Build Thread Rebuilding my 5x10 CNC Router - Page 2


Page 2 of 2 FirstFirst 12
Results 21 to 39 of 39

Thread: Rebuilding my 5x10 CNC Router

  1. #21
    Member
    Join Date
    Sep 2005
    Location
    Indonesia
    Posts
    1098
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Here is a video of the plasma running. I dont have a touch off setup so I just zero the z at the top of the tubing.
    You do not need THC also for this cut ? Does your plasma cutter have pilot arc ?



  2. #22
    Member
    Join Date
    Apr 2013
    Location
    United States
    Posts
    30
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    @asuratman if you ever get the chance to operate a plasma CNC, you will soon learn that THC aka Torch Height Controller is generally needed only for thinner or corrugated metal where the surface can wrap or is already irregular. Most metals (1/4” and thicker) can normally do without THC. Also, if you’re Torch Voltage isn’t calibrated correctly, THC can be a big time PITA...causing unexpected nose dives leading to wastage of material or worst yet, breaking your Z axis.



  3. #23
    Member
    Join Date
    Sep 2005
    Location
    Indonesia
    Posts
    1098
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router



    I cut hole on 1/4" H beam with plasma with no good result. I thought I have no THC. I will do some excercises cutting with plasma so I can make good frame like Scott has and faster... big frame and rigid ...



  4. #24
    Member
    Join Date
    Jan 2013
    Location
    Germany
    Posts
    75
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    I don't like it when I see someone build such a flimsy machine! You really need to rethink how rigid this machine won't be when you are slamming the gantry back and forth at about 1,000,000 inches a minute. It might handle 500,000 inches a minute but now way in hell will it be able to hit 1,000,000 inches minute. Just sayin!!!!!!!



  5. #25
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Quote Originally Posted by Black Forest View Post
    I don't like it when I see someone build such a flimsy machine! You really need to rethink how rigid this machine won't be when you are slamming the gantry back and forth at about 1,000,000 inches a minute. It might handle 500,000 inches a minute but now way in hell will it be able to hit 1,000,000 inches minute. Just sayin!!!!!!!
    This cracked me up! Sorry for the lack of updates here. Hoping to make some progress soon. I will keep the thread updated when I can.

    Thanks


    Sent from my SM-G950U using Tapatalk



  6. #26
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Final some progress on the cnc rebuild. Lots of great life events taking priority over working on the cnc.

    Moving parts from the old machine to the new machine. I used the thermwood ballscrew supports which I think are going to work great. I have 1 per side but have the notches in the tubing to add 1 more per side if needed.

    I have decided to lube the bearing blocks and the ballnuts with air/lube to each unit. It's similar to what thermwood and onsrud do on some of their machines. I have always found grease difficult to keep clean, in fact I have the slightest spalling on the far ballscrew. It was harder to get to on the old machine so I didnt clean/lube it as much as I should have. I think the ballscrew has plenty of life left but does have the slightest ticking noise when it runs.

    I had new ballnut support blocks mounted. I got both the x axis screws in place and did some quick PID tuning the x axis runs great it was awesome to see it moving under it's own power. I will take some video of things running. The closest ballscrew will have to come back out in order to install home switch and limit switches. But I really wanted to see it run so I went ahead and mounted it.

    Got the new Y to Z plate cut and mounted. Waiting on some bolts and a new timing belt for the Y then will get that up and running.

    Alright homing and limit switches. On the old machine I had prox switches for homing and lever style limit switches. For this machine I'm struggling to find a good spot to mount the lever switches in X. I'm thinking of combining X-- and X home into one prox switch. X++ would be another prox. Thoughts on this? Any issues sharing a switch?



    Sent from my SM-G950U using Tapatalk



  7. #27
    Member Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    4047
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Looking good !

    Prox switches: Depends on your controls. It is possible to use the same switch with some control systems. Nothing wrong with doing it that way.

    Jim Dawson
    Sandy, Oregon, USA


  8. #28
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Got the X axis limit and homing prox switches installed these will sense the ballnut carrier. Also made the ballscrew protection for X. This should do nicely to keep dust/chips off and also allow plywood and other materials to be slide across.

