Z cutting deeper the more time goes by ??? - Page 3


Page 3 of 6 FirstFirst 123456 LastLast
Results 41 to 60 of 105

Thread: Z cutting deeper the more time goes by ???

  1. #41
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Collet manufactur

    PreciseBits from Tinker&Tinker

    Attached Thumbnails Attached Thumbnails Z cutting deeper the more time goes by ???-d4118055-c953-4ad0-80e8-c654c7a827b2-jpg  


  2. #42
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by GME View Post
    HH,

    I haven't heard of CandCNC before. Wow, are their electronics prices high. CNCRouter parts has a Nema 23 plug & play box for $1,545 that's in wide use with service that's second to none. It's a shame a business can charge as much as these folks do and provide no service at all. Since your system is 10 years old, maybe it's time to think about replacing the electronics with something more up to date?

    If you are using Mach3, In guessing that it's just their propriety XML file. If so, you could probably replace it with a generic one, like one on CNCRP's site. I'd reach out on the zone, explain your electronics situation, and see if anyone else has modified their systems. Since it's been 10 years, there are bound to be folks with some knowledge about it.

    I'm sorry that I can't help on the electronics front. I built my own box using the PMDX 126 main board, PMDX 107 spindle control and Warp9 ESS smoothstepper, which connects to the computer with a standard network connection. I also upgraded from Mach3 to Mach4, which has been a delight. It is light years ahead of Mach3 and you can get most any question answered on the Mach forum. For me, the upgrade was easy, but I know of some folks who have had difficulty. Mostly, it was just a matter of not following the excellent setup directions Warp9 puts out. Warp9 has a big interest in Mach4, because most of the setup is done in the Smoothstepper plugin for Mach4.

    A lot of other folks like UCCNC's electronics and control software. I've never used it, but it has a good following. There is also Acorn. It also has its followers and also detractors. I wouldn't buy it, because it is limited to 4 axes. In my book, that takes upgrading to a rotary axis off the table.

    FWIW, I am also a Vectric user. I am using the latest version of Aspire and love it. I also use Fusion 360 (CAD & CAM packaged) (free to the hobbist; a kings ransom for businesses), but its learning curve can be brutal. It's full 3d design software. I guess if I had to say which I use more, it's about 60/40 in favor of Aspire. Fusion does some things better, while Aspire does other things better. You can't beat Vetric products for V carving and sign making.

    Gary
    Hi Gary,
    When I was building my CNC Gecko was just bringing out their 251X drivers at an initial reduced price so I picked up one for X, 2 for Y, Z and A. And i called Gecko and asked them to recommend a supplier for the electronics and they recommended CandCNC, i purchased their components and made a box wired everything up. Both CandCNC and Gecko supplied very good wiring diagrams, and i was able to put everything together and it worked. I was still fuzzy on setting up the motors what with motor tuning and all, kept playing and finding out where things stall/fail at and put everything in a safe place and it has worked for me,
    Im running Mach3, VcarvePro 8.5, vectric 3D, and Vectric Photo or something like that, but only VcarvePro (VCP). If I need something out of VCP i send it to a friend that has Aspire and he helps me out, I bought a copy of BobCad with Bobart and that has been my cad program.
    I'm a big follower of CNCNUTS, he really does an excellent job of communicating in all hit YouTube site. He switched over to UCCNC a bit ago and if i switched I would seriously consider it because of him. But at 79 not sure i could go through what it takes to build another unit. already I'm frequently struggling to come up with words that are on the tip of my tongue. But i need to keep active doing things.
    The MDF poster i'm making is for the Womens Fellowship at church, saw them today (my wife is Chinese and poster is in Chinese) and explained that there was too much text (Chinese characters) in the fle and the small closed text like O and A but Chinese was flaking off because the text was too small. they agreed to remove some then i could enlarge the remaining and not have flaking, ill run the modified file in the morning and see how it works.
    Anyway, thanks for your comments, ill tell more about my machines stability in the morning,
    thanks again
    HH



  3. #43
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    U
    Quote Originally Posted by GME View Post
    HH

    Here are a few photos to illustrate the testing I described. I used some of my 80/20 to make a mock up of a CNC.

