CNC MACHINE TECH needed in Northern California - Page 2
Close
Login to Your Account
Page 2 of 2 FirstFirst 12
Results 21 to 38 of 38
  1. #21
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    3,992
    Post Thanks / Like
    Likes (Given)
    750
    Likes (Received)
    2093

    Default

    Quote Originally Posted by [email protected] View Post
    ....At this point I think I may have been suckered in to thinking it's the X because it's loud (and feels gravely like a bad bearing when turning the screw by hand). With an indicator and a large bore, I can see the typical axis reversal dip, but as big as .0005", maybe .0006", but then it looks like the axis over corrects, then oscillates a couple times until it settles in to a nice round arc again. Seems more electronic to me than mechanical, but that's why I wanted a pro to take a peek.....
    Right after reversal, when the axis is moving slowly on the arc, oscillation can be caused by a bad screw/nut assembly or thrust bearings. Usually gravely sounding bearing rotate with a "lumpy" feel. When the servo motor encounters a tight spot, more current is needed to overcome that. Then once past the tight spot the motor may overshoot a tiny bit. Once the motor is turning faster this phenomenon is less noticeable or eliminated

    My position is you never try to tune an axis or otherwise fiddle around with drives, parameters, etc, if there is any doubt about the mechanical condition of the axis. If the mechanical side checks out perfect and you still have poor axis behavior then it is time to dig in.

  2. Likes Milland, machtool, eaglemike liked this post
  3. #22
    Join Date
    Sep 2008
    Country
    UNITED STATES
    State/Province
    California
    Posts
    1,423
    Post Thanks / Like
    Likes (Given)
    887
    Likes (Received)
    803

    Default

    Quote Originally Posted by [email protected] View Post

    Got .0003" max with a larg-ish screwdriver. Returned to zero within .0001" both ways.
    Got .0001" max both ways, so .0002" total movement. Seems mechanically sound to me?
    Those values aren't excessive. I would redo the backlash compensation, and if I had time I'd replace the thrust bearings. Would appear to be .0002" backlash in the thrust bearings, and .0001" in the leadscrew. (Wish my Fadals were that good. )

    Did you check the Y-axis to see if there's any play in the leadscrew/thrust bearings?

    If you circular interpolate a hole do you get any steps at 90/120/180/360 positions? If you do your backlash comp is off. If the circle is smooth at 90/120/180/360 then your comps are good.

    There's probably a procedure in the Haas manual for properly adjusting backlash. For instance Fadal has a short program that moves the table, and then returns it to the original position, any deviation would be removed in the backlash compensation. Fadal (and I'm sure it applies to other MC makers) specifically says don't use the jog wheel to determine backlash comp.



    Good luck!
    Last edited by triumph406; 04-13-2019 at 10:42 AM.

  4. #23
    Join Date
    Sep 2008
    Country
    UNITED STATES
    State/Province
    California
    Posts
    1,423
    Post Thanks / Like
    Likes (Given)
    887
    Likes (Received)
    803

    Default

    If your backlash comps are good, and holes are .0007" out of round, could you have play in the linear rails? Putting an indicator on the main casting, and the needle against the part of the Y-axis that moves you could determine whether you had any play in the Y-axis rails, and putting an indicator on the cross slide and the needle against the table you could see if there's play in the X-axis rails.

  5. #24
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Triumph, I didn't get a chance to check for linear rail slop. The tech ball barred it first thing and sure enough it matched exactly what I was seeing on the vision system, which is nice to see. The plot was egg shaped and the tech says, "that's twist". Fuck me, the thing wasn't leveled correctly.

    Between leveling and then replacing X thrust bearings it sounds like it's new again, (EDIT) has less than .00005" backlash and .00015" reversal spikes (/EDIT), but still has .0004" out of round in both ball bar and an aluminum test piece. Tech worked on backlash and screw linear comp settings for an hour and that's as good as it got. I'm thinking the Y is not great either, but for now this will have to do.

    Also asked the tech about the dreaded maincon board "upgrade". He said I was lucky in that I had the 15" LCD (2007 machine), so all the "upgrade" is for this machine is the new maincon board and some new wiring to the pendant. Figure $10k for that. And, he said he's collecting a list of parameters and fixes for those parameters because the maincon "upgrade" rarely goes well, and can take another 2 days to get the machine running well again.

    Great job, Haas.

