTF has frying potatoes got to do with a Sequence Restart Skip programming sequence???
    i don't know ... really have no clue ... ... is it grammatically incorrect?

    with a bit of luck this thread can turn into a war zone ... again

    Attached Files Attached Files
    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  2. #14
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    Quote Originally Posted by broby View Post
    you and your programming method
    this post is only a modest example; nothing more ...

    ( . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . )

    beside structure from post7, i also log data for each code variant

    if enough functionality is added to consider a new code variant, than i keep the old one, and make modifications inside a new copy; for each variant i highlight differences; in the attached image, variants 5..7 are for restart procedures, and they are labeled "unsafe"

    latest variants may be machinable, or demo-tests ... like this i can always go back to a previous safe state and continue from there

    programs, and not only, i always back-up on a memory stick and also at home ... just in case, if one PC crashes, i have a spare 1 meter near, or at home, that has all i need to continue without interruption

    for data backup there are also clouds, but i don't like them, because they require net connection ...

    i use " free file sync " ... here : FreeFileSync - Free Backup and File Synchronization Software

    Attached Thumbnails Attached Thumbnails detect restart sequence ?-01-jpg  
    Last edited by deadlykitten; 06-13-2016 at 04:11 PM.
    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  3. #15
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    lL + play / skipping behaviour solved
    if "IL+play" skips something, than just insert skipped code @ each operation begining

    in my case :
    .... tailstock to safe position, when tailstock is used, otherwise key must be switched before restart to right, and after restart, to left
    .... M867 to allow M66

    note : code add-ons are not neccesary when restarting by " search + restart button + play "

    Code:
    O....
    
    send tailstock to safe position
    
    variables initialization
    
      ( * )
    
    IF [ VRSTT NE 0 ] NEND
    
    M867
    safe position 1
    < index M66 > < S / SB ... > < tool approach M63 >
    cutting
    safe position 2
    
    NEND
    
    RTS


    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  4. #16
    Registered
    Join Date
    Jul 2010
    Location
    united states
    Posts
    269
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    Mill guys are familiar with the number search+cycle start method, however mills don't require the IL button be pressed to do so.
    On a lathe, safety can only be achieved by not running the machine. If you want it completely safe, turn off the power, kick the breaker, shut off the lights and go home. Respectfully.
    I completely understand "dummy proofing" a setup, but as soon as you do, they make a dumber dummy.
    That said, during a restart with an NC tailstock especially (or programmable tow along for that matter) it acts "weird" sometimes. But only to the operator.
    Best thing to do is create safety checks for each operation checking each specific thing that must be true for something to be safe and if it isn't true, do something else first.

    In other words:

    PART.MIN

    CALL OOP1
    CALL OOP2
    CALL OOP3
    CALL OOP4

    M2

    OOP1
    LOAD VARIALBES
    CHUCK CLAMPED?
    TAILSTOCK ENGAGED?
    COOLANT ON?
    SEATS AND TRAY TABLES IN FULLY UPRIGHT POSITION?

    FACE PART

    RTS

    OOP2
    LOAD VARIALBES
    CHUCK CLAMPED?
    TAILSTOCK ENGAGED?
    COOLANT ON?
    ARMS AND FEET INSIDE THE RIDE?

    CENTER DRILL

    RTS

    OOP3
    LOAD VARIALBES
    CHUCK CLAMPED?
    TAILSTOCK ENGAGED?
    COOLANT ON?
    YOU SURE YOU'RE WIFE'S OUT OF TOWN?

    ENGAGE TAILSTOCK IF NOT ALREADY...
    (OR)
    IF ENGAGED, SKIP
    CALL OOP2
    SKIP
    (END OR)
    TURN PART

    RTS

    OOP4
    LOAD VARIALBES
    CHUCK CLAMPED?
    TAILSTOCK ENGAGED?
    COOLANT ON?
    GUESS YOU WERE WRONG.

