Page 1 of 2 12 LastLast
Results 1 to 15 of 19
  1. #1
    Join Date
    Apr 2009
    Location
    Kingswood
    Posts
    930

    Default CNC Mill Striations

    I reported on my first test piece on a newly commissioned CNC earlier today.

    Job went Ok, but the final finish off a new end mill shows interesting striations.

    Piece was Cast Iron, 10 mm thick and about 44 x 51 mm, spindle at about 750 RPM sourced from a VFD set to 25 Hz, feed at 200 mm/min.
    First pass was a Ripper 12 mm cutter, 0.4 mm depth and various width 0.5 to 10 mm.
    Second pass was a 12 mm new end mill, 0.4 mm depth and 0.5 mm width, to produce final size of 42 x 49 mm.
    The photos show the general appearance of the piece, and a close view of the edge striations.

    The striations are present on all 4 sides,
    The 'feel' is good, my fingernail cannot feel the striations.

    Note that the striations are constant across numerous passes, 0.4 mm cut depth to 10 mm is about 21 passes.
    So, this not a random effect, it has got to be linked to something repetitive.
    Obvious possible sources are the VFD, the mill steppers and the VFD.

    The VFD/motor will have harmonics, but there is a belt drive and so unlikely to be synchronized with the cutter for such long path lengths.

    The ripper cutter left corrugations along the piece, not striations, so print-through of that seems unlikely.

    This leaves the steppers.
    The code would be absolutely repeatable at every path point around the piece, so the steppers will also be absolutely repeatable.
    There is a belt, but that is toothed, quite stiff and under constant tension.
    Later tonight I will try to calculate the stepper step rate to see if that coincides with the observed striations.

    Can anyone else report similar striations ?
    And any corrective actions ?
    John
    First Test Piece Striations compr.JPGFirst Test Piece Final compr.JPG

  2. #2
    Join Date
    Sep 2012
    Location
    York, North Yorkshire UK
    Posts
    6,439

    Default

    Hi John,

    Looking at your pictures, I would say that the feed rate is too fast 8" inches a minute is high, try half that.
    Best Regards:
    Baron J.

  3. #3
    Join Date
    Jun 2016
    Location
    Sydney
    Posts
    283

    Default Chatter?

    The 1st pic looks like chatter to me. The 2nd pic is just weird.
    I would want to check for any play or flex in the table or the spindle.

    Cheers
    Roger

  4. #4
    Join Date
    Dec 2014
    Location
    melbourne
    Posts
    72

    Default

    Looks to me like your end mill has slight run out.
    If you know your feed rate and spindle speed and can measure the pitch of your lines it should work out to be one line per rev.

    Robert

  5. #5
    Join Date
    Mar 2009
    Location
    Melbourne
    Age
    54
    Posts
    825

    Default

    Hi John,
    When you run your tool path are you climb or conventional milling?
    My understanding, and experience with a little 6040 CNC router, is climb cutting tool paths run much smoother and give better finishes with CNC.
    200mm/min in CI does sound aggressive unless you have the spindle power and rigidity to support it.
    Cheers,
    Greg.

  6. #6
    Join Date
    Apr 2012
    Location
    Healesville
    Posts
    2,129

    Default

    Hi John, what I think......
    The lines that you are getting are likely a result of climb milling, if you mill in a conventional manner and finish with .03 to .05mm DOC you will nearly get a mirror finish.
    But in saying that I have zilch CNC experience, sure in theory with zero backlash you should be able to climb mill but practice always beats theory so you might have to have a play around to find what the mill likes and is capable of.
    I nearly always get the longitudinal shinny lines when side cutting with an end mill, I have never investigated the cause and have thought the side flutes on the end mill must have a slight difference in sharpness along the flutes.
    Thanks for this post I like to see other peoples cutting results.
    cheers, shed

  7. #7
    Join Date
    Apr 2009
    Location
    Kingswood
    Posts
    930

    Default

    The cut is conventional, not climbing.
    End mill runout sounds possible, but it is not likely that the rotational position of the cutter would repeat exactly for 25 passes around the job.

    Checked the VFD setting, 24 Hz on a 1500 RPM 1.5 kW motor with pulleys set for 950 RPM, so about 470 RPM.

