It is currently Sat Jan 20, 2018 9:59 am

 

Can't figure this out...

LDR plasma table discussion forum
comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Can't figure this out...

Postby comeoutswingin » Fri Nov 11, 2016 5:07 pm

Table is a 5x10 LDR. I cannot figure out why I am having this issue... I have modified it to use Early Torch off which has corrected a majority of the problem with totally ruining the parts at the end of the cut, but I still have this ridge issue on EVERYTHING I cut and I'm getting pretty sick of it..

This shows what I am talking about. The alignment of the line at the top is not even close to being straight with the other. This is where the lead in comes in at. These drawings are perfectly aligned from cad so I know its not a drawing issue. I am not so certain its a mechanics issue because it only happens where the start and end of a cut happen.

1.jpg


Before early torch off this is what I got out of my parts. Pretty neat.

20161002_170018_1478901443729_resized.jpg

You do not have access to download this dxf file.
Become a Contributing Member to gain access to the thousands of files shared by our members. Click Here

5x10 LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Fri Nov 11, 2016 8:38 pm

Send me your sheetcam Job file and I will try to help.

Thanks
Dan
LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Fri Nov 11, 2016 8:42 pm

I know this sounds crazy, but are you certain the piece is not moving do to heat right before the end of the cut?
I ask because the inside cut does not seem to have the issue. I would assume if it was mechanical it would be present on all cuts?

Dan
LDR
Last edited by dhelfter on Sat Nov 12, 2016 12:10 pm, edited 1 time in total.

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sat Nov 12, 2016 12:10 am

Dan the inside does in fact have the issue. I will take pictures of parts I have that also have the problem.
5x10 LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sat Nov 12, 2016 7:24 am

It seems much much more pronounced on the outside cut in the pics? The holes an inside square don't appear to bad?
Please email me your gcode and job file so I can help.

Thanks
Dan
Ldr

User avatar
acourtjester
Elite Contributing Member
Elite Contributing Member
Posts: 2889
Joined: Sat Jun 02, 2012 6:04 pm
Location: Pensacola, Fla

Re: Can't figure this out...

Postby acourtjester » Sat Nov 12, 2016 9:33 am

I would attach a pen to your torch and mark paper to see what you get. If it is a mechanics problem the pen tip is much smaller then the kerf and will show up better for troubleshooting.
Also increase the size of the box if your axis calibration is off slightly it will increase as the size goes up.
DIY 4X4 Plasma/Router Table
CandCNC DTHC-IV & Feather Touch
Hypertherm PM65 Machine Torch
Drag Knife and Scribe
Miller Mig welder
13" metal lathe
Small Mill
Everlast PM 256 tig

User avatar
tnbndr
Elite Contributing Member
Elite Contributing Member
Posts: 1368
Joined: Mon Jan 09, 2012 4:30 pm
Location: New Berlin, WI
Contact:

Re: Can't figure this out...

Postby tnbndr » Sat Nov 12, 2016 9:54 am

Have you tried a lead out?
Dennis
LDR 4x8, Scribe, DTHCIV
PM45
DeVilbiss Air America 6.5HP, 80Gal., 175psi, Two Stage
[email protected], [email protected]
Miller 211
RW 390E Slip Roll
AutoCAD, SheetCAM, Mach 3
https://www.facebook.com/ikescreations?ref=hl
http://ikescreations.com

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sat Nov 12, 2016 1:27 pm

I've tried every possible combination of lead in/ lead out.. All adding a lead out does is dramatically waste my nozzle life. With my previous system I've never needed to use a lead out and never have had any issue because of not having one, so I have a really hard time believing that a lead out is the solution.
5x10 LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sat Nov 12, 2016 5:47 pm

Really? I have never seen good cuts without a lead out. I am not sure why a lead out on that size piece would burn your nozzle? I can see that on a large leadout inside offset and the profile is small enough the slug falls away.
If you have tried every combination, then guess I would mount a pen very rigidly and "draw" the part. This will tell us if it is mechanical issue.

Dan
LDR

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sat Nov 12, 2016 5:49 pm

Dan I've been trying to get a hold of you today. I have some serious problems going on unrelated to this issue...
5x10 LDR

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sun Nov 13, 2016 2:02 am

I mounted a pen best I could. You can see the the start and end points are not lining up... Now, it does not matter what drawing I use.. they all do it..

Measured with my digital calipers, the two lines are .0375 apart....
pentest.jpg

You do not have access to download this dxf file.
Become a Contributing Member to gain access to the thousands of files shared by our members. Click Here

5x10 LDR

User avatar
acourtjester
Elite Contributing Member
Elite Contributing Member
Posts: 2889
Joined: Sat Jun 02, 2012 6:04 pm
Location: Pensacola, Fla

Re: Can't figure this out...

Postby acourtjester » Sun Nov 13, 2016 11:23 am

From this test you can determine which axis is not correct, calibrate that axis steps per inch and redo the pen test to verify the problem has been corrected. When you run the step calibration use the longest distance that you can move the axis. There is an outside chance you have a shaft slipping a small amount too.
Here is a post about how to get to the axis calibrations function of Mach 3 if you are using a different screen for Mach 3.
viewtopic.php?f=55&t=20237&p=117675&hilit=steps#p117675
DIY 4X4 Plasma/Router Table
CandCNC DTHC-IV & Feather Touch
Hypertherm PM65 Machine Torch
Drag Knife and Scribe
Miller Mig welder
13" metal lathe
Small Mill
Everlast PM 256 tig

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sun Nov 13, 2016 12:08 pm

I don't believe it is an axis calibration as your start and end point would still come out correct, just the piece would be incorrect in size. If it is mechanical it must be backlash.
I would try three things. First get rid of any lead in our out just draw a square and using the pen do no offset. This makes it as simple a shape and program as possible.
Second is program lead in on left or right side then another at top or bottom.
Third is draw a straight line maybe two inches then down one inch then back two inches then down one inch and so on. Using a ruler line up all the direction change marks. If it is backlash it will grow with each direction change.