    Z cutting deeper the more time goes by ???-img_0498-gif

    Step 1. This roughly where I mounted the indicator. I zeroed it off the top of the gantry.



    Z cutting deeper the more time goes by ???-img_0500-jpg


    Step 2. Here, the mock up shows the gantry lowered by the G91 distance.



    Z cutting deeper the more time goes by ???-img_0501-jpg


    Step 3. This photo shows the gantry raised by the same G91 distance as it was lowered in Step 2.


    Z cutting deeper the more time goes by ???-img_0502-jpg

    Step 4. Note the reading in Step 3. In this example, I mocked up a reading of -.1035, which would indicate that the screw was slipping, or the stepper was losing steps.

    Repeat a total of 10 times.

    Here is the other suggestion was to mount the indicator to Z and zero it to the spoilboard. After zeroing it, it's the same steps as above, only your first move is up, and the second is back down.

    Does this make what I described a little clearer? I hope so.

    Gary

    Sorry Gary I just saw this thanks for the phots of the example, i like photos, LOL.
    OK what does G91 mean? Your taking a lot for granted here buddy. I’m thinking it means a certain distance but I need to know for sure.
    I started to run the entire file again this PM, first cut 1/4” pocket 0.010” deep but the V but just left scratches, (my paper Zeroing method not good) It then cut 7 characters perfectly in the file, then went crazy deep. Stopped the program and ran the 1/4” pocket that should be 0.010” deep (without Z zeroing) and it cut it 0.178” deep. Bummer.
    I put those same seven characters in a new file in their relative same position and ran them three times all perfect. ????

    After I get the G91 clarified I’ll run it.

    Thanks for your patience
    HH



  4. #44
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    HH. G91 is an incremental move. So, if you were to enter: G91 Z-6, Z will move downward 6 inches (assuming you are using inches instead of millimeters). G91 Z6 would move the Z axis up 6 inches.

    In Mach3, find the MDI button and click it. The screen will change. Just above the reset button, it says "Input." You enter the command on the space directly to the right of "Input." Assuming you have jogged the Z axis up and zeroed to your dial indicator, you would type in a G91 command and hit enter. As indicated above, the syntax is G91 Z(the number of inches you want Z to move). A negative Z number will move Z down the number of inches indicated; a plus Z number will move Z up the number of inches indicated.

    The number of inches depends upon how much Z movement your machine allows between the full up position and the spoilboard. For the sake of discussion, I'll assume you have 7 inches of total movement. If you have 7 inches, you wouldn't want to risk having your router hit the spoilboard, so you would pick a number that is something short of 7 inches. To allow a safety margin, I'd go with something like 6 inches movement from the top. So, in that MDI screen I would enter the following:

    G91 Z-6

    This moves Z down 6 inches. Then, I would enter the following:

    G91 Z6

    This moves Z back up 6 inches. You would note the dial indicator reading, and repeat the sequence. After 10 repetitions, you should get an idea about what is going on. The second routine I suggested would be the same routine, except you would zero the indicator at the spoilboard and start the G91 moves in the position direction, i.e., G91 Z6 followed by G91 Z-6, and read the indicator changes.

    Here's a screen shot that may help:

    Z cutting deeper the more time goes by ???-mdi-screen-jpg


    I am sorry to hear that your Z height is still screwing up. In intermittent nature of the problem is puzzling. How hot do your steppers run? If too hot to touch I suppose it could affect the magnets.

    One sensible test I've seen is to swap one of your other steppers with the one currently on Z and see if that solves the problem. If it does, you probably have a bad stepper. If it doesn't solve the problem, you've ruled out the stepper as the culprit.

    I'm back to thinking there may is some slippage going on. I'm thinking your idea to put a flat on the screw is a good one. What to do first?

    I'd start with whatever is easiest and work my way to what takes the most work.

    Gary

    Attached Thumbnails Attached Thumbnails Z cutting deeper the more time goes by ???-mdi-screen-jpg  


  5. #45
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by GME View Post
    HH. G91 is an incremental move. So, if you were to enter: G91 Z-6, Z will move downward 6 inches (assuming you are using inches instead of millimeters). G91 Z6 would move the Z axis up 6 inches.