    Rebuilding my 5x10 CNC Router-20191218_064426-jpg
    Rebuilding my 5x10 CNC Router-20191218_132604-jpg
    Rebuilding my 5x10 CNC Router-20191218_132609-jpg
    Rebuilding my 5x10 CNC Router-20191218_132614-jpg

    Attached Thumbnails Attached Thumbnails Rebuilding my 5x10 CNC Router-20191218_064410-jpg   Rebuilding my 5x10 CNC Router-20191218_064426-jpg   Rebuilding my 5x10 CNC Router-20191218_132604-jpg   Rebuilding my 5x10 CNC Router-20191218_132609-jpg  

    Rebuilding my 5x10 CNC Router-20191218_132614-jpg  


  9. #29
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Got the lube distribution system installed. I built a tank out of 4x6 box tubing and some angle iron. Drilled and tapped for 16 4mm tubing outlets and one inlet. The goal of the large chamber is to keep the oil suspended in air. I have not tried it out yet but my plan is to check for oil at each bearing by letting the air blow at the paper and confirming oil. I may need to clean up the tube routing once the rest of the Z axis is installed.

    The lube will turn on when the system is enabled. I have initially set the air pressure at 8psi and will adjust where needed. The other regulator is for the Z axis counterbalance. I pushed the air tank into the gantry beam for now but will need to come up with some foam to hold it as it hangs around in there.

    Sent from my SM-G950U using Tapatalk



  10. #30
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Alright I gave the lude system a try, I get air but no oil. I upped the drips of lube coming out of the oiler. Still no oil just air. I only checked 4 lines but I was hopeful the distribution tank would allow oil/air to go to all the tubes. The main line going into the tank has oil in it when I unhooked it. So oil is going somewhere but not to all tubes.

    Anyone out there have experience with air/oil lube systems?

    Some specs of what i have
    Regulator set to 8psi (I tried going up in pressure but not down)
    Mobile spindle oil in the oiler
    Currently at 1 drop per second

    Any help is appreciated

    Sent from my SM-G950U using Tapatalk



  11. #31
    Member Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    4047
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    I would think that you would need to atomize the oil in the air stream, create a fog micron sized droplets in the distribution box that hopefully will remain suspended in the air. That implies high velocity air through a small nozzle. I think at 8 PSI there is not enough velocity possible because you can't get enough differential pressure.

    It might be possible to do this using standard oil burner nozzles, might be worth a try. Maybe put an oil burner nozzle in the main line to the distribution box or build it into the box itself.

    https://www.amazon.com/s?k=Oil+Burne...ref=nb_sb_noss

    https://www.amazon.com/Nrpfell-Burne...ustrial&sr=1-8

    Jim Dawson
    Sandy, Oregon, USA


  12. #32
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Quote Originally Posted by Jim Dawson View Post
    I would think that you would need to atomize the oil in the air stream, create a fog micron sized droplets in the distribution box that hopefully will remain suspended in the air.

    https://www.amazon.com/s?k=Oil+Burne...ref=nb_sb_noss
    Jim this is exactly what I was thinking needs to happen also, I was thinking this is what the Wilkerson oiler would do. The 8psi was a starting point based on info I could find on thermwood routers but there isn't much information out there. For the oil burner nozzle, that would take a small pump for the oil?

    For testing does holding the tube in front of paper make sense? I would think if there is oil in the air it should show on a white sheet of paper. When I pulled the 3/8 line off tank/manifold there was decent amount of oil in the air.

    Sent from my SM-G950U using Tapatalk



  13. #33
    Member Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    4047
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Quote Originally Posted by ScottDamman View Post
    Jim this is exactly what u was thinking needs to happen also, I was thinking this is what the Wilkerson oiler would do. The 8psi was a starting point based on info I could find on thermwood routers but there isn't much information out there. For the oil burner nozzle, that would take a small pump for the oil?

    For testing does holding the tube in front of paper make sense? I would think if there is oil in the air it should show on a white sheet of paper. When I pulled the 3/8 line off tank/manifold there was decent amount of oil in the air.

    Sent from my SM-G950U using Tapatalk

    That should be a good test. You should see oil at the oiler tubes if the oil is suspended in the air. But if the oil is falling out and collecting in the bottom of the box then at best the distribution would be uneven.

    It might require an oil pump, but those are readily available. Higher pressure = smaller droplets. I think the best case would be a diesel injector pump and an injector, but that may be way overkill and stupidly expensive. Not really sure what the most reasonable solution is. The waste oil burners use an air mix nozzle and don't use much air, but they are fed with a small pump.