  6. #25
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Final ball bar result, ran at 80 IPM.

    untitled.jpg

  7. #26
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Something is still wrong. Renishaw uses #558 (X) & #559 (Y) for "stylus offset" to compensate for runout at the stylus during calibration. I indicate a gage ring to within .0001", set G59, then run XY calibration. All goes well except #558 = -.0007" and #559 = +.00035 even though the probe has .0001" max runout every time I've checked it.

    I then probed gage ring, indicate, move away, probe, indicate, move away, etc. and get excellent results, repeats to within .0002" as it should.

    I then set #558 & #559 to zero. Repeat above test. Now X moves -.0007" and Y moves +.0003" every time. OK, this makes sense and all numbers match, but still not sure why it's comping X and Y when there is almost no runout, but whatever. Cool.

    ------------------------------------------

    Now I move to machining a bore, then probe the bore and update the offset, then machine again, probe, machine, etc. I restore #558 = -.0007" and #559 = +.00035, which is what worked well in previous test. Now results have X moving -.0007" and Y moving +.0006 every time I repeat the test.

    Set #558 & #559 to zero. Repeat above test. Now X repeats dead on and Y moves +.0002" every time. Exactly opposite the results I got with the gage ring.

    does-not-compute.jpg

  8. #27
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    3,992
    Post Thanks / Like
    Likes (Given)
    750
    Likes (Received)
    2093

    Default

    Quote Originally Posted by [email protected] View Post
    Something is still wrong.......
    The probe trigger point is not equal in all directions. Particularly with mechanical probes and less so with strain gauge probes. Probes use a 3 point kinematic design yet you are working with 4 points.

    Are the gage lengths of the indicator, probe, and boring bar all roughly equal? Head droop can cause some confusing testing if the lengths are not pretty close.

  9. Likes dstryr liked this post
  10. #28
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Quote Originally Posted by Vancbiker View Post
    Are the gage lengths of the indicator, probe, and boring bar all roughly equal? Head droop can cause some confusing testing if the lengths are not pretty close.
    Eyeballed gage lengths:
    Endmill used to interpolate bore: 3.25"
    Probe: 5.25"
    Indicator: 8.25"

    I'll see if I can get them more similar and see if test results change. Thanks Vanc!

    And when the Selway tech was here, he leveled the machine and then swept for tram. X is good, Y is .0025" out over 12". I had him leave it alone because I figured that wasn't much on a 2" facemill, but the spindle is drooped for sure and will have to be shimmed to correct. Not sure how that happens over time.

  11. #29
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    3,992
    Post Thanks / Like
    Likes (Given)
    750
    Likes (Received)
    2093

    Default

    If the column is square and the tram is out that far then you have pretty serious head droop. With a 2" change in gage length that tram error will account for .0004" of position error alone.

  12. Likes DavidScott liked this post
  13. #30
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    How very interesting: Evened up the gage lengths as best I could to:

    Endmill used to interpolate bore: 5.06"
    Probe: 5.25"
    Indicator: 4.8"

    #558 = -.0003" after probe X/Y re-calibration (probably my fault, DTI setup this short was tough to use)
    #559 = 0.

    Machine, probe, machine, probe test redone. Now X moves -.0003" every cycle and Y doesn't move at all. What'dya know. So with zero runout, those variables really should be zero, and I obviously need to shim the Z! Shit!

    Thanks again Vanc. I'm impressed you figured head droop in all the numbers I posted.

  14. Likes dstryr liked this post
  15. #31
    Join Date
    Jan 2013
    Location
    Gilbert, AZ
    Posts
    5,586
    Post Thanks / Like
    Likes (Given)
    6819
    Likes (Received)
    6968

    Default

    Quote Originally Posted by [email protected] View Post
    The other guy, who I hired as an "advanced operator" is getting it slowly, but only after scrapping the entirety of a $12,000 job, and we only found it after plating. I walked away from it feeling like he was comfortable, and he was. Sizes of features were all fine, however he somehow failed to blow out the soft jaws well enough (or at all??), leaving all kinds of chips for the parts to get dented on, to the point that locations were mostly out of tolerance because the parts weren't seating. It was a mess. I know that's all on me, but god damnit I never dreamed I would have to tell an "advanced operator" to 1) blow the chips out of the jaws, and 2) look at your parts and if there are so many dents from chips that the damn things look like the surface of the moon, stop!
    Is this dudes name Tommy?!

    I hope things turn around for you soon Matt. I got it kinda bad right now. But, I can't even imagine what you are going through!

    I wish I was local. You could come run shit on my Brother. It is just sitting there.
    I'm supposed to be starting light production back up end of May. But, the whole Hydro/ransomware deal pushed the material out who knows how long?