    In Mach3, find the MDI button and click it. The screen will change. Just above the reset button, it says "Input." You enter the command on the space directly to the right of "Input." Assuming you have jogged the Z axis up and zeroed to your dial indicator, you would type in a G91 command and hit enter. As indicated above, the syntax is G91 Z(the number of inches you want Z to move). A negative Z number will move Z down the number of inches indicated; a plus Z number will move Z up the number of inches indicated.

    The number of inches depends upon how much Z movement your machine allows between the full up position and the spoilboard. For the sake of discussion, I'll assume you have 7 inches of total movement. If you have 7 inches, you wouldn't want to risk having your router hit the spoilboard, so you would pick a number that is something short of 7 inches. To allow a safety margin, I'd go with something like 6 inches movement from the top. So, in that MDI screen I would enter the following:

    G91 Z-6

    This moves Z down 6 inches. Then, I would enter the following:

    G91 Z6

    This moves Z back up 6 inches. You would note the dial indicator reading, and repeat the sequence. After 10 repetitions, you should get an idea about what is going on. The second routine I suggested would be the same routine, except you would zero the indicator at the spoilboard and start the G91 moves in the position direction, i.e., G91 Z6 followed by G91 Z-6, and read the indicator changes.

    Here's a screen shot that may help:

    Z cutting deeper the more time goes by ???-mdi-screen-jpg


    I am sorry to hear that your Z height is still screwing up. In intermittent nature of the problem is puzzling. How hot do your steppers run? If too hot to touch I suppose it could affect the magnets.

    One sensible test I've seen is to swap one of your other steppers with the one currently on Z and see if that solves the problem. If it does, you probably have a bad stepper. If it doesn't solve the problem, you've ruled out the stepper as the culprit.

    I'm back to thinking there may is some slippage going on. I'm thinking your idea to put a flat on the screw is a good one. What to do first?

    I'd start with whatever is easiest and work my way to what takes the most work.

    Gary
    I ran the test for Z axis speeds of 30, 60, and 15 IPM Ill try and attach the data.in I hope the correct way, tried to attach a .xlsx but the system said not axcceptable file type I try sending it another way. Test was. ran this afternoon before i saw tour instructions above, let me know if the method was incorrect.
    yesterday i tried to run the file again and everything was fine until the 7th character, and ut went deep again. immediately rand the 7 characters again and it ran fine.
    ill send data file now.
    thanks Gary

    well I wasnt able to send it as a private message, ill look around and see what i can do. Sorry HH

    Last edited by Mr.Chips; 03-29-2019 at 01:48 AM.


  6. #46
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by Mr.Chips View Post
    I ran the test for Z axis speeds of 30, 60, and 15 IPM Ill try and attach the data.in I hope the correct way, tried to attach a .xlsx but the system said not axcceptable file type I try sending it another way. Test was. ran this afternoon before i saw tour instructions above, let me know if the method was incorrect.
    yesterday i tried to run the file again and everything was fine until the 7th character, and ut went deep again. immediately rand the 7 characters again and it ran fine.
    ill send data file now.
    thanks Gary

    well I wasn't able to send it as a private message with the data, ill look around and see what i can do. Sorry HH
    The steppers are barely warm even after running 2+ hours. withe this testing I'm trying to catch it in the act, the tests i ran was 20 up and down at 30 IPM, 20 more of 60m and 20 more of 15IPM all readings when moved up 5" were zero (5 inches) and when the axis returned to the bottom the first 9 readings were zero, and all the other bottom readings were +0.0005" meaning 0.0005" lower that the original z zero position.
    I ran 60 cycles and did not reset any of the indicators after setting it on the first of the 60 runs.
    It looks very good but something is not right.



  7. #47
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    I’m thinking of doing something, starting with replacing the Z acme thread. Not being able to pin point this problem is driving me crazy.
    I was in a Woodcraft store the other day and the Z thread was connected directly to the stepper with a connector which held the entire weight of the Z assembly and the Z assembly was attached to the acme thread with a delrin nut.

    Under my Z stepper is a spring alignment connection, this would not support the Z assembly, so under that is a flange bearing set screwed to the acme thread securely bolted to a frame member, this supports the weight of the Z assembly.
    Am I over building, or is only a a coupling set screwed to the stepper and the acme thread enough to carry all the Z assembly. Guess I need to look at some more builds before I start making modifications.

