584,842 active members*
4,621 visitors online*
Register for free
Login
Page 84 of 107 3474828384858694
Results 1,661 to 1,680 of 2130
  1. #1661
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    I don't know if I'm being dumb but I can't figure out how to reference the axes in mach3 without actually touching the switches?

    I'd like to be able to leave the machine zeroed on a fixture, note down the g53 coords, jog back to g53 x/y/z zero on startup and reference without actually touching the switches so they don't throw off the X & Y position. I've tried clearing the ref all button script but then it doesn't do anything. Anyone know how to do it?


    Sent from my iPhone using Tapatalk

  2. #1662
    Join Date
    Jun 2010
    Posts
    4252

    Re: OmioCNC report

    Nomenclature, that's what it is.
    You have switches - OK. But are they LIMIT switches or HOME switches?
    If they are Limit switches and NOT Home switches I think you can zero the system by clicking on the Ref All Home button, left of the DROs.
    Is this what you want?

    Cheers
    Roger

  3. #1663
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Quote Originally Posted by RCaffin View Post
    Nomenclature, that's what it is.
    You have switches - OK. But are they LIMIT switches or HOME switches?
    If they are Limit switches and NOT Home switches I think you can zero the system by clicking on the Ref All Home button, left of the DROs.
    Is this what you want?

    Cheers
    Roger
    Sorry, they're home switches. No matter what I remove from the Ref All Home button script, it automatically bounces off the switches, unless I delete the whole script in which case it does bugger all (doesn't zero the machine)


    Sent from my iPhone using Tapatalk

  4. #1664
    Join Date
    May 2015
    Posts
    1422

    Re: OmioCNC report

    It should be hitting the home switch, setting machine zero on that axis, THEN backing off.

    Have a look at your machine vs workspace coordinates on page 1 of Mach3, you should see it zero. Unplug an axis, move everything, plug it back in and re-home you should see it re-zero.

  5. #1665
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Quote Originally Posted by dharmic View Post
    It should be hitting the home switch, setting machine zero on that axis, THEN backing off.

    Have a look at your machine vs workspace coordinates on page 1 of Mach3, you should see it zero. Unplug an axis, move everything, plug it back in and re-home you should see it re-zero.
    Yeah, that's what it does. But the switches have a lot of potential for error and the fixture uses dowels so the holes need to line up properly, which is why I want to home manually by jogging to g53 x0/y0 and not using the switches.

    Just had an idea though, maybe unplugging the switches will work? Or jog it to 'home' position then unplug the X and Y axis, press ref all and manually trigger the switches with me hand. There should be a less cumbersome way though?


    Sent from my iPhone using Tapatalk

  6. #1666
    Join Date
    May 2015
    Posts
    1422

    Re: OmioCNC report

    Two options:
    1. Swap out the microswitches for hall effect sensors or opto switches, much better repeatability etc.
    2. What I do - hit ref all just to get rid of the error code and make things green, then ignore whatever the hell that was and go and register properly against dowel or dowel hole or whatever.

  7. #1667
    Join Date
    Sep 2016
    Posts
    506

    OmioCNC report

    Quote Originally Posted by dharmic View Post
    Two options:
    1. Swap out the microswitches for hall effect sensors or opto switches, much better repeatability etc.
    2. What I do - hit ref all just to get rid of the error code and make things green, then ignore whatever the hell that was and go and register properly against dowel or dowel hole or whatever.
    I already do #2, but the tool is the same for the first op and the last op, so it would be convenient to just g53 g0 x* y* z* in the MDI / ref all / load code and go upon startup, rather than messing around taking the tool out, chucking up the DTI, locate zero, tool back in, set Z... etc.

    Not the end of the world I suppose.

    In happier news, the OMIO and everything I've bought for it will have paid itself off after a few runs of this particular (and always repeating) job, which I never expected it to end up doing, so that's a bonus.


    Sent from my iPhone using Tapatalk

  8. #1668
    Join Date
    Jun 2010
    Posts
    4252

    Re: OmioCNC report

    Ah, I see.
    Perhaps your problem is that your Home switches are not sufficiently repeatable? This can happen so easily.

    In fact, finding Home switches (at an acceptable price) which can match the potential accuracy of the machine is very difficult. This is why I do not use Home switches at all. I ref the machine to the workpiece instead, to within 5 microns, and go from there. Yes, that takes time, but it WORKS.