    ENGAGE TAILSTOCK IF NOT ALREADY...
    (OR)
    IF ENGAGED, SKIP
    CALL OOP2
    CALL OOP3
    SKIP

    THREAD PART

    RTS

    I've also used an operation counter via local/common variables to count how many operations have been done and based what it can/will do on that number

    My advice, simplify your programs.
    Yes, a machine can be crashed.
    I've crashed a machine. Today probably.
    But, you cannot prevent it.
    My recommendation is make it as easy as possible for an operator to restart by limiting his options and at a certain point, the part will be worth less than the time it takes to restart it (possibly) or the costs of crashes due to incomplete parts being restarted.
    IE: if it is incredibly difficult to restart a part once the tailstock has been engaged, the down time of repair and the repair itself will cost way more than whatever part you're making.
    I HIGHLY discourage the down arrow/number search then IL+C/S method of restart. Talked to a guy a year or so ago who had a mill guy running a lathe do that on a lathe that was doing 3 dozen safety checks but he didn't know how to do a lathe restart.
    LB3000 w/ 400hrs on it. Total. He crashed so hard into the spindle he broke the X axis slide of the machine because of it.
    At some point, your macros and safety checks become the CAUSE of crashes rather than the PREVENTION of them.

    Just my two pennies.



  5. #17
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    If you want it completely safe, turn off the power, kick the breaker, shut off the lights and go home. Respectfully >>> that's the spirit ; don't lose it
    CHUCK CLAMPED ? >>> machine checks that by default
    TAILSTOCK ENGAGED ? >>> machine checks that by default
    COOLANT ON ? >>> is it possible to verify if, after M08, operator stopped cooland from panel ?
    SEATS AND TRAY TABLES IN FULLY UPRIGHT POSITION? >>> i always instruct my operators to move turret up & right before a restart ... i guess i may verify from code stuff like that, but i bet on operator's attention

    YOU SURE YOU'RE WIFE'S OUT OF TOWN?
    this is nice ... sneaking jokes

    once i was talking with a friend about old cinema cartoons parents were going with their kids, so not to leave a child alone .... so far so good

    well, guys from cinema, from projector, where inserting xxx scenes during the cartoons these scenes were very short, very fast, so out of kids reaction time, but just enough for the adulst to start thinking ... is that kind of timing, when things happen so fast, and you wake up long after

    i just rememebered something : i was breaking because traffic lights were red ... it was night ... i was looking straight ahead, and i saw a light shape on my rear-mirror ... light shape, not head lights ... so i changed my eyes to the mirror ... and then i saw in front of my car, on the near street lane a nissan gtr ... this guy in a fraction of a second reduced speed from whatever and zig-zaged between 3 lanes ... also, even if breaking was hard, there had been no tire sounds ... also machine front did not lean ... it was so natural ... at green it launched ... i was looking and i continued my way home thinking about that

    I've crashed a machine >>> me too ; so far i recorded 2 crashes : i remember 1st one : i was looking how a big drill was going to hit the chuch, and i did not know what to do so i just continue looking ... i knew i will be in trouble
    it is incredibly difficult to restart a part once the tailstock has been engaged >>> why ? it's so easy & boring ... i have such a setup right now
    At some point, your macros and safety checks become the CAUSE of crashes rather than the PREVENTION of them ... it may be true ; so called " unsafe sentiment of safety " ; when you do something wrong, being sure is correct
    all the best teahole

    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  6. #18
    Registered
    Join Date
    Jul 2010
    Location
    united states
    Posts
    269
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    Quote Originally Posted by deadlykitten View Post
    CHUCK CLAMPED ? >>> machine checks that by default
    TAILSTOCK ENGAGED ? >>> machine checks that by default
    COOLANT ON ? >>> is it possible to verify if, after M08, operator stopped cooland from panel ?

    Yes. This is easy. I do it all the time.
    Check VORD[0095E]

    SEATS AND TRAY TABLES IN FULLY UPRIGHT POSITION? >>> i always instruct my operators to move turret up & right before a restart ... i guess i may verify from code stuff like that, but i bet on operator's attention

    I will simply state that I don't bet on any operator doing anything. Ever. My idea of safety checks include verifying the coolant is on.

    this is nice ... sneaking jokes

    You're welcome.