    Jim Dawson
    Sandy, Oregon, USA


  14. #34
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Alright a small update. I really cranked the oil output of the lubricator up and was able to get lots of oil to each component but it was/is way to much oil. I think I may have the wrong type of oiler, mine is a mist oiler and I think I may need a microfog oiler. For now I was able to get some lube to things so I could keep moving.

    I have been trying to button up odds and ends on the machine while I wait for a timing belt for the Y axis. I wanted to mount the control panel to the machine that way if I ever need to move it I could more easily move it. Unfortunately I could not come up with a good way to mount with the space I have available to I mounted it to the wall with some unistrut. I have the main power just temporarily installed until I can re-run it permanently. I need to mount a cooling fan in the cabinet yet. On the previous machine I had the controls up high and just always ran with the door open. Early on I also wanted to see what the drives were doing so having it open helped. Now I am comfortable in how the control operates I think I can have the door shut.
    Rebuilding my 5x10 CNC Router-20200104_214320-jpg

    Next was to finally level the machine, I used 3/4 inch leveling feet from mcmaster one at each corner. The leveling went ok using 12 inch starrett machinist levels having the whole table machined made leveling very easy. One thing I wish I had done differently is to use fine thread leveling feet. It was everything I could do to turn the wrench when I had to jack a corner of the machine. I actually got the floor jack under the one corner to ease some weight on the leveling bolt.

    I ran the gantry servo motor auto tuning and learned I still need to bolt this machine down even though it probably is in the 7000lb neighborhood. With the gantry running 2 inches and an acceleration of 150 in/sec2 it walked the machine all around. The auto tuning also failed about 70% through I think this was do to the walking, it was causing error/overshoots/etc. In order to get the system to finish the tuning I dropped the accel to 100 for now so that I could get a good initial PID tuning from the autotuning function. I will need to re-level the machine once I am ready to bolt it in place.

    Here is a video of the X axis running 600IPM and still the 100 accel



    One thing I am struggling with, the X axis servos will fight each other a bit. I had this trouble on the old machine but because the x axis linear rails moved a bit (they had no reference and were just bolted to epoxy) it really wasnt a concern. On this machine there is no movement so the servos are really straining. What I mean by fighting each other;

    When the machine is first powered on and I enable Mach4 both servo controls are reading roughly the same torque output (I have the display set to torque) the display shows + or - 1 or 2. Then I home the machine. During the home jog the 2 x axis servos read close to the same torque call it -20. Once the machine comes to a stop after the home sequence the output is no longer the same one might show -20 other shows +20 and these numbers fluctuate. When the machine is disabled there is an audible bump that comes from this fight being released also the DRO moves a bit as the servo (assume Motor 0) rolls back. When it is re-enabled there is another bump as they each drive to their previous position. To me it is as if the 2 servos are homing to a different spot or something however that shouldnt be possible.

    I have just 1 home switch for the x and both motors are mapped to it
    Motor 4 is slaved to Motor 0 in Mach
    In the CSLABS config I have home to index turned off
    In Mach 4 controls the motors are setup exactly the same
    All mechanical hardware has been checked and re-checked

    Any servo experts out there have some thoughts? My initial thinking is something is not set right in the servo drives themselves. I have gone through every perimeter and made sure they are the same on both drives. I also did the voltage 0 procedure on both drives which didnt seem to help. The odd part to me is that before the homing they are running very in sync so something during the homing causes this.

    Any help is greatly appreciated, hopefully it is just something I am overlooking.

    Attached Thumbnails Attached Thumbnails Rebuilding my 5x10 CNC Router-20200104_214320-jpg  


  15. #35
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    10768
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Quote Originally Posted by Black Forest View Post
    I don't like it when I see someone build such a flimsy machine! You really need to rethink how rigid this machine won't be when you are slamming the gantry back and forth at about 1,000,000 inches a minute. It might handle 500,000 inches a minute but now way in hell will it be able to hit 1,000,000 inches minute. Just sayin!!!!!!!
    You should rethink what you just posted , the machine below can run at 2500 IPM and is similar construction, it can cut at over 500 IPM



    Mactec54


  16. #36
    Member Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    4047
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Quote Originally Posted by ScottDamman View Post

    Any servo experts out there have some thoughts? My initial thinking is something is not set right in the servo drives themselves. I have gone through every perimeter and made sure they are the same on both drives. I also did the voltage 0 procedure on both drives which didnt seem to help. The odd part to me is that before the homing they are running very in sync so something during the homing causes this.

