Help with Standard 4x2 Accuracy


Results 1 to 16 of 16

Thread: Help with Standard 4x2 Accuracy

  1. #1
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Help with Standard 4x2 Accuracy

    Hi everyone,

    I have been running a CNCRP Standard 4x2 machine for about a year and a half and have been thoroughly impressed. I am running it with a Hitachi M12VC, ethernet smoothstepper, MX4660 driver with NEMA 23 motors. As I understand it, this is the "light" setup and it is a beast.

    I use this for a variety of purposes, from cabinet panels to guitars and the odd prototyping job that comes along. I haven't been overly concerned with the accuracy because nothing I was cutting required tight tolerances up until a couple of weeks ago.

    I have been cutting parts from baltic birch and one critical cut is a slot that is 0.48" wide. I've been roughing that with a climb cut at 0.3" DOC, and then cleaning it up with a conventional cut at full depth to get it to the correct width. I've consistently gotten a cut that measures 0.46-0.47" which to be honest is acceptable for what these parts are being used for, but I'd like to figure out why I am not getting closer to 0.48". These slots are generally cut so that the 0.48" dimension is measured along the Y axis. When the slot is positioned on the machine where the dimension is measured more along the X axis, I get even smaller measurements (closer to 0.44-0.45"). Again, totally acceptable *for what these parts are used for*.

    Just the other day I got a request to submit a bid to a company who is requiring tolerance of +/-1mm (which I am already within) but I want to get the machine as accurate as possible. CNCRP claims a tolerance of +/- 0.005" and I am sitting at 0.01 - 0.02" so I know there can be some improvement. And if I want to do more business with this company down the road, I need to improve the tolerance.

    The first thing I've checked is router runout. Without a collet installed, I get runout of approximately 0.003" inside the shaft of the router. I wouldn't say this is great, but it's acceptable and likely isn't causing the problem. I figure if anything, the runout would be causing my slot to be oversized rather than undersized, due to the router bit cutting away 0.003" more of the material.

    I'm not exactly sure what to look at next, so I guess this is what my question really is: where should I begin to look at mechanically if there were something off with the machine? I don't think it is a steps per issue in Mach3, but is that a possibility as well? I currently have it set to 1736.237 as per recommended by CNCRP.

    I have tried to email CNCRP but haven't heard back (been about a week) but it could be I went about it the wrong way. Under their contact page, I only had the option to fill out a form and no one responded to that.

    Thanks for taking a look and reading through my post.

    Similar Threads:


  2. #2
    Member ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    35538
    Downloads
    1
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    If you tell the machine to move 24", does it move exactly 24"?

    Gerry

    UCCNC 2017 Screenset
    [URL]http://www.thecncwoodworker.com/2017.html[/URL]

    Mach3 2010 Screenset
    [URL]http://www.thecncwoodworker.com/2010.html[/URL]

    JointCAM - CNC Dovetails & Box Joints
    [URL]http://www.g-forcecnc.com/jointcam.html[/URL]

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)


  3. #3
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    Quote Originally Posted by ger21 View Post
    If you tell the machine to move 24", does it move exactly 24"?
    If I remember correctly, yes...but I'm measuring with a measuring tape so it isn't exactly the most accurate method. I will go out and check in a few minutes.

    I forgot to mention, the other thing I checked is motion with the dial indicator. I set the indicator to 0, tell the machine to move 1", and then back again. When I do this I don't get the dial indicator back to 0...I get something like 0.008" off or something. I was kind of rushed when I checked so I will do it again.

    EDIT: I just tested 24" of travel on both the X and Y axes. Both seem to be dead on 24"...I had a harder time measuring the Y axis due to the design of the machine and my approach to measuring, so the Y axis could be not *quite* dead on.

    Again, I did this with a tape measure so I can't say how precise that is. I did this by removing my soft limits, and moving the X and Y axes step by step right up to the black rubber bumpers. I put a flashlight behind the bumpers and the bearing carriages so I could see exactly when the light disappeared. Then I measured from the rubber bumper to the bearing carriage on both axes. I also got out my dial indicator again and I suspect I either suck at reading them/I'm using it incorrectly, or it is busted. It is telling me I'm off by 0.01" when I tell the machine to move 1", but when I check it with my digital calipers, I get a reading of 0.996". 0.004" is obviously beyond the 0.005" tolerance figure CNCRP gives, so that's good. My methods may not be the best either but I think they are sufficient. If they aren't, please let me know the correct way to make these measurements.

