[mrbroms]

I have a new one. My router (cncshark) isn't following the gcode on the screen (from vcarve). I'm just cutting a circle and some stars.The numbers I'm reading are correct but the router is way off the position and wood. The weird part is it started following the path perfectly for two rounds of depth cuts and then seems to stretch or shift on the x axis. Is there a way to fix this. Could it be caused by cutting too deep too quickly. I'm at .15" depth per cut but I'm using a 1/8 " bit. Also (and maybe related) what is the best way to deep clean my machine?
Any help would be great.
broms

[yohudi]

Hi there,

I had a similar thing happen once on the Y-axis...

turned out that the pre-load adjuster on the Y axis ballnut had worked loose... so it was developing backlash problems (only on the affected axis) as the cut progressed...

seemed to be fine on the first pass .. 2nd pass a bit off.. by 3rd and 4th pass was noticably different... and so on..

tightened up the pre-load adjuster and everything went back to being as right as rain... If both your other axes are fine.. and you know the code is fine.. then the drive mechanics is the first place I'd look.. in my case was pre-load... but anything that might introduce backlash into the affected axis is worth a look... couplings... ballnut... motor drive.. but usually something that can wear or work loose...

EDIT: Other thing that was symptomatic and that might help confirm was... The 'standard' Offset that had been tested and used for months to set the machine Zero was also constantly drifting.. but only in the affected axis... so I would reset the machine zero... apply a new offset.. then jog the axes about a bit... then return to zero... and true enough the y-axis offset had gone west... it's an obvious effect once you know that there's backlash in the screw.. but before you start ripping things apart it might help confirm the problem.. and also that it's nothing to do with the code etc...

hope this helps

Danny

[mrbroms]

thanks Danny-

I will give it a try and let you know how it goes. I really appreciate it!!

chris

[mrbroms]

Ok. I started by using your suggestion. I set up a new 0,0,0 and jogged around the table and then had it go to 0,0,0. It was way off (some of the times). I realized a few things.
1) It was the Y axis.
2) The Y stepper motor was not responding sometimes Y+ and sometimes Y-

I then took off the top of the table and looked at the attachments from the stepper motor. they were solid!!

I then followed up the cable to the connector that I had attached a few weeks ago when I moved the table. I noticed one of the male connections was being pushed out of the back. I figured that it hadn't set right and snapped out. SEE PICT. I pushed the connector into its proper depth after opening the connection and inspecting everything. I thought I had solved it and could happily move forward......... NOT. When I did the jogging routine again, It had closer results some of the time but its far from usable. Very far!! So I removed the Y stepper from the cnc frame and tried to jog it without any pressure from the machine. I noticed that it would rotate in the direction I wanted it to but sometimes when I hit the rotate again it would sound like it was turning but it wasn't. It was very random but only happened when I first hit start. Is anyone still with me? I have lots of questions coming up.

So here we go.

1) If the 4 prong connector was the problem, why didn't it fix it? Is it because while the connector was broken it started to wear down the stepper motor and teeth and now the stepper is broken?
2) Should I replace the connector or the stepper motor or both?
3) in the picture, I noticed the connection goes -
stepper to rbgb wires to double ybrb wires to 4 prong connector to single ybrb wires. Why? Single to double to single? I plan on giving the CNCshark people a call tomorrow.
4) Where can I buy a 269 oz/in 1.8-degree steppers
for a 24V (cncshark specs)
4B) should I use a more powerful stepper for the y axis. I don't want to blow a fuse or board though

Thanks for pointing me in the right direction. More pointing would be great from anyone on any of the topics discussed. I hope someone can learn from my stupidity ... I know I learned a lot.

Thanks for any help,
chris

[yohudi]

Hi there,

glad to see you're making progress.. of sorts...

Also glad to hear the suggestion to check the machine zero manually helped identify the actual source of the problem...

At this point I think you have to see what the manufacturer says re: the type of problems that can arise as a result of the wiring fault you found..