    Thanks
    HH



  8. #48
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    I’ll try sending a photo of the data.

    Attached Thumbnails Attached Thumbnails Z cutting deeper the more time goes by ???-image-jpg  


  9. #49
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Have decided to take the first step.
    Remove the flange bearing and place a bronze oil impregnated washer (yellow) above and below the horizontal frame member, then place a steel split collar with acme internal threads (blue) above and below the washers. This would be the simplest approach and still keep the spring coupling. And see if that solves the problem. Anyway at 10 years that flange bearing and poor set screw clamping has served a goo life.

    Attached Thumbnails Attached Thumbnails Z cutting deeper the more time goes by ???-c2367921-6f6c-462b-8443-6db26f60ccd2-jpg  


  10. #50
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    HH

    Puzzling.

    I went back and took another look at the photos you posted earlier. When I looked at them when you posted them, I failed to fully understand how your Z axis was put together. More particularly, I didn't appreciate the role of the flange bearing. Not to be overly critical of the design, I believe using a flange bearing in this orientation was a mistake. Flange bearings are meant to be oriented such that the shaft is held horizontally. They were never meant to bear weight vertically, which loads the bearing from the end. Also, there are four types of flange bearings. One, like yours, relies on a set screw to hold the shaft. On a horizontal orientation, a set screw only has to keep the shaft from moving side to side, and for that purpose, it works fine. However, in your orientation, you are asking a single set screw to hold the entire weight of the axis. IMO, that is asking too much of a set screw. Two of the other three types of flange bearings have a much more positive lock on the shaft, but you still have the problem of bearing end loading. End loading is a recipe for bearing failure. Having said that, I cannot explain how some sort of bearing failure would translate to an intermittent dive of .080 inches.

    FWIW, what follows are photos of a more conventional arrangement. The aluminum block in the photos houses a conventional radial ball hearing. The main block houses the bearing, and the plate at the left side of block holds it captive. Going from right to left, the screw passes though the collar, then through thrust bearings and washers, then through the bearing block, then though the second thrust bearing set, and finally into the coupling. When installing, the aluminum bearing block screws to a fixed part of the Z axis like your flange bearing, except that the block in this case has screws going through a horizontal plane. You flange bearing mounts in a vertical plane.

    The lower half of coupling hold the weight of the Z axis. In the right photo, you can clearly see the screw that tightens the coupling to the stepper shaft. What you cannot see is another screw on the lower half of the coupling that tightens the coupling around the acme screw. The thrust bearings prevent any binding when the screw turns.
    ,

    Here is a picture of the coupling I believe you were talking about having on you machine. It's the same type I used on my previous machine. It works the same way as the coupling in the above pictures. It's not as though the spring part of the coupling is holding the axis weight. Just the lower part holds the acme screw, and the weight transfers to the aluminum block above that hold the radial bearing. In the arrangement show, there is no end loading on the bearing inside the aluminum block, and the bearing can do its job of providing smooth motion. Unfortunately, the arrangement shown would require a design change on your machine and may not solve your problem. In any event, it's a proven design, and if I were faced with the question, I would make the change. At worst cases, you would have a better setup. At best, it would solve your problem.



    To minimize the redesign. I would probably drill out the thread on the bearing block used to hold the cover in place and drill and tap new holes in the horizontal member that holds the flange block.

    I'm still inclined to believe your problem has to do with something slipping. As I mentioned earlier, you could swap your z stepper with one of your others. Run a job and see if the problem shows up again. If it doesn't, it's probably the stepper you had on Z. If it shows up with a different stepper, it's almost certainly not a stepper problem. Given that your stepper does not run on the hot side (excess heat wears them out), I doubt that the stepper is the problem.

    If you're inclined to redo your Z, the parts shown above are available from CNC Router Parts and Fineline Automation. Without going back and looking at prices, I believe they run a bit higher at CNC Router Parts (CNCRP), but their service is second to none. Fineline has the poorest service you can imagine. It's a one-man show, and the owner is completely unresponsive. You pay your money and are left to wait, and wait and wait, wondering where your parts are. More often than not, the owner will not respond to inquiries and will leave you sitting in the dark. IMO, you are much better off dealing with CNCRP. (As an aside, I looked at your profile and surmise that you live in Washington, as do I. You seem to be in the same time zone, and seeing that you were an aircraft machinist, Boeing came to mind. CNCRP is located in North Bend. I live in Gig Harbor). I would be willing to come to you, if you would like some help.