    Last edited by canman77; 02-17-2020 at 08:17 PM.


  4. #4
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    Just wanted to keep a log here for those who may have similar problems later.

    Ahren contacted me from Avid CNC and is helping me troubleshoot. I knew they would eventually contact me, sometimes these things just slip through the cracks. No worries there at all. He recommended checking for straightness in the rails as well as making sure the bearings are all snug. I found that several bearings on my Y axis rails were loose, so I tightened them up to the rails but did not find this to be the solution. I also did not find any curve or bend in my rails, so we are still looking into the issue.

    For what it's worth, I am consistently off by 0.01" in the Y axis on movements at least up to 5-6", as that is how big my digital calipers are. Whether I move 0.48" or 4" I come up short 0.01". I am not sure how to test this on a larger scale without bigger digital measuring equipment. Perhaps a worthy investment.

    Also worth noting, when I measure movement on the X axis with digital calipers, I am well within the 0.005" tolerance.



  5. #5
    Member ger21's Avatar
    Join Date
    Mar 2003
    Location
    Shelby Township
    Posts
    35538
    Downloads
    1
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    A caliper that measures in .005 increments is really only accurate to .01", unless you have a caliper that measures in .001" increments.

    Chasing .01" can be very difficult, especially without an accurate way to measure large distances.

    Try cutting two squares, one climb cut, and one conventional. Measure the difference. This is how much your machine is flexing, and it's very difficult to try to "dial out" flex.

    Gerry

    UCCNC 2017 Screenset
    [URL]http://www.thecncwoodworker.com/2017.html[/URL]

    Mach3 2010 Screenset
    [URL]http://www.thecncwoodworker.com/2010.html[/URL]

    JointCAM - CNC Dovetails & Box Joints
    [URL]http://www.g-forcecnc.com/jointcam.html[/URL]

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)


  6. #6
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    OK, I just cut 4 test squares. 2 20" squares and 2 5.5" squares. Naturally, the 20" squares were measured with a tape measure, and any measurement short/long of 20" was estimated with calipers. The 5.5" squares were measured with calipers. And for reference, my calipers do measure in increments of 0.001" so I think they are a fairly good measuring tool for this job.

    So, the results...I have listed the Y axis measurement first, since this was the one I was most curious about:

    20" Conventional:
    Y: 19.97" (appx. 0.03" short)
    X: 20"

    20" Climb:
    Y: 20.02" (aapx. 0.02" long)
    X: 20.03" (appx. 0.03" long)

    5.5" Conventional:
    Y: 5.472"
    X: 5.488"

    5.5" Climb:
    Y: 5.506"
    X: 5.523"

    It appears that the difference in Y for the 20" square (climb vs conventional) is 0.05" and the difference in X is 0.03". The difference in Y for the 5.5" square is 0.034" and the difference in X is 0.035". Can I interpret these numbers to mean that I have approximately 0.03" of flex in the machine? I am definitely not getting cuts that are out by that much, since I'm often using stock to leave with climb cuts and then cleaning them up with a full depth conventional pass.

    Is that 0.03" figure about right for these machines or am I doing something wrong?

    Ahren replied to me last night and suggested my pinions are worn out and need to be replaced, as they are causing backlash. I checked for backlash and I am only getting roughly 0.002" of backlash on either axis so I am not sure if the pinions are actually worn.




  7. #7
    Member
    Join Date
    Aug 2016
    Location
    United States
    Posts
    122
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    I have Standard 4x2 as well. I am using a Bosch 1617. I haven't checked mine very closely in a while, but I recently upgraded to Mach 4 and did a quick check. My results were better than that. I think about +/- 0.010" is typical for my setup. When I was setting up my machine, I looked at several videos on Google about the subject. A lot of them were focused on steps and such. I got the impression from the Avid info that there wasn't much to gain there, but it might be worth looking at. I tend to move the machine pretty slow compared to what I read other folks do, Not sure that's a factor. I built a holder for a dial indicator and checked using that originally. The results were pretty good. I'm thinking less than the 0.010" I was shooting for. I'm not really sure what to look at. I have had dimensional issues when I failed to secure he work piece real well. Any bad bearings? I had a cracked bearing that would cause funny stuff once. I tend to follow the rule of thumb for depth of cut. Oh, and I added a precision collet - great investment. Wish I had more ideas for you. I sure hope you get this figured out. I know it is frustrating. My main thought for responding was that I have a similar setup and without any real fussing get better results, so I would think yours would too. Let us know what you figure out. Oh, one last thought, I have read about people reporting missing steps and such due to electronic interference (via dust collector, etc.). Material and tooling can play a part as well. Is the dust clearing out of the cut?