    EDIT: and of course, I feel you on the shitty haas parts/service situation. The main attractive attributes of haas ownership are gonzo!

  16. #32
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Quote Originally Posted by Vancbiker View Post
    If the column is square and the tram is out that far then you have pretty serious head droop. With a 2" change in gage length that tram error will account for .0004" of position error alone.
    The tech didn't check column squareness, only tram, so I don't know if it's in the head or the column. Good point. I need to get somebody back here and check / fix that stuff. I don't have the knowledge or equipment to do it.

  17. #33
    Join Date
    Mar 2016
    Country
    UNITED STATES
    State/Province
    California
    Posts
    52
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    23

    Default

    Did he run the ballbar on X/Z and Y/Z?

    That will easily show any head nod.

  18. #34
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Quote Originally Posted by readymix View Post
    Did he run the ballbar on X/Z and Y/Z?

    That will easily show any head nod.
    No. He zeroed in on the X because it was noisy (and because the X was my initial complaint) and really didn't look at anything else.

    I've got a call in to both service places suggested. CMS is "swamped" but is checking in to more local places.

  19. #35
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    3,992
    Post Thanks / Like
    Likes (Given)
    750
    Likes (Received)
    2093

    Default

    Quote Originally Posted by [email protected] View Post
    .....Thanks again Vanc. I'm impressed you figured head droop in all the numbers I posted.
    After 40+ years around CNC machines a few things have stuck .

  20. #36
    Join Date
    Mar 2016
    Country
    UNITED STATES
    State/Province
    California
    Posts
    52
    Post Thanks / Like
    Likes (Given)
    2
    Likes (Received)
    23

    Default

    Quote Originally Posted by [email protected] View Post
    No. He zeroed in on the X because it was noisy (and because the X was my initial complaint) and really didn't look at anything else.

    I've got a call in to both service places suggested. CMS is "swamped" but is checking in to more local places.
    I have never ran only XY. It takes maybe 5 more minutes to run XZ and YZ once you have it set up.

    Hopefully you can get someone out to check soon. If I was closer I'd show up today.

  21. Likes [email protected] liked this post
  22. #37
    Join Date
    Jan 2014
    Country
    UNITED STATES
    State/Province
    Washington
    Posts
    3,992
    Post Thanks / Like
    Likes (Given)
    750
    Likes (Received)
    2093

    Default

    Quote Originally Posted by [email protected] View Post
    ......And when the Selway tech was here, he leveled the machine and then swept for tram.......
    Quote Originally Posted by Vancbiker View Post
    If the column is square and the tram is out that far then you have pretty serious head droop......
    Quote Originally Posted by [email protected]
    The tech didn't check column squareness, only tram
    This has been bugging me a bit. I don't know what the bed and leveling screw arrangement on a Haas is like so I may all wet...

    How does a tech figure he has "leveled" a VMC that has more than 4 leveling screws if he did not check the column for squareness?

    On a brand new smallish machine I can see getting away with just checking the tram if one is OK assuming that the spindle to Z axis geometry is good. I have seen many used VMC installs where the column is leaned forward due to too much weight on the screws at the back of the column. The installer had rationalized that since there is more weight in the column those screws should have more weight on them. The bad tram is then rationalized as just being wear in the Z axis or result of a crash or something.

    IMHO, your "tech" did not do a proper leveling on your machine. Possibly there were constraints either on his side or yours that prevented that, I don't know.

    A reasonably good granite square is not expensive and a really handy tool....

    For a small machine....

    Precise Precision Granite Square 10" x 6" x 1" Grade 00 - 303-916 - Penn Tool Co., Inc

    For medium machines.....

    Precise Precision Granite Square 15" x 10" x 1.5" - 303-917 - Penn Tool Co., Inc

  23. #38
    Join Date
    May 2004
    Country
    UNITED STATES
    State/Province
    California
    Posts
    2,493
    Post Thanks / Like
    Likes (Given)
    885
    Likes (Received)
    1611

    Default

    Awful picky of you Vanc, he's just a senior tech at Selway, geez...

    I just got off the phone and Selway will be here as soon as they can (don't know when that is yet). I'm having them just start from square 1 and go through the damn thing like a new install. I called the other techs suggested and every one of them said Selway was my best bet, time wise.

    I just want a pro to fix the thing so we can get back to work, but it's obvious I need to get much more involved with PM'ing these things concerning level and squareness. One big problem is I have no idea how to go about that efficiently when both machines have subplates on them that cover 90% of the tables.

  24. Likes wheelieking71 liked this post

Tags for this Thread

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
  •