    Gary



  11. #51
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by Mr.Chips View Post
    Have decided to take the first step.
    Remove the flange bearing and place a bronze oil impregnated washer (yellow) above and below the horizontal frame member, then place a steel split collar with acme internal threads (blue) above and below the washers. This would be the simplest approach and still keep the spring coupling. And see if that solves the problem. Anyway at 10 years that flange bearing and poor set screw clamping has served a goo life.

    See my post above. I was typing it while you were posting. (Crossed in the mail, so to speak). Sounds like you are proposing something similar to what I posted.

    Gary



  12. #52
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by Mr.Chips View Post
    Have decided to take the first step.
    Remove the flange bearing and place a bronze oil impregnated washer (yellow) above and below the horizontal frame member, then place a steel split collar with acme internal threads (blue) above and below the washers. This would be the simplest approach and still keep the spring coupling. And see if that solves the problem. Anyway at 10 years that flange bearing and poor set screw clamping has served a goo life.
    I've never seen a threaded split collar, but then, it never occurred to me to look for one.

    Thrust washers vs. oilite washers??

    Gary




  13. #53
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    You checked my profile, must have been someone else. I live in Gilbert AZ a suburb of Phoenix. I retired from 3M after 30 years of service in California and Texas.

    I used those thrust ball bearings on my two Y axis ball screws. Was thinking that the Z might be too much weight for that type small bearing. Guess I got lucky using that flange bearing. Did you check out the picture I posted of the data, think it looks pretty good. Here’s a photo of the test setup. The test procedure is at the bottom of the photo of the data.

    I could use a design like yours in the red background, but instead of the coupling touching the top thrust bearing I would use a collar pressing down on the upper thrust bearing.

    Currently looking for parts.

    Thanks Gary
    HH

    Attached Thumbnails Attached Thumbnails Z cutting deeper the more time goes by ???-image-jpg  
    Last edited by Mr.Chips; 03-29-2019 at 05:48 PM. Reason: Edit


  14. #54
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by Mr.Chips View Post
    You checked my profile, must have been someone else. I live in Gilbert AZ a suburb of Phoenix. I retired from 3M after 30 years of service in California and Texas.

    I used those thrust ball bearings on my two Y axis ball screws. Was thinking that the Z might be too much weight for that type small bearing. Guess I got lucky using that flange bearing. Did you check out the picture I posted of the data, think it looks pretty good. Here’s a photo of the test setup. The test procedure is at the bottom of the photo of the data.

    I could use a design like yours in the red background, but instead of the coupling touching the top thrust bearing I would use a collar pressing down on the upper thrust bearing.

    Currently looking for parts.

    Thanks Gary
    HH

    HH

    So much for my sleuthing talents. When I checked your profile, it didn't say where you lived, it did say your were an aviation machinist. Seeing that were are probably in the same time zone, and knowing how large a presence Boeing is in this area, I put 2 and 2 together and came up with the wrong answer.

    Yes, I looked over the data. I agree. It looks very good - as good as you could ever expect it to be. I like the way your set up to measure. Very efficient.

    Thrust bearing have more strength than you might expect. I had them on my Z axis and they worked very well. I don't see any problems with using a collar on top of the thrust bearings. although it looks to me like the bottom part of the coupling does exactly the same thing as the collar. Isn't it really the same thing, except the part is larger.

    The coupler in the pictures with the red background is the one more commonly in use these days. While it isn't obvious in the picture, it is in two parts that separate at the center. Its like using two collars, one upper and one lower. That is what I am using on the Z axis of the new 4 x 4 machine I'm presently building.

    Have you made the modifications you mentioned, and if so, have you tested them? I'm anxious to know whether the mod takes care of the problem. If it doesn't, I'm left scratching my head. I doubt your stepper is the problem, but swapping the Z stepper for another should give you the answer. I doubt even more that it's your BOB. For the BOB to make the stepper suddenly run deep would require it to put out more step pulses than were programmed (think 'make it up out of whole cloth'). That seems pretty unlikely to me.