  8. #8
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    Quote Originally Posted by Goonk View Post
    I have Standard 4x2 as well. I am using a Bosch 1617. I haven't checked mine very closely in a while, but I recently upgraded to Mach 4 and did a quick check. My results were better than that. I think about +/- 0.010" is typical for my setup. When I was setting up my machine, I looked at several videos on Google about the subject. A lot of them were focused on steps and such. I got the impression from the Avid info that there wasn't much to gain there, but it might be worth looking at. I tend to move the machine pretty slow compared to what I read other folks do, Not sure that's a factor. I built a holder for a dial indicator and checked using that originally. The results were pretty good. I'm thinking less than the 0.010" I was shooting for. I'm not really sure what to look at. I have had dimensional issues when I failed to secure he work piece real well. Any bad bearings? I had a cracked bearing that would cause funny stuff once. I tend to follow the rule of thumb for depth of cut. Oh, and I added a precision collet - great investment. Wish I had more ideas for you. I sure hope you get this figured out. I know it is frustrating. My main thought for responding was that I have a similar setup and without any real fussing get better results, so I would think yours would too. Let us know what you figure out. Oh, one last thought, I have read about people reporting missing steps and such due to electronic interference (via dust collector, etc.). Material and tooling can play a part as well. Is the dust clearing out of the cut?
    As I understand it, cutting conventional parts will generally come out a little smaller, and climb cutting will produce parts that are a little bigger, so I'm not overly concerned with the approximate 0.02" variations in my 5.5" squares. Do you get results that are dead on when you both climb and conventional cut test squares?

    When I climb cut and leave a bit of stock, and then conventional cut, my results are usually pretty dead on, aside from the slot issues mentioned above where I routinely am about 0.01" off. And if you are +/- 0.01" regularly then maybe I shouldn't feel too bad about not getting down to the 0.005" tolerance, but I'd still like to get to that point.

    I did have one bearing break and that caused all sorts of crazy problems, but they went away once I replaced the bearing. I don't see any other cracked bearings but I did tighten up some bearings that weren't really making contact with the steel rails, which did not seem to make a difference. Ahren suggested backlash in his email and recommended new pinions, but when testing for backlash I am within 0.002" which I'd consider good. And I might be missing something but I don't really see how 0.002" of backlash could cause a measurement to be out by 0.01"?

    I haven't tried slowing the machine down, maybe that is worth a try today. What speed do you generally run cuts at? On non-critical parts (cabinet panels, for example, where 0.02" tolerance would be acceptable) I cut 350ipm, 0.25in DOC. On guitar parts I am usually cutting 100-150ipm at 0.125" DOC. My rapids are 800ipm. Perhaps I'll slow my rapids down and test again with calipers to see what happens. My last test in the MDI section of Mach3 was done with the rapids.

    As far as dust goes, I do run dust collection and the dust does get removed from the cuts. It's actually a pretty clean operation so I don't think I can blame dust in the cuts. I did look at precision collets and router runout as well because I thought that might be the problem with my Hitachi M12VC (known for not great runout) and after thinking about it again, I suppose it could be a contributing factor. I will have to test runout again today, I forget what the exact numbers were.



  9. #9
    Member
    Join Date
    Aug 2016
    Location
    United States
    Posts
    122
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    Wow! I'm a big chicken I guess, but I rarely go over 100 ipm and 0.125 depth of cut. I think my rapids are at 100. Like you, I'm kinda miffed at what/where your problem could be. I didn't try climb vs conventional cutting. I use Vcarve and I think it is set to climb by default, so I rarely change it. One test I saw was where a square was attached to the spoil board on one end of the machine and a dial indicator was zeroed there. Then, the machine was moved all the way to the other end and back to see if the indicator was still reading zero. I do know that my circles are never perfect, but darn close. I played with the step count stuff, but I don't think it helped much. Is your gantry square to the opposite axis? Does the same cut measure the same on both "sides" of the axes? I mean cut a long grove on Y=0 then do the same thing on y=24 and check them. I'm thinking out loud here! Does the machine act the same cutting vs not? Is the tooling good? I dunno, maybe some folks with more experience an knowledge will jump in.