    once i was talking with a friend about old cinema cartoons parents were going with their kids, so not to leave a child alone .... so far so good

    well, guys from cinema, from projector, where inserting xxx scenes during the cartoons these scenes were very short, very fast, so out of kids reaction time, but just enough for the adulst to start thinking ... is that kind of timing, when things happen so fast, and you wake up long after

    i just rememebered something : i was breaking because traffic lights were red ... it was night ... i was looking straight ahead, and i saw a light shape on my rear-mirror ... light shape, not head lights ... so i changed my eyes to the mirror ... and then i saw in front of my car, on the near street lane a nissan gtr ... this guy in a fraction of a second reduced speed from whatever and zig-zaged between 3 lanes ... also, even if breaking was hard, there had been no tire sounds ... also machine front did not lean ... it was so natural ... at green it launched ... i was looking and i continued my way home thinking about that

    The GTR is a vehicle that shook automotive thinking pretty hard. A Nissan. This is a Japanese vehicle. They make garbage. (Circa 1970's thinking). The introduction of the Skyline, which has several iterations, the GTR made people do a double take. Between that an the Honda NSX, it was a paradigm shift. Ferrari and Lamborghini weren't safe any more. I recall a friend who was stationed in Germany years ago was driving on the Autobahn when he was passed by a Ferrari like he was standing still. Only after it passed did he realize there was an NSX about a foot behind it. He estimated they were doing north of 120mph. The GTR really came to world wide notoriety with the introduction of the R34 GTR with the RB26DETT strait 6 and AWD (In my opinion. Not that it wasn't popular in cult followings. See: Datsun). The R34 to me encompassed the spirit of taking a platform and improving it and at a point looking back to realize "crap, we made something amazing. Where do we go from here?". This is evident to me in the R35, which, while impressive is a complete departure from the spirit of the R32/33/34 Skylines. And it's ugly. This was designed to be amazing, and is. But it isn't good that is amazing. Don't get me wrong, I'd rock a G car any day (Infinity G35/37 2door) but the VQ35DE doesn't have that thrill. And taking it and throwing twin turbos on it, makes it thrilling, but there's just something about a straight 6 that makes my heart skip a beat. Hence my love of the Germans and the E46 M3, but that's another side tangent for another day.

    all the best teahole
    What were we talking about again?



  7. #19
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    I don't bet on any operator doing anything. Ever. My idea of safety checks include verifying the coolant is on.
    i understand, and i don't go like this all the time; i believe that ideas about this issue are between these 2 limits :
    ......1) this is the operator, thank god he come to work, let it work how he can, because cnc spins without him; hope this time he won't forget to push green button
    ......2) jesus is operating

    i go with (1) on stable setups, easy parts, and i use unqualified opeartors, like young girls
    i go with (2) on unstable setups : here i use veterans ...

    i have attached parts crafting chart, for 3 days, each with 2 shifts : yellow is veteran ; blue is newbie

    difference is huge, and so i will take actions to raise newbies productivity ... so i bet on operators , but also do my best to raise their odds, only when necessary

    PS : if someone has an idea how to auto-generate this charts from cnc is more than welcomed

    Attached Thumbnails Attached Thumbnails detect restart sequence ?-untitled-jpg  
    Last edited by deadlykitten; 06-18-2016 at 10:20 AM.
    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  8. #20
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    Quote Originally Posted by OkumaWiz View Post
    you can also number search or scroll to a point in the program that you want to restart at and then press INTERLOCK and CYCLE START at the same time to begin from that point
    i asked myself many times why is there such a restart method ? i think that it is there so to reduce restart duration when you are sure what you do

    otherwise, you can see how life goes by, while the "normal restart sequence" goes through the program, through each code line

    however, meanwhile i worked on this, and i can say that :
    ... i use VRSTT in such a way that "normal restart sequence" jumps over operations, but not over critical points
    ... so i can "restart normally" almost as fast as "INTERLOCK+CYCLE START", and still being safe no more life waste

    ... however, "INTERLOCK+CYCLE START" rocks



    Okuma may optimize the "normal restart sequence", so to make it perform considerably faster, but this involves much more than salting a program with VRSTT

    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  9. #21
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    hy in attached archive you may see a movie where i restart operations 7 and 6, by using normal restart method

    code is skiped, but not over critical points ( in this case : the tailstock )

    thus, after a restart, machine behaves like running a new program from the specified line : its almost as fast as IL+cycle start, but being safe ; kindly !