    I have been milling the ends of 3" x 3" 80/20 extrusions this afternoon. First I mill the ends each piece square. Then I determine the shortest piece, and mill each piece to length. 80/20 has a .002" per inch cut tolerance for square and .015" for length. In my experience, 80/20 is virtually always at or slightly worse than the stated tolerances. It was interesting figuring out the support structure to allow me to mill up to 68" pieces of extrusion, but I came up with something that works. It helped having a 12" machinist precision level.

    Gary




  15. #55
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    You know I went into my profile and it has aircraft machines, but I wasn’t and had no idea how it got there, couldn’t find where to change it. I retired from 3M after 30 years in quality control. No machinest background. I’m self taught, I bought a Mickey Mouse little mill from Harbor Freight, and on old 6” Craftsman lathe and used them to build my CNC. The plastic drive gears in the mill lasted 8 yeats before they gave up. So I used the CNC to cut all the parts to convert it over to direct drive. I have lusted over a Bridgeport mill but I’m cheap and the good ones are expensive. I did upgrade my Craftsman lathe, for a South Bend 9”.
    This afternoon I couldn’t get ant response from the CO that had the acme thread collars, so it will will have to wait till next week probally for them to return my call.
    Now I did swap the X axis stepper for the Z and While doing that I may have found something. The Z stepper shaft is not going very far into the spring connector, and the spring connector doesn’t have set screws that screw into the stepper shaft, they clamp around. If that connection slipped while pulling up then the axis would be lower than it should be. I think I made a mistake in making the stepper vertical supports too high. I going to disconnect the spring top side from the stepper shaft and check it again, and take the two vertical stepper supports off and mill them down to get more contact between the stepper shaft and the spring coupling. I’ll do that in the morning.
    Will let you know what I find out.
    Thanks
    HH



  16. #56
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by Mr.Chips View Post
    You know I went into my profile and it has aircraft machines, but I wasn’t and had no idea how it got there, couldn’t find where to change it. I retired from 3M after 30 years in quality control. No machinest background. I’m self taught, I bought a Mickey Mouse little mill from Harbor Freight, and on old 6” Craftsman lathe and used them to build my CNC. The plastic drive gears in the mill lasted 8 yeats before they gave up. So I used the CNC to cut all the parts to convert it over to direct drive. I have lusted over a Bridgeport mill but I’m cheap and the good ones are expensive. I did upgrade my Craftsman lathe, for a South Bend 9”.
    This afternoon I couldn’t get ant response from the CO that had the acme thread collars, so it will will have to wait till next week probally for them to return my call.
    Now I did swap the X axis stepper for the Z and While doing that I may have found something. The Z stepper shaft is not going very far into the spring connector, and the spring connector doesn’t have set screws that screw into the stepper shaft, they clamp around. If that connection slipped while pulling up then the axis would be lower than it should be. I think I made a mistake in making the stepper vertical supports too high. I going to disconnect the spring top side from the stepper shaft and check it again, and take the two vertical stepper supports off and mill them down to get more contact between the stepper shaft and the spring coupling. I’ll do that in the morning.
    Will let you know what I find out.
    Thanks
    HH


    HH

    The helical coupler you have is typical. I haven't seen one with a set screw into the shaft. As far as I've seen, they all clamp around the shaft only. As far as coupler contact with the stepper shaft is concerned, can't you just loosen the set screw on the flange bearing, put upward pressure the lower end of the Z axis, and turn the acme screw in the up direction, so that the coupler moves up and seats higher on the stepper? Easier than milling parts. That's essentially what I did when I had the slipping problem.

    Do you think that your intermittent problem could be slipping at the stepper end of the coupling? Dollars to donuts, when you build your machine, you ensured the coupling seated well up on the stepper shaft. When a couple slips, it will tend to slide downward, which is what you observed. I'm thinking you may have found problem.

    I have a bench top mill as well and am self taught. It's a Precision Matthews PM-25MV, which is a higher quality version of the Grizzly G0704. Unlike the G0704, which also has plastic gears, mine is a direct drive.