    It didn't look fast and I had a tool-maker and expert machinist (both retired) observing the cut.
    The swarf looked entirely normal.
    A 4-flute tool at 470 RPM has chips at 4x470 = 1880 chip/min.
    At 200 mm/min each chip is 200/1880 mm = 0.11 mm.
    My 'Big Green Book' (TAFE text) lists CI at 0.15 to 0.23 mm per chip, so the feed is not excessive.

    The mill is a Bridgeport clone, a light industrial machine with adequate power/rigidity for this test.
    Chatter would be a random effect without any patterning, and inevitably detectable by feel.
    These striations repeat exactly over multiple cuts and cannot be felt.

    Compared the striations with a thread pitch gauge, they are 0.4 mm pitch.
    The ballscrew is 10 mm pitch so the noted striations are 25/pitch.
    The steppers are 200 steps/rev with 10 microsteps and 16/32 toothed pulley reduction, giving 400 microsteps per mm.
    The noted striations would be repeating every 160 microsteps.

    The numbers do your head in.

    There are obvious vibrations in the drive pulleys.
    These may be dependent on mechanical resonances and the software set-up, so possibly correctable.

    Setting up a CI cylindrical test of comparable diameter with the same conditions.
    It will be interesting to see if the striations appear as the steppers would be at constant speed only for very short periods.

    My initial head-scratching exercise is how to most effectively use an electronic edge-finder to determine the centre of a circular object.
    Thankfully the ball tip is resiliently mounted (already had a couple of hard contacts).

    John

  8. #8
    Join Date
    Sep 2009
    Location
    Newcastle
    Posts
    341

    Default

    Why are you cutting 0.4mm depth of cut (DOC) and 0.5mm width of cut (WOC)?

    Generally with CNC a big DOC with smaller WOC is used. This spreads tool wear amongst other benefits.

    You could try some different feed rates and see what happens to the pattern.

    Gibs tight?

  9. #9
    Join Date
    Aug 2008
    Location
    Melbourne
    Age
    34
    Posts
    1,075

    Default

    First thing I'd do is try reducing the chip load; either drop your feed or bump up your RPM, my feed/speed app (FSWizard, it's free for Android and iOS and is VERY good) says you should be around 6-700 RPM and 0.07-0.08 mm/tooth feedrate for a 4-flute HSS endmill in cast iron.

    If you're referring to Table 2 on page 345 of the green book for your feeds, just note that those numbers are for carbide; if you're using HSS tooling, 1/3 to 1/2 is probably a good place to work from.

  10. #10
    Join Date
    Apr 2009
    Location
    Kingswood
    Posts
    930

    Default

    pippin88,
    I agree that the DOC and feed could be varied to improve/change the finish.
    I have an admission, it appears the first test was a piece of MS, not CI.
    I did some cuts on genuine CI today, and the swarf is quite different.

    Remember, this first test was just that, the 'first' cuts on a newly configured CNC machine.
    The objective was to try and measure the backlash.

    elanjacobs,
    My values were taken from Table 1 on the same page, it is labelled 'High-speed steel'.

    Yesterday I set up a CI work piece (pulley) with the aim of making a circular job of 36 mm.
    The objective was to see if the striations show on a circle.
    Job went well, but I wanted a little more depth, so:
    - manual input of another 0.4 DOC with code including G91,
    - then restart the program at the last sequence and got the cutter spiralling into the job and nut.
    Post-analysis revealed G90 in the code had been overwritten by the G91 !
    A good lesson learned.

    Quite a divot out of job/nut before the cutter broke !
    The only new 12 mm end mill I had (damnit).

    Today set up another CI cylinder and a new 12 mm Carbide 6-flute end mill.
    Cut the circle 75 mm diameter with the same conditions of speed, DOC and feed.
    Inspection shows the same striations at the quadrant points, ie, only one stepper in operation as the backlash comes into play.
    Where both steppers are moving, the striations get lost in the background, but it did appear that the pitch of the striations was not varying.

    Of great interest is the pronounced finish changes as the Y-cutter swaps directions.
    On the pull the striations are obliterated and the job has grooves, on the push the striations are in the background of a typical scuffed surface.

    Also noticeable is the lack of roundness at the quadrant points.
    I don't have a 75 mm micrometer, but the vernier does reflect the backlash effects.
    Hoping that some backlash compensation will cure most of these sorts of things on real jobs.

    I will now try some linear moves on CI at aggressive speeds, DOC and feed, more like what is needed to get jobs done.
    Along the way I am going to vary the microstepping rate on the steppers to see what effect that might give, it is currently set at 10.
    The gibs had been set-up, but now after some running it will be worth checking.