    Any help is greatly appreciated, hopefully it is just something I am overlooking.
    I don't think you have mentioned what servos you have. It almost sounds like the motors are trying to twist the frame. Is there a deadband setting in the parameters? Maybe the deadband could be set a bit wider for one of the motors?

    Jim Dawson
    Sandy, Oregon, USA


  17. #37
    Member
    Join Date
    Jan 2005
    Location
    USA
    Posts
    10768
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Quote Originally Posted by ScottDamman View Post
    Alright a small update. I really cranked the oil output of the lubricator up and was able to get lots of oil to each component but it was/is way to much oil. I think I may have the wrong type of oiler, mine is a mist oiler and I think I may need a microfog oiler. For now I was able to get some lube to things so I could keep moving.

    I have been trying to button up odds and ends on the machine while I wait for a timing belt for the Y axis. I wanted to mount the control panel to the machine that way if I ever need to move it I could more easily move it. Unfortunately I could not come up with a good way to mount with the space I have available to I mounted it to the wall with some unistrut. I have the main power just temporarily installed until I can re-run it permanently. I need to mount a cooling fan in the cabinet yet. On the previous machine I had the controls up high and just always ran with the door open. Early on I also wanted to see what the drives were doing so having it open helped. Now I am comfortable in how the control operates I think I can have the door shut.
    Rebuilding my 5x10 CNC Router-20200104_214320-jpg

    Next was to finally level the machine, I used 3/4 inch leveling feet from mcmaster one at each corner. The leveling went ok using 12 inch starrett machinist levels having the whole table machined made leveling very easy. One thing I wish I had done differently is to use fine thread leveling feet. It was everything I could do to turn the wrench when I had to jack a corner of the machine. I actually got the floor jack under the one corner to ease some weight on the leveling bolt.

    I ran the gantry servo motor auto tuning and learned I still need to bolt this machine down even though it probably is in the 7000lb neighborhood. With the gantry running 2 inches and an acceleration of 150 in/sec2 it walked the machine all around. The auto tuning also failed about 70% through I think this was do to the walking, it was causing error/overshoots/etc. In order to get the system to finish the tuning I dropped the accel to 100 for now so that I could get a good initial PID tuning from the autotuning function. I will need to re-level the machine once I am ready to bolt it in place.

    Here is a video of the X axis running 600IPM and still the 100 accel



    One thing I am struggling with, the X axis servos will fight each other a bit. I had this trouble on the old machine but because the x axis linear rails moved a bit (they had no reference and were just bolted to epoxy) it really wasnt a concern. On this machine there is no movement so the servos are really straining. What I mean by fighting each other;

    When the machine is first powered on and I enable Mach4 both servo controls are reading roughly the same torque output (I have the display set to torque) the display shows + or - 1 or 2. Then I home the machine. During the home jog the 2 x axis servos read close to the same torque call it -20. Once the machine comes to a stop after the home sequence the output is no longer the same one might show -20 other shows +20 and these numbers fluctuate. When the machine is disabled there is an audible bump that comes from this fight being released also the DRO moves a bit as the servo (assume Motor 0) rolls back. When it is re-enabled there is another bump as they each drive to their previous position. To me it is as if the 2 servos are homing to a different spot or something however that shouldnt be possible.

    I have just 1 home switch for the x and both motors are mapped to it
    Motor 4 is slaved to Motor 0 in Mach
    In the CSLABS config I have home to index turned off
    In Mach 4 controls the motors are setup exactly the same
    All mechanical hardware has been checked and re-checked

    Any servo experts out there have some thoughts? My initial thinking is something is not set right in the servo drives themselves. I have gone through every perimeter and made sure they are the same on both drives. I also did the voltage 0 procedure on both drives which didnt seem to help. The odd part to me is that before the homing they are running very in sync so something during the homing causes this.

    Any help is greatly appreciated, hopefully it is just something I am overlooking.
    Try only having the Home setup for ( 1 ) motor then the other should follow

    Mactec54


  18. #38
    Member
    Join Date
    Jan 2016
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    A couple updates:

    I tried just mapping the master motor to the home switch Mach4 errors out for motorkit 4 homing
    Since it seemed ok before homing I tried to run the machine some without homing. Not cutting just some MDO moves. As I watched the drives I noticed that they were also binding during these moves slightly. When I stopped the machine ~60in from its start position there was a small amount of release of energy. So it seems the problem is present even without homing.