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sun Nov 13, 2016 3:47 pm

Did the line thing, and drew a line with a ruler as accurately I could... its off, and it seems to be off in the x axis?
back.jpg

You do not have access to download this dxf file.
Become a Contributing Member to gain access to the thousands of files shared by our members. Click Here

5x10 LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sun Nov 13, 2016 4:03 pm

If you rotate your drawing 90 degrees and try that will tell us about the y axis.

I would check on the x axis
Set screws on both small pulley and large pulley
Belt tension
Pinion gear engagement in rack

Looking at your original pics the thing I find odd is the lines that do not line up are on the top which if cut in the orientation pictured would indicate an issue with y axis not x

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sun Nov 13, 2016 4:45 pm

I've checked all the pulley set screws, and the pinion is engaged tight to the rack. I can move the pinion, but that that point i'm applying so much force to make that happen that it is just moving because the spring is stretching.

in the picture above, left to right is X, and top to bottom is Y. I flipped the drawing but after numerous attempts at getting a good drawing I'm just getting deflection in the pen itself.
5x10 LDR

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sun Nov 13, 2016 5:22 pm

The only slop I can find seems to be in the motor itself.. I took this video with the motors powered up..

https://youtu.be/-eHegZLhTOI
5x10 LDR

tcaudle
Posts: 841
Joined: Tue Feb 28, 2012 6:47 pm

Re: Can't figure this out...

Postby tcaudle » Sun Nov 13, 2016 5:38 pm

Locked motors do not have slop. Either the motor is not locked (freewheeling ) or the belt pulley is slipping on the shaft. The cuts look like your gantry is not square.

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sun Nov 13, 2016 6:19 pm

Looking at the video, think you find the problem! As Tom said, locked steppers do not have slop. So if in that video you had power supplied to motor, then either the motor is bad or the set screws on small pulley are not firm. I would remove the set screws from the motor pulley, and make sure that one of them is aligned with the flat on motor shaft. Then tighten them up and check again. If still has slop then a new motor is needed

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sun Nov 13, 2016 6:32 pm

Well, I tightened the set screws... One of them, I snapped an allen wrench off getting one loose, the other the set screw gave way before the allen did.. so I couldnt get that one out.. and probably never will. I aligned the good set screw with the flat spot and tightened it down.. It helped, but still in a line test its not perfect....

I can probably pull the stepper off the machine and put it on the mill and try to get that stripped set screw out, but i'd put money on it that I will just wreck the pulley itself...

I'm not the right track now I guess.. but that leads me to my next problem... mach 3 is over riding my cut speeds in the Gcode, cutting one part possibly correctly and then reverting the rest to 35ipm...
5x10 LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sun Nov 13, 2016 7:46 pm

Small set screws should have heat supplied to break the loctite bond.
As for the cut speed, going to have to get some other folks input. i have never seen mach3 override a federate by itself. If it is not able to move as fast as programmed it will of course slow down but not to 35 ipm. Maybe to at candcnc can help us.

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Sun Nov 13, 2016 7:52 pm

At first I thought I messed up and applied the wrong tool, but I went back in sheetcam and double checked. I went as far as to reboot everything and reexport the post. The majority of the parts would cut at the 35ipm when they were programmed for 90ipm. I manually bumped the feed rate up, but it would carry over and the Z would run at the increased feed rate and usually crash. Then it would jump over to another part, and would multiply the feed rate up from 90(correct) to a percentage up like 150-200%, which would just skim across the material.

I loaded another .tap file that has been cut 100s of times correctly and it exhibited the same behavior???
5x10 LDR

dhelfter
Posts: 109
Joined: Wed Sep 09, 2009 11:18 pm

Re: Can't figure this out...

Postby dhelfter » Sun Nov 13, 2016 8:16 pm

Can you video your mach screen while cutting a part?

comeoutswingin
Elite Contributing Member
Elite Contributing Member
Posts: 122
Joined: Fri Oct 16, 2009 10:34 pm

Re: Can't figure this out...

Postby comeoutswingin » Wed Nov 16, 2016 2:19 pm

I will video the issues next time I am able to cut parts... It's back to the real job time for me now.
5x10 LDR

tcaudle
Posts: 841
Joined: Tue Feb 28, 2012 6:47 pm

Re: Can't figure this out...

Postby tcaudle » Wed Nov 16, 2016 8:00 pm

comeoutswingin wrote:At first I thought I messed up and applied the wrong tool, but I went back in sheetcam and double checked. I went as far as to reboot everything and reexport the post. The majority of the parts would cut at the 35ipm when they were programmed for 90ipm. I manually bumped the feed rate up, but it would carry over and the Z would run at the increased feed rate and usually crash. Then it would jump over to another part, and would multiply the feed rate up from 90(correct) to a percentage up like 150-200%, which would just skim across the material.

I loaded another .tap file that has been cut 100s of times correctly and it exhibited the same behavior???


Sounds like the motor tuning on at least one axis is honkered. The machine will only go as fast as the slowest velocity setting in Motor tuning no matter how fast you tell it in code. The feedrates are in the code as an "F" command and MACH simply reads them and sends the number to the program. If code ran before and not now then its a setting in MACH.


Return to “LDR Motion Systems”

Who is online

Users browsing this forum: No registered users and 1 guest