    The best thing I did for my mill was adding optical scales and DRO. I would have much preferred precision magnetic scales, but I couldn't part with the extra money. Dealing with the backlash was a major PIA. The DRO makes it a non issue, since you are always reading the actual position on the DRO. I bought my set through Amazon. You send a message to the seller telling them the scale lengths you want, and they come direct from China. A real pain to install, though. As so many of the YouTube videos tell you, you have to fabricate your own mounting brackets, and on the smaller machines, you are working in very close quarters. I'd do it again in a heartbeat, though.

    I know what you mean about wanting one of the big ones, like a Bridgeport. I doubt I would ever use one enough to justify the cost, and I really don't have the floor space. I have a good sized shop, over 1000sf, but I have a lot of large tools. I've been filling my shop for years. I have what amounts to a cabinetmaking shop. I also have a professional TIG welder, MIG welder and plasma torch.

    I wish I had a lathe, but haven't sprung for one yet. I really haven't had enough projects to push me to purchase. I'm sure a lathe must be in my future, 'if only I can find a place to put it. I want to convert my mill to CNC first, though. Sad, but the total cost of converting will end up being more than I paid for the mill.

    I'm looking forward to following your modifications and ultimately finding a solution.

    Gary




  17. #57
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Make and Model of Control: CandCNC UBOB REV 6A, Mach3 V2, Gecko Drivers 521X, VcarvePro 8.5

    You can check on previous postings for the Z axis data, I just completed going over the Z axis doing the followin.
    Originally T had tightened the two set screws on the Flange bearing collar on to the acme threads.
    I removed the Z acme screw and milled two flats so that the set screws contact squarely and solidly in t he shaft.
    I reduced the two plates supporting the Z stepper thus making the stepper shaft and the top of the Acme shaft go furtger into the spring coupling.
    While the Z assembly was apart i checked the travle of the four trucks on the rail, the travle was smooth without any roughness or binding.
    I reassembled the CNC this afternoon and again ran the same Z test looking for lost steps or any variation in readings. twenty cycles were ran at 60 IPM and there were five instances where the readings were +0.0005' the remaining readings were all zero. Only a slight improvement of data taken before repairs.

    After I ran the test I loaded the original file and ran it again, and it repeated the exact problem that I had before I did the repairs. When the V bit went deep into the MDF i stopped the program and the V bit was Im estimating, 0.175'" into the material, however the Z readout on the screen showed -0.0203", something is not right.
    Prior to it failing it cut six characters perfectly. I stopped the program in the middle of the seventh character and the line number was 5920N5920, I looked back through the code to where it started cutting the 7th character and that line number was 5920N5920. I extracted that range of code and there wasnt any Z axis moves, but a portion of the character was properly cut. Don't understand why i didn't see any Z commands.

    I really thing this is not a mechanical problem, but something when i created the file and generated the code. I really would like someone to evaluate the gcode and see if there is something amiss there.

    A little background on the file development. The Characters started out as printed text approximately 3/4" tall. they were then scanned at a high resolution so as to prevent pixilation, creating a JPG file. The JPG is imported into VcarvePro and converted to vectors. Then converted to curves. They are then arranged and size adjusted to the best fit. The file is then ran showing the finish cut. Everything looked normal.

    I have ran dozens of files like this in the past but suddenly Im experiencing the V bit cutting deep into the material.

    Today after the file failed again I then created a small file of text using the text editor in VcarvePro and it ran perfectly without any hint of the V bit going deep.

    If you would like the entire file or the range of code where the problem occurred, give me you email and ill send it to you.



  18. #58
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Here's the photo of the problem file. The blue circled section was cut before optimizing the Z axis and the red circled area is after, exactly the same problem, which I think is in the gcode.
    the "Now is the time" text was cut 10 minutes after the red section was cut, no problems.
    Im creating the JPG file from scratch, this time without "curve fit" from Vcarve.