    Jim I dont specifically see anything called out as deadband is this something that would tell the drive to ignore some of the input? Or ignore parts of the resolver? If there was a setting would it be trial and error to adjust it? The servos are chinese servos from bstlinear

    I checked all mechanical connections again on both X axis servo/ballscrews/linear rails. Nothing was found to be loose.

    Couple of things I am thinking (pulling at straws...):
    The base servo tuning in the servo drives dont match each other therefore one motor is responding differently to input and resolver feedback. I will need to learn alot more about servo drives and motors before digging further into this
    Maybe one drive is coming on before the other when they are initialized. I have the delay set to 100ms on both (have tried adjusting this but I have always kept them the same for both drives)
    For a minute I was thinking what if the ballscrews were not the same i.e. one was out of spec. But I think if this was the case it would bind while it traveled but after coming back to home position it would be bind free

    Anyway open to thoughts and things to try
    Thanks for the help on this


    Working on vacuum hold down plan - got my hands on a refurbished 15hp motor. I also have a roots blower being shipped in that I purchased used on ebay. I think in some cases 15hp will be marginal but I think if I employ good zoning and leak coverage it will be adequate. The blower I got can take up to 25hp but I dont have the power to run 25hp motor. With dust collection, air compressor, vacuum pump, 10hp spindle, and the cnc running I think I will be close to the 100amp sub panel that I have in the garage. I plan to put a current probe on and see where I am falling when each thing running individually, dont want to trip the main 100amp while this thing is running. I will run the vacuum pump off a 40hp ABB VFD at 440volts. I gave it a quick paint job even though it had new paint after being refurbished. This way it matches the machine.
    Rebuilding my 5x10 CNC Router-20200109_213609-jpgRebuilding my 5x10 CNC Router-20200107_144015-jpg

    Also been working on the vacuum bed plan. It is a work in progress but the plan right now is to make Qty 6 - 20 inch wide by 64 inches long sections. These will be bolted down to the steel frame and are being designed to be easily removable so in the future I can add a small water table for the plasma or a T slot table on part of it for cutting aluminum. The plan is to make each section out of 3/4 birch 11 ply with solid core PVC sheet glued to it. Then route the vacuum channels into the top using the machine. I also plan to put strips of T slot into the table so that I can clamp things down when vacuum wont work. I have not decided what to do for that T slot, ideally they stick up through the sacrificial board but the concern is some lost area for vacuum. If I keep them below the MDF that will be ontop I would have to take the MDF off each time I want to use the T slot and also risk damaging the PVC vacuum plenum as I slide material on an off the table. I had hoped to use aluminum for the base and vac plenum but the price was just too far out of the price range.
    Rebuilding my 5x10 CNC Router-vacuum-bed-jpg
    Rebuilding my 5x10 CNC Router-vacuum-bed2-pngRebuilding my 5x10 CNC Router-vacuum-bed3-jpg

    Attached Thumbnails Attached Thumbnails Rebuilding my 5x10 CNC Router-20200109_213609-jpg   Rebuilding my 5x10 CNC Router-20200107_144015-jpg   Rebuilding my 5x10 CNC Router-vacuum-bed-jpg   Rebuilding my 5x10 CNC Router-vacuum-bed2-png  

    Rebuilding my 5x10 CNC Router-vacuum-bed3-jpg  


  19. #39
    Member Jim Dawson's Avatar
    Join Date
    Dec 2013
    Posts
    4047
    Downloads
    0
    Uploads
    0

    Default Re: Rebuilding my 5x10 CNC Router

    Jim I dont specifically see anything called out as deadband is this something that would tell the drive to ignore some of the input? Or ignore parts of the resolver? If there was a setting would it be trial and error to adjust it? The servos are chinese servos from bstlinear


    Deadband may be called something else, hard to say, or there may not be any way to adjust them. Yes, I think it would be trial & error. Do you have a link for a manual?

    Not sure if it would work, but maybe an adjustable coupling between one of the motors and the ball screw would work, to get it in time.

    Something like this

    Rebuilding my 5x10 CNC Router-25044-500x375-jpg
    Or maybe you could just rotate one of the couplings on the shaft with the torque on it and get the system to relax?


    Attached Thumbnails Attached Thumbnails Rebuilding my 5x10 CNC Router-25044-500x375-jpg  
    Jim Dawson
    Sandy, Oregon, USA


Page 2 of 2 FirstFirst 12

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

Rebuilding my 5x10 CNC Router

Rebuilding my 5x10 CNC Router

Rebuilding my 5x10 CNC Router