![]() | |
| Home Page | Mark Forums Read | Today's Posts | My Replies | Classifieds | Reviews | Photo Gallery | Web Links | Share Files | Advertise With Us | Ad List |
| |||||||
| CNC Plasma and Waterjet Machines Discuss building, operating CNC Plasma, waterjet and EDM machines here! |
| This forum is sponsored by: |
![]() |
| | LinkBack | Thread Tools | Search this Thread | Display Modes |
|
#2
| |||
| |||
| ...sounds like a mechanical problem. Does it have ball screws or is it rack & pinion? First thing i'd check is the motor coupling or screw coupling if belt driven. If the encoder is on the tail end of the motor, check it's coupling also. Same goes if it's mounted separate from the motor. The shape of the hole points you to the axis with the problem. Typically the axis is traveling short of the commanded distance, while the readout will display correct position. If it's stepper driven open loop, or no encoder, you can concentrate on the motor coupling, screw or pinion. You left out how far off the holes are, so I'm assuming a large enough error that it's visual, which should be more serious than a backlash adjustment. |
|
#4
| |||
| |||
| Ingersoll Rand.... very common to have to rebuild the ballscrews... since the force of pressures of 56K it puts alot of pressure on the ball screw.. there is a place in grand rapids michigan that regrinds ball screws good luck |
|
#5
| |||
| |||
| Its belt driven using linear shafts imbedde in a aluminum extrusion. Guide wheels on ecentric cams for tension adjstment. Thw belts ,I have torqued to proper setting I have whatched the CMC drives for backlash dont see any ,Have not tried using indicator and jogging avis yet,I will try that tommorrow and reject all drive cable connection also, The holes are off very much :} Visualy is an under statment LOL. I do sometimes get a skew limit error when homing and i adjust limit switches until it is corrected . THNX for the help ALL im a machinist of 20 years>>>>>> but first time at the JET :} |
| Sponsored Links |
|
#6
| |||
| |||
| Could be a couple of issues. You should "map" the screw to see if you have 1:1 motion through the whole path of travel. Uneven motion would suggest uneven wear which would indicate a need to rebuild the screw. Keep moving in 1 direction to eliminate any bearing slop from creeping in. If the ball screw reaction bearings are worn, you'd see real problems at the direction change points. These are typically at 12, 3, 6 and 9 o'clock. Chances are the worn bearings would be on the longer axis of the eliptical shape. DON'T just look to see if it moves 0.100 and then returns. Look to see if you have 1:1 motion. You may find that the motion is NOT 1:1 initially and then "catches up" and moves OK. With properly preloaded ball screw support bearings, you should be able to get 0.001" of travel for an equivalent 0.001" screw movement. In our case, we could actually put an indicator in our spindle and move the table to see "slop". If it is that bad, you'll have no hope of cutting a round hole. When all that (ball screw and support bearing service) is said and done, you may find that servo tuning is going to be needed. Once you remove the slop from the screw and/or support bearings, you have to retune any electronic compensation that was "added" at any point in time to cover up the mechanical slop you may have hopefully eliminated. Keep in mind that if the machine has gibbs or even linear guides, wear here could result in out-of-round holes due to ball screw motion being gobbled up by table slop. Ultimately, the problem will not have a silver bullet solution - there's probably a bunch of things that are adding up that need to be fixed. |
|
#9
| |||
| |||
Sir, You should try to setup the machine at a fixed location in the operating area. Then attach a dial indicator to the cutting head pointing along the X axis and apply some force in that axis. Apply force in one direction, note the indicator reading, then go in the opposite direction. Note the difference in dial readings. Repeat for the Y axis. This should show up any lost motion. Then you have to find the cource and correct it. Could be a drive sprocket/pulley on the gearbox output shaft. Where do you have the encoder mounted for each axis? We used to mount the encoder SEPARATELY, on its own pinion and driven off the gear rack (in your case it would be the belt), to put the drive motor and gearbox inside the feedback loop. If you have mounted the encoder on the motor, then likely there is backlash in the gearbox. If the encoder is driven off the gearbox OUTPUT shaft, look for lost motuion there. Also, see that your gain settings for each axis are high enough that your system in just below the point of oscillation. Regards, Jack C. |
|
#11
| ||||
| ||||
| ^^I was at StoneExpo a few years ago and talked to the engineer that built the PTC machine. They use Mach3 for the controller, at least the one that was at the show. I looked it over it's a good machine uses linear motors so you will never need to worry about wearings out ballscrews or belts maybe just the guide bearings. The max speed the linear motors can do is 100 feet per second or something like that. It is all stainless so will take a while to rust One person locally got a Flow flying bridge waterjet, I was pretty disappointed after looking it over a little. I could wobble the Z axis very easily and all the machine controls are by keyboard like Mach3 but plus $50k. A waterjet would be the simplest machine to build (motion system!) out of all the other cutting machines out there, there is no feedback force from the cutting tool other than a few pounds. Now I just need to convince some one that they need a waterjet so I can build one.
__________________ Dennis |
![]() |
| Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
| Thread Tools | Search this Thread |
| Display Modes | |
| |