    The cracked bearing thing really threw me for a while. I kept hearing a pop, but it was not consistent. I casually mentioned it in an email layout something else to Avid and they said that was probably a bearing..and, of course, it was. Mine was on the Z carriage thing.

    I've routed circuit boards on mine with very tight spaces and it did real well. No issues. Those were around 2" square probably. I rarely use the last 24" of mine. Sometimes, I regret that I got the longer one, but then when I need it, its there. I have a very small shop and this thing has a big footprint.

    I haven't milled much recently I upgraded to Mach 4 and have spent a lot of time learning to program it. I have several buttons I programmed to do handy things for the way I work. Nothing fancy, but the new setup took me a while to learn. Everything seems to be working now.



  10. #10
    Member
    Join Date
    Feb 2005
    Location
    United States
    Posts
    829
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    I ran a Standard 4896 and was normally getting accuracy to within .003-4, sometimes less. I would routinely cut at 200+IPM .500 in deep in plywood, several projects in aluminum came out in the same order of accuracy as well.

    The standard machines must have all bearings in contact with the rails, whenever I found my accuracy suffering it was one of two problems.

    1. either a broken bearing, happened 2 times.
    2. or some bearings were not in contact with the rails, frequent.

    I was constantly fighting the y-axis carriages, and ussually the inside right hand carriage would not be contacting the inside of the rail. no idea and would frequently adjust to keep it in contact.



  11. #11
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    Wow! I'm a big chicken I guess, but I rarely go over 100 ipm and 0.125 depth of cut. I think my rapids are at 100. Like you, I'm kinda miffed at what/where your problem could be. I didn't try climb vs conventional cutting. I use Vcarve and I think it is set to climb by default, so I rarely change it. One test I saw was where a square was attached to the spoil board on one end of the machine and a dial indicator was zeroed there. Then, the machine was moved all the way to the other end and back to see if the indicator was still reading zero. I do know that my circles are never perfect, but darn close. I played with the step count stuff, but I don't think it helped much. Is your gantry square to the opposite axis? Does the same cut measure the same on both "sides" of the axes? I mean cut a long grove on Y=0 then do the same thing on y=24 and check them. I'm thinking out loud here! Does the machine act the same cutting vs not? Is the tooling good? I dunno, maybe some folks with more experience an knowledge will jump in.

    The cracked bearing thing really threw me for a while. I kept hearing a pop, but it was not consistent. I casually mentioned it in an email layout something else to Avid and they said that was probably a bearing..and, of course, it was. Mine was on the Z carriage thing.

    I've routed circuit boards on mine with very tight spaces and it did real well. No issues. Those were around 2" square probably. I rarely use the last 24" of mine. Sometimes, I regret that I got the longer one, but then when I need it, its there. I have a very small shop and this thing has a big footprint.

    I haven't milled much recently I upgraded to Mach 4 and have spent a lot of time learning to program it. I have several buttons I programmed to do handy things for the way I work. Nothing fancy, but the new setup took me a while to learn. Everything seems to be working now.
    I got the machine for cutting cabinet panels (I do the occasional cabinet job here and there and can produce far superior cabinets with CNC than with table saw!) so power and speed were the big selling points for me. This machine has more power than I'd ever need and I haven't had any issues running 350ipm+ in birch plywood. Granted, I don't go much deeper than 0.3" but it still hauls.

    Anyways, I haven't checked that both sides of the machine cut the same distance (ie, everything is square) but I can measure some cabinet parts I just cut to see. I haven't noticed anything looking off as far as square shapes go, but I will still look at it and see. I think nlancaster's post may explain the source of my problem though..

    I ran a Standard 4896 and was normally getting accuracy to within .003-4, sometimes less. I would routinely cut at 200+IPM .500 in deep in plywood, several projects in aluminum came out in the same order of accuracy as well.

    The standard machines must have all bearings in contact with the rails, whenever I found my accuracy suffering it was one of two problems.

    1. either a broken bearing, happened 2 times.
    2. or some bearings were not in contact with the rails, frequent.