    Note: you can ref to a base plate of course. I often do that, and it then is stable for cycles of production.

    Cheers
    Roger

  9. #1669
    Join Date
    Sep 2009
    Posts
    1856

    Re: OmioCNC report

    The diy hall effect switches work realy well and only cost a few dollars each to make,
    http://danielscnc.webs.com/

    being disabled is not a hindrance it gives you attitude
    [SIGPIC][/SIGPIC]

  10. #1670
    Join Date
    Jun 2010
    Posts
    4252

    Re: OmioCNC report

    Got any idea what their long term accuracy and repeatability is? Better check when milling steel too.

    I have considered optical switches with thin fibres. When the whole fibre diameter is 0.25 microns, analog level detection should be good for a very small movement. The problem is of course dust.

    Cheers
    Roger

  11. #1671
    Join Date
    Sep 2009
    Posts
    1856

    Re: OmioCNC report

    There's a thread in the zone with all that in it and yes of course metal would be a problem
    http://danielscnc.webs.com/

    being disabled is not a hindrance it gives you attitude
    [SIGPIC][/SIGPIC]

  12. #1672
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Yeah, these mechanical home switches aren't the best. I've measured them quite a bit and the smallest error I've recorded is 0.04mm, which isn't too bad, but the worst is somewhere around 0.32mm, and the error isn't repeatable, it's any random amount between those numbers.

    I might try rig up some better switches but if I can find a way to reference without touching switches at all that would be easier.

    I have optical switches on my lathe, they repeat absolutely bang on to 0.00mm every single time. The motors are all encoded, but I don't think they make a difference because their position resets to 0.0 whenever you power on the machine no matter the position of the axis, so they wouldn't know the position of the switch until you've referenced off it. On the lathe though the switches are hidden behind the way covers and out the way of chips and dust, not so easy on a router.


    Sent from my iPhone using Tapatalk

  13. #1673
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Has anyone noticed big differences in repeatability over the range of travel? This is the first time I've used the full bed for something, and my Y axis is traveling further than it did the other day for the same movement. This is a board of cast acetal (how flaming much!!! £££) which I drilled and faced on Friday and just came to finish off with chamfers and tap the holes today, pics show the spindle is lining up with the holes at the 'home' (Y0) side perfectly but at the other end (Y- side) its traveling past the centre of the hole, you can see how the chamfer has been cut deeper on the Y- side of the hole than the other. This error graduates as the machine moves further in Y-.

    Double checked code, zero position, nothing untoward, so it's got to be the machine that is moving differently?

    Pics (the one with the gold bolt is at the home position, other is at the far end of the bed)




    Sent from my iPhone using Tapatalk

  14. #1674
    Join Date
    Apr 2018
    Posts
    34

    Re: OmioCNC report

    I had the same thing happen to me along the y axis.

    I think it's because the machine bed is made of aluminum.

    But the Y axis ballscrew is made of steel. The aluminum frame/bed/table expands at a much greater rate than the steel ballscrew, hence the positioning "stretch".

    I let a part run all day for a 9 hour roughing run.

    Then the next morning I came in and it was out 0.15mm - about 0.005".

    The shop temp had dropped from 20C at the time it was done roughing ...down to 4C the next morning when I ran the finishing cycle.

    I forgot that these machines don't have thermal expansion coefficients applied.

    I left the parts on the machine and continued running them at the end of the day when the temp was back up and everything matched up fine.

  15. #1675
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Quote Originally Posted by CaistorAl View Post
    I had the same thing happen to me along the y axis.

    I think it's because the machine bed is made of aluminum.

    But the Y axis ballscrew is made of steel. The aluminum frame/bed/table expands at a much greater rate than the steel ballscrew, hence the positioning "stretch".

    I let a part run all day for a 9 hour roughing run.

    Then the next morning I came in and it was out 0.15mm - about 0.005".

    The shop temp had dropped from 20C at the time it was done roughing ...down to 4C the next morning when I ran the finishing cycle.

    I forgot that these machines don't have thermal expansion coefficients applied.