    Attached Files Attached Files
    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




  10. #22
    scaring mice @ north pole deadlykitten's Avatar
    Join Date
    Jun 2015
    Location
    Antarctica
    Posts
    1731
    Downloads
    0
    Uploads
    0

    Default Re: detect restart sequence ?

    Quote Originally Posted by deadlykitten View Post
    its almost as fast as IL+cycle start, but being safe
    hello all comparing to my last post in this thread, i have reduced with more than 50% the duration of the normal restart sequence ( when using restart button, and not IL+ cycle start, which is a bit risky ) ; video attached

    each operation is inside a soubroutine : OS01, OS02, OS03, etc, and the main program calls each one of them :
    NN01 CALL OS01
    NN02 CALL OS02
    NN03 CALL OS03
    * NNxx represents the adress of each operation, so to be able to restart as desired; content of each soubroutine looks like this:

    Code:
    OSxx
        N****
        IF [ VRSTT NE 0 ] NEND
        // soubroutine content
        NEND
    RTS ( . . . . . . . . . . . . . . . . . . . . . . . )
    this approach jumps over entire operations, like nothing is there

    the key to avoid a failure, is to create the program in such a manner, that each operation can be restarted as an individual

    this involves managing the [ turret safe positions, which are particular to each operation ] and [ machine modes : G270, G271 and G272 ]

    there is a trick : for example, if 2 neighbour operations are using :
    ... both C axis, than C axis is not disengaged and reangaged
    ... C axis and S axis, than C axis is disengaged
    ...... thus there is an optimized transition between machine modes that is performed automatically ( osp300L : dynamic safe positions + adaptive CTR )

    only exception is when the tailstock is involved and i handle it like this :

    Code:
    OSxx
        N****
        IF [ VRSTT NE 0 ] NEND
        // soubroutine content with tailstock codes
        NEND
        IF [ VRSTT EQ 0 ] NSTLS
             tailstock codes ( again, repeated )
        NSTLS
    RTS ( . . . . . . . . . . . . . . . . . . . . . . . )
    to be more exact, code looks like this :

    Code:
    OSxx
        NGHOST G305 (*2) 
        IF [ VRSTT NE 0 ] NEND
        // soubroutine content with tailstock codes
        NEND NOEX (*3)
        IF [ VRSTT EQ 0 ] NSTLS ( N skip tailstock )
             tailstock codes (*1)
        NSTLS NOEX (*3)
    RTS ( . . . . . . . . . . . . . . . . . . . . . . . )
    
    (*1)
    ( tailstock codes are repeated outside the code jumped during the restart sequence, so to make the control aware of tailstock state )
    
    (*2)
    ( in my case, if NGHOST is missing, than restarting operation "n" will lead to execution of operation "n+1", not "n" )
    ( this is because restarting at adress N_random may lead to program execution from the line that begins )
    (with N_random, or from the 1st line after the line that begins with N_random )
    ( this depends at least on NOEX and/or VRSTT presence on the line that is being restarted; maybe other factors are involved as well )
    
    ( G305 is there to avoid stopping the program at a non-execution line during a step-by-step procedure )
    (*3)
    ( someday i will explain more stuff about NOEX )
    as a result, during a restart sequence, only tailstock codes ar being taken into consideration

    i have done this, because i go crazy when i wait near a cnc to perform it's restart sequence, especially when long operations are involved all is build depending on cnc restart flag : VRSTT; another way to use this system variable is when using multi-purpose toolholders on Y turrets ( image attached ), so to manage turret safe positions, depending on VRSTT value, but this is a story for another time / kindly

    Attached Thumbnails Attached Thumbnails detect restart sequence ?-01-png  
    Attached Files Attached Files
    Last edited by deadlykitten; 01-12-2018 at 05:21 PM.
    Dub FX 'NO REST FOR THE WICKED' feat. CAde & Mahesh Vinayakram, https://www.youtube.com/watch?v=rBmMzabdEKQ




Page 2 of 2 FirstFirst 12

Tags for this Thread

Posting Permissions

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


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

detect restart sequence ?
detect restart sequence ?