If those wires carry step and direction voltage to the motors.. it would seem position drift was because of intermittent direction power... If remaking the connection didn't fix the problem you have to question if power is still being sent down the repaired lead... from what you say this doesn't seem to be the case.. or the motor would perform as it should on the bench... generally speaking failure to get step or direction power shouldn't really harm the motor.. faulty wiring is more likely to cause a problem in electronics..... obviously you can't take this as gospel...

the intermitent nature of the fault suggests it might be faulty wiring or power feed.. but without being physically present it's hard to say. you mention recently extending the lead in question.... so I'd check it's still OK at the other end.. I think you probably thought of this already though... If you can use a meter it might be helpful to check if voltage is being supplied to the cable that was faulty... this would at least let you know the area the fault lies in....

I'm not familiar with the cncshark machine.. so can't really offer anything constructive re: the way they have it wired. I would therefore speak to the manufacturers and pursue the questions you raised in your last post...

Sorry I can't be more help....

Danny

[lumberjack_jeff]

Assuming all the steppers are the same, try swapping the z stepper for the y.

On edit:
http://www.mechmate.com/forums/forumdisplay.php?f=26
is a good educational resource.

[Montabelli]

I would check to see if the motor behaves correctly when not mounted. if it does, check your y lead screw and see how easily it turns. if it turns freely, then check all quick connects on your plugs to be sure they are making good contact. if every thing checks out ok and the motor still won't operate properly then swap leads to change drivers to the axis and see if the problem shifts axis. if it doesn't then the problem is between the drive and the axis (wire/motor/screw) if it does change the problem goes the other direction.

also, I almost forgot, check out the motor coupling to the screw.

good luck

[mrbroms]

The search continue. I took the advice of Jeff "Assuming all the steppers are the same, try swapping the z stepper for the y." I found that all the stepper motors faltered on the 'Y' connection from the controller, and the 'Y' stepper worked great when connected to the other controller connections. The good news is that I don't have to buy a new stepper motor. The bad news is that it is probably an issue with the controlled itself. Not good because I know nothing about the controller ... except that it is supposed to work. Here's a pict of the controller.

I am in contact with Tim from Next Wave Automation (CNC Shark) and he responding quickly to the connection issue. I hope the controller is an easy fix.

Any thoughts on what kind fix is needed? This is a perfect case of Murphy Law in action.

Also, is there a way to retrofit my setup to add a 4th axis? If so any leads and links would be great.

Thanks again to everyone for helping me out!!!
chris

[15mgtar]

have you try to connect it directly to the controller without that 4 prong connector?

[lumberjack_jeff]

If you're still concerned about the connector, try moving the y pigtail to another controller.

I can't quite tell from the photo, but sometimes the terminal block plugs into a connector on the circuit board - you might not need to unscrew any of the wires to test this theory.

Typically, the next step would be to switch the axis the suspect controller is running, but with a ribbon cable, this seems a little problematic.

Basically we're just switching things around to isolate the problem.

[mrbroms]

Update.... I love warranties. I plugged the wires from the stepper motor directly to the controller and had the same results. At about the same time Tim from CNCshark was very helpful and told me to ship the motor, power supply and controller back to him so he could diagnose and fix it. He said everything was still under warranty. That took a lot off my chest. When I get it back I will find out and post what the issue was so the next guy can learn from this. Thanks to everyone for there tech-support. This is the most knowledgeable and helpful community online!!!
chris

[Magnusan]

Hi

English is not my strong side but I hope you can understand that I mean!


I have a similar problem with my machine (SCM Tech z30). The G-code is generate by MasterCam and then converted by a postprocessor to the PGM-file. I can not se anything wrong with the code and the pp has work fine over two years now so I think it is the machine. I'm going to test the g-code in another machine to but it can may take a while so I have a few question when I wait.

I wondering if you know what it was that caused the error and if you run the same program several times, is the error moving around or is it always on the same location?

Thanks for all help I can get!