    I left the parts on the machine and continued running them at the end of the day when the temp was back up and everything matched up fine.
    I did think about that but I'm not sure that's the case here. The conditions inside and out are almost identical to the day I did the first ops, and this has moved about 0.5mm which would seem quite a lot for thermal shrinkage/expansion. That chamfer is 0.75mm and you can see it's barely kissed one side of the hole at the dodgy end.


    Sent from my iPhone using Tapatalk

  16. #1676
    Join Date
    Apr 2018
    Posts
    34

    Re: OmioCNC report

    One thing I DID do when I got my machine was to slow down the homing speed.

    Mine would home at something like 3000 mm/min. Way to fast.

    Our cnc's here just crawl along to home position.

    I slowed mine down to 500mm/min. Yes it's slow, but I take a second to park it

    50mm away from home before I shut it down.

    Maybe worth a try?

    I'd also suggest checking with a dial indicator for any end play in the y axis bearings, although personally I haven't encountered that problem on mine.

    Curious to see how it all plays out...

  17. #1677
    Join Date
    Apr 2018
    Posts
    34

    Re: OmioCNC report

    oh, and something else to check...

    is it possible your y axis lost a few steps?

    Mine did, and I had to get into the acceleration settings.

    Linear acceleration was factory 600mm/sec2
    Curve acceleration was factory 600mm/secs

    I brought down the curve acceleration to 70.7% to 420mm/sec2 and it stopped losing y steps.

    Same reason you have to slow down in a curve.

    Maybe worth looking into...

  18. #1678
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Losing steps - I don't think so. It's overshot, so it would have to have gained steps? Also the motors make a racket when they lose steps don't they?

    Y bearing/screw/motor play, good shout. Checked by hand and it feels solid at each point, around 0.5mm I would expect I would feel click back and forth? Can be checked with a dial indicator tomorrow (don't have it here with me).

    Home switches, yes they are **** and not to be trusted. I have a reference on the board (a reamed hole which I spin a DTI around in to find centre) which I use to zero G54 and that is in position to within 0.005mm.

    It's worth noting the X axis is completely unaffected and is on point at each end of its travel.

    I've done stuff like this before (albeit much smaller) where I've had stuff on the table across different days and different set ups and never had an issue with position or alignment, but something's not happy.

    Or perhaps it's stress in the acetal from machining it and caused it to move? I spent a lot more on cast stuff to hopefully eliminate this from being an issue but I don't understand enough about that sort of stuff to rule it in or out.




    Sent from my iPhone using Tapatalk

  19. #1679
    Join Date
    Apr 2018
    Posts
    34

    Re: OmioCNC report

    mine didn't make any noise when it lost steps...and the part was only 150mm long by 40mm wide...lost 0.5mm each depth of cut on the y.

    good work on the G54 datum hole. never miss with that one. I check mine on my fixture everytime I fire it up...

    I wonder if maybe the machine lost a few steps when you did the holes last week and then, after shut down...went to the correct position...

    Food for thought here....I have been playing with different lubes for the ballscrews, and came across something that works fantastic.

    amazon carries it. 400g tub of Superlube synthetic grease with PTFE teflon. mine likes it and it quieted right down on the linear bearings.

    my point being that if its lubed with good stuff then the motors don't have to work as hard during rapid moves, that's where mine lost a few steps..

  20. #1680
    Join Date
    Sep 2016
    Posts
    506

    Re: OmioCNC report

    Hmm... don't think it can be lost steps, here's a photo of the drilling toolpath, if steps were lost there would be a point along there where each hole was off by the same amount regardless of its actual position along the Y axis. The error increases gradually the further the machine travels in Y-.



    Nice idea with the lube. I had just been giving the rails and screws a regular spray with silicone lube, purely because they're all exposed and the grease was a magnet for dust and chips, but that stuff looks a much better option.


    Sent from my iPhone using Tapatalk

Page 84 of 107 3474828384858694

Similar Threads

  1. Report files
    By RP Designs in forum Centroid CNC Control Products
    Replies: 7
    Last Post: 11-22-2009, 06:51 AM
  2. Machining Report
    By Tielegin in forum Esprit
    Replies: 0
    Last Post: 05-05-2009, 06:25 PM
  3. Fresh man report
    By ice in forum Commercial CNC Wood Routers
    Replies: 0
    Last Post: 09-28-2006, 11:45 AM

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
  •