    As an aside, cutting with CNC does not provide the same satisfaction as twiddling manual knobs.

    John

  11. #11
    Join Date
    Aug 2008
    Location
    Melbourne
    Age
    34
    Posts
    1,075

    Default

    Quote Originally Posted by electrosteam View Post
    elanjacobs,
    My values were taken from Table 1 on the same page, it is labelled 'High-speed steel'.
    I saw that one, but I figured it was referring to arbor mounted cutters for horizontal milling due to the lack of a specific endmill column as they have in the other table.

    EDIT: Just checked the Guhring book and their coated carbide endmills have a lower recommended feed than the numbers you were working to. I'm very sure that Table 1 is not what you need here.

  12. #12
    Join Date
    Apr 2009
    Location
    Kingswood
    Posts
    930

    Default

    Other than at TAFE, I have only ever operated relatively small manual machines by the feel.
    This new CNC monster will need a much more rigorous approach.

    Just did a cut on some CI:
    - 6-flute solid carbide end mill,
    - 1600 RPM,
    - cut 18 mm axial x 0.5 mm radial,
    - feed 500 mm/min.

    Sweet as you like with long curled tubes of swarf.
    A regular faint pattern, that I will examine in the morning.

    Guys, if I repeat the same cut, what speed and feed might be appropriate ?

    I have discovered a real problem with the mill, the motor pulleys are higher than the spindle pulleys.
    Looks like a motor removal and a pulley pulling exercise.

    John

  13. #13
    Join Date
    Aug 2008
    Location
    Melbourne
    Age
    34
    Posts
    1,075

    Default

    https://fswizard.com/

    Plug in all your parameters and it will spit out feed/speed

  14. #14
    Join Date
    Jan 2004
    Location
    Mackay North Qld
    Posts
    6,446

    Default

    Hi,
    Something I remember from years back. When the VRD control lathes were in their infancy I recall lots having this trouble with machining vibration. One of the causes then was the harmonics in the different types of belts.

    I have no idea of whether it applies at all as I know little of CNC mills.

    A good mate whose advice on motor vehicles was always sound, used to say, "Look for the simple things first".

    I can't tell you how many times his mantra bore fruit.

    Just a thought then. would the belt tension have an effect. I recall one of the lathe fixes was to change to segmented belts which seemed to help in damping the belt harmonics.

    I hope it helps

    Grahame

  15. #15
    Join Date
    Apr 2009
    Location
    Kingswood
    Posts
    930

    Default

    Grahame, thanks for input.

    Checked the Wizard, it says 10,000 RPM at 1.9 mm cut over the 18 mm length !!!
    My mill will go to 5000 RPM, but the belt misalignment makes me reluctant.

    Did a cut on the CI piece:
    - Dia 12 mm, 6-flute carbide end mill,
    - 2500 RPM,
    - conventional milling,
    - 1mm DOC over 18 mm,
    - 500 mm/min.

    Sounded furious, swarf coming off like rockets.

    Diagonal grooves downwards in direction of cut.
    Could be flute contact moving down as tool progresses along job.

    Repeat above with:
    - DOC 0.2 mm,
    - climb milling.

    Dust cloud from the cutter.
    Now I have vertical striations at 0.8 mm pitch (feels smooth).

    As Graham reminded me, check the simple things first.

    John

Page 1 of 2 12 LastLast

Similar Threads

  1. Hercus o mill Vertical mill head lubrication
    By china eyes in forum THE HERCUS AREA
    Replies: 0
    Last Post: 14th Oct 2018, 06:57 AM
  2. Selecting a small Mill / Mill/Drill
    By Wombat200 in forum METALWORK GENERAL
    Replies: 31
    Last Post: 15th Mar 2012, 10:58 PM
  3. Arboga EM825 mill vs Hercus model 0 mill?
    By neksmerj in forum METALWORK GENERAL
    Replies: 23
    Last Post: 14th May 2011, 03:16 PM
  4. Face Mill V End Mill
    By eskimo in forum METALWORK GENERAL
    Replies: 8
    Last Post: 31st Aug 2010, 03:19 AM
  5. No. 0 Mill Serial Number and Mill at Work Photos
    By Anorak Bob in forum THE HERCUS AREA
    Replies: 1
    Last Post: 30th Dec 2008, 03:31 AM

Posting Permissions

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