    Attached Thumbnails Attached Thumbnails Z cutting deeper the more time goes by ???-jpg  


  19. #59
    Member
    Join Date
    Apr 2016
    Location
    United States
    Posts
    841
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by Mr.Chips View Post
    Make and Model of Control: CandCNC UBOB REV 6A, Mach3 V2, Gecko Drivers 521X, VcarvePro 8.5

    You can check on previous postings for the Z axis data, I just completed going over the Z axis doing the followin.
    Originally T had tightened the two set screws on the Flange bearing collar on to the acme threads.
    I removed the Z acme screw and milled two flats so that the set screws contact squarely and solidly in t he shaft.
    I reduced the two plates supporting the Z stepper thus making the stepper shaft and the top of the Acme shaft go furtger into the spring coupling.
    While the Z assembly was apart i checked the travle of the four trucks on the rail, the travle was smooth without any roughness or binding.
    I reassembled the CNC this afternoon and again ran the same Z test looking for lost steps or any variation in readings. twenty cycles were ran at 60 IPM and there were five instances where the readings were +0.0005' the remaining readings were all zero. Only a slight improvement of data taken before repairs.

    After I ran the test I loaded the original file and ran it again, and it repeated the exact problem that I had before I did the repairs. When the V bit went deep into the MDF i stopped the program and the V bit was Im estimating, 0.175'" into the material, however the Z readout on the screen showed -0.0203", something is not right.
    Prior to it failing it cut six characters perfectly. I stopped the program in the middle of the seventh character and the line number was 5920N5920, I looked back through the code to where it started cutting the 7th character and that line number was 5920N5920. I extracted that range of code and there wasnt any Z axis moves, but a portion of the character was properly cut. Don't understand why i didn't see any Z commands.

    I really thing this is not a mechanical problem, but something when i created the file and generated the code. I really would like someone to evaluate the gcode and see if there is something amiss there.

    A little background on the file development. The Characters started out as printed text approximately 3/4" tall. they were then scanned at a high resolution so as to prevent pixilation, creating a JPG file. The JPG is imported into VcarvePro and converted to vectors. Then converted to curves. They are then arranged and size adjusted to the best fit. The file is then ran showing the finish cut. Everything looked normal.

    I have ran dozens of files like this in the past but suddenly Im experiencing the V bit cutting deep into the material.

    Today after the file failed again I then created a small file of text using the text editor in VcarvePro and it ran perfectly without any hint of the V bit going deep.

    If you would like the entire file or the range of code where the problem occurred, give me you email and ill send it to you.


    HH

    I would agree that it could be something in the G-Code, but for your earlier report. You said it went deep on the 7th character (same as the most recent attempt). You then immediately reran it, and it cut okay. If the problem was with the G-Code, I would expect it to go wrong at the same spot every time. No harm in evaluating the code, but if there is something wrong in it, I would at a complete loss to explain why it would fail to cut the 7th character wrong one time, but would successfully cut on the next attempt.

    Sorry, but I'm not the one to evaluate your file. I know the basic stuff to manually move the machine around, but I've not taken the time to learn the code well enough to evaluate code and diagnose problems in it.

    Gary




  20. #60
    Gold Member Mr.Chips's Avatar
    Join Date
    Sep 2003
    Location
    Gilbert Az
    Posts
    1469
    Downloads
    0
    Uploads
    0

    Default Re: Z cutting deeper the more time goes by ???

    Quote Originally Posted by GME View Post
    HH

    I would agree that it could be something in the G-Code, but for your earlier report. You said it went deep on the 7th character (same as the most recent attempt). You then immediately reran it, and it cut okay. If the problem was with the G-Code, I would expect it to go wrong at the same spot every time. No harm in evaluating the code, but if there is something wrong in it, I would at a complete loss to explain why it would fail to cut the 7th character wrong one time, but would successfully cut on the next attempt.

    Sorry, but I'm not the one to evaluate your file. I know the basic stuff to manually move the machine around, but I've not taken the time to learn the code well enough to evaluate code and diagnose problems in it.

    Gary
    When I said I ran it again, was kind of misleading.
    I opened the file and copied a portion of the vectors/text where it went bad plus a little more then pasted it into a new file, then regenerating the Vcarve and converting that new portion and ran it. There’s probally nothing wrong with the artwork/vector file.

    Would you suggest I start a new forum posting stating that I have a problem that is associated with the code generated and would like someone to take a look and see if the code looks OK?

    Thanks Gary for all you hard work and suggestions I appreciate it.

    HH



Page 3 of 6 FirstFirst 123456 LastLast

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

Z cutting deeper the more time goes by ???

Z cutting deeper the more time goes by ???