    I was constantly fighting the y-axis carriages, and ussually the inside right hand carriage would not be contacting the inside of the rail. no idea and would frequently adjust to keep it in contact.
    It appears I have the same two problems...the first problem was one broken bearing, which was VERY obvious and a quick fix. The cuts were significantly out when the bearing broke and movement was not smooth at all.

    I definitely have the second issue as well...I was trying to adjust the bearings on my Y axis and the right and left hand sides seem to keep pulling each other out of whack. I loosened the steel riser bolts to adjust the bearings and once I tightened the bolts, things would shift. I wonder if I need to break down the gantry and try doing a bit of a rebuild. How parallel were your steel rails?

    Seems like the bearings are the weak link of the machine...things have to be pretty spot on in order to get the 0.005" tolerance. And to be honest, that's not a huge weak link. Without getting super crazy with the build and making sure everything is absolutely perfect, I am getting 0.01" accuracy which is a HUGE improvement over my last machine. I had no name drivers that were so out of whack and I could never get the steps per revolution right. Scaling was a nightmare, but this machine is killer. Chasing down the 0.005" tolerance is mostly a pride thing for me now more than anything!



  12. #12
    Member
    Join Date
    Feb 2005
    Location
    United States
    Posts
    829
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    canman, I spent a large amount of time during assembly to build the machine as square and parrellel as possible.

    My process to square the rails to the gantry, and get all bearings in contact with rails.

    1. assemble carriage and riasers.
    2. attach gantry to both risers.
    3. losen all carriages and use clamps to keep risers from seperating from carriages.
    4. square gantry to table rails.
    5. tighten outside carriages to the rails
    6. tighten inside carriages to the rails

    do all of the above while maintaining gantry squareness to table rails.

    I used bar clamps to pull the carriages into the rails.

    This should "should" result in all bearings in contact with rails.

    To set steps per unit, I used a tape measure and my best vgroove cutter as a pointer.

    1. align tape to axis being measured, X or Y axis.
    2. position pointer as close to axis minimum as possible and align tape mark to pointer.
    3. use mach3 steps calibration wizard to move axis to as near maximum travel as possible.
    4. use pointer and tape measure to verify distance traveled and input into calibration wizard.
    5. repeat until you get as near perfect travel as possible.

    The above steps calibration will get you about as good as you possibly can measure then distance. Because you are using a long move, in my case 48x96 inches, that gives the longest baseline to measure steps against actual movement.



  13. #13
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    Thanks for all that information, I appreciate it. I am planning on spending some quality time squaring up the rails this weekend. I was hoping to have time during the week but work has kept me too busy. I spent a lot of time making sure the base was square when I first assembled the machine but somehow I must have missed a little bit when I installed the rails. Luckily that shouldn't be an ordeal to get squared away...sounds like I can loosen all the bolts and manipulate everything until it is perfect.

    As far as steps per unit, I think I have those pretty well set. I am using the numbers given on CNCRP's website and I think I want to dial in the machine before I start messing with steps per unit. It seems more likely that my rails got a little out of square when building the machine. When I get the rails squared I will post here with results.

    Out of curiosity, what number did you end up with for steps per unit? I wonder how close it is to the number given by CNCRP.



  14. #14
    Member GigaWatt's Avatar
    Join Date
    Apr 2015
    Location
    United States
    Posts
    130
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    I had width problems on a slot cut and found my 0.25" end mill measured 0.2485".



  15. #15
    Member
    Join Date
    Dec 2015
    Posts
    140
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    That's something I should look at as well...I haven't really measured my endmills. I'll do that this week and see what I find. I haven't had a chance to re-align everything on the machine, this was a hectic weekend. The second half of the week and this weekend should be a lot more free so I am hoping to make progress. Stay tuned. Thanks for all the suggestions and advice.



  16. #16
    Member
    Join Date
    Jan 2014
    Posts
    29
    Downloads
    0
    Uploads
    0

    Default Re: Help with Standard 4x2 Accuracy

    To attain the precision that you are shooting for you can throw away any carpentry type measuring tools.
    I have great accuracy on my machine but carefully check everything with machinist measuring tools.
    You can even buy a cheap digital DRO Slide and mount it to your table temporarily and measure a longer distance. This was something I did early in my build with an 18" DRO Slide designed for machine tools. Cost less than $100.
    You will never get any precision on your machine using a tape measure to check anything.



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

Help with Standard 4x2 Accuracy

Help with Standard 4x2 Accuracy