Machsupport Forum
General CNC Chat => Share Your GCode => Topic started by: lokey on April 07, 2009, 01:08:12 PM
-
Hi, after the recent purchase of a Zenbot, Mach3, and LazyCam I discovered a problem of a accumulating positive shift on all axis that was not being displayed on the DRO’s of Mach3. This was really apparent within projects containing multiple circles being cut around a central point. After trying to solve the problem using the many setting available within Mach3 and not being able to solve the problem I was able to correct the problem by writing a macro. Thank you ArtSoft for providing the training videos. Using the macro has greatly increased the accuracy of the machine. If any one knows how to solve the problem without using the macro please let know.
Best Regards
-
lokey, I may be wrong but it sounds to me like your steps per aren't set right.
Brett
-
Hi, I am new to CNC and I tried many times to correct the problem using that method. I used a dial indicator. I would set code to move 2 inches and it move 2 inches. When I repeated the process there was a accumulating positive shift of 0.003 after each test run that was not showing up on the DRO's. When I use the macro the shift does not occur.
I will keep trying to correct the problem using your suggestion. It would be nice not to use the macro, shorter cut time. Thanks for your feed back.
Best Regards
-
Is it possible this is to do with active high or low on the step or dir pin not being compatible with your drivers?
Gaining or losing a step on each direction change?
Greg
-
Hi, I tried active low and active high, still had the same problem. I ran test code of a series of circles around a central point without using the macro and pressed goto z to return to the start location and it was off to the positive on the x and y by a noticeable difference. I ran the same code using the macro and it returned to the correct start location. The problem appears every time the axis moves and there is a accumulating shift of 0.0015 to the positive that does not show on the DRO‘s. There maybe a setting within Mach3 to correct the problem, I read through the doc files but I am unable find it. I will keep trying and thank you for your feedback. Best regards
-
Art, the author of Mach3 is quite clear that the DROs show only the pulses that have gone out the printer port.
If the DRO shows it has moved one unit then the PP has put out one unit worth of pulses. You will find that the problem lies somewhere after that fact.
Most likely mechanical but may be electrical. Start by halving the rapid speed and acceleration rate and retest.
Greg
-
Hi, I try the suggestion and there is still a positive accumulating shift of 0.0015 after all axis movement. I am new to Mach 3 and if there are any other possible solutions to the problem please let me know. I attached the part .dfx file, it is the only moving part of the project and needs to center. Even with the macro the inner circles cut fine but the outer circles are offset.
Best regards
-
When you say you have a shift of .003 are we talking inches and is the position getting nearer to the home position every run?
Graham
-
Hi, yes inch’s. I am using a Zenbot with Mach3, LazyCam, and TuboCad. In the system I found there is a accumulating positive shift after each axis movement that doesn‘t show up on the DRO‘s. . I am new to the software and there maybe a setting to deal with the problem but I am unable to find it. The problem becomes really apparent with a project that include cutting circles around a central point. The last outer circle would be so far off center that the part would be useless.
I was able to write a macro that solved the problem. I have to run the macro after each axis movement. Because the Zenbot has drive belts the macro works really well and it has increased the precision of the Zenbot to a surprising level. Great bang for the bucks when your just starting out. I would not use the macro if the machine had ball screws.
If you know of a solution to the problem without using the macro please let me know. Thanks for feedback and best regards.
-
In the system I found there is a accumulating positive shift after each axis movement that doesn‘t show up on the DRO‘s.
Does this lost position happen on every axis or only one? Logic says that it must be one of two things.
First is mechanical slippage. Somewhere between the stepper shaft and the axis.
Second is lost steps on the stepper. This can be mechanical load being beyond the steppers capability. Maybe a tight spot. Maybe acceleration and or rapid too fast.
What stepper motors, driver, power supply are you using? Are you using a break out board?
After ruling out mechanical issues cut the acceleration and rapid speed in half and see if that makes a difference.
Greg
-
Hi, I tried the suggestions and still have the problem. The accumulating positive shift occurs on all axis. I did notice that after performing G28.1 code that some axis do return to zero and then display +0.0009 or -0.0009 on the DRO's without any additional code. That is indicating that it is software related and is why I had to include so many reset DRO's to zero commands in the macro.
I am not software engineer but I am thinking that it maybe a truncation issue. I am still searching for a solution that does not require the using of the macro. If other possible solutions to the problem come to mind please let me know. Thank you for your efforts and best regards.
-
Hi, I tried the suggestions and still have the problem. The accumulating positive shift occurs on all axis.
OK cutting acceleration and rapid in half did not fix it then cut it in half again and retry.
You did not say what drivers etc you are using. The drivers may need a longer pulse.
Also try "Sherline mode"
Greg
-
Hi, the system is a Zenbot 12x16 CNC router. I tried all the above suggestions and the problem remained un-solved.
I noticed that when I used 28.1 G code that the DRO's would not reset zero but would reset to a + or - 0.0009. I rewrote the macro to adjust for the DRO display error and run the macro after each chain is cut and the machine now cuts the project very well.
The problem could be a truncation issue with the software. Thanks to everyone for the support.
Is there away to change the directions of the cuts within LazyCam? If anyone knows please let me know.
Best regards.
-
Lokey you need to get past thinking the pulses are not lining up with the DROs
The DROs only count the pulses as they go out the LPT. They do not report pulses that Mach thinks it has sent out. Only pulses that do go out.
The problem exists somewhere after that fact.
If after cutting a job the DROs return to zero and the machine is off position then you have lost steps for whatever reason.
Greg
-
Hi, I understand what you saying. The DRO's report what has been sent to p-port. After I use a 28.1 G code command the DRO's should re-set to zero. They don't, they re-set to a + or - 0.0009.
-
That will depend on what steps per unit you have for that axis.
If it a nice round number like 1000 steps per unit the DROs will read zero.
More likely you have a number like 567.689 steps per unit.
In the background Mach does all its calculations to ten decimal places.
My router has 106.7653 steps per mm and it will cut complex jobs over many hours and return to zero within a thou.
The software will not accumulate a measurable error after every cut segment you must look beyond that point.
It might be electrical, pulses may not be getting to drivers reliably, drivers may be missing pulses. You still have not told us your driver, power supply, BoB etc setup.
Or you are missing steps because of performance tuning or mechanical issues or even mechanical slippage.
Greg
EDIT: This is a recent purchase. What have the people that you bought it from said about this.
Also what is the rapid speed and acceleration set to? And what is the lowest values you have tested it at?
-
Hi, the step per unit are 1112 as recommended by the manufacturer. After I use G-code G28.1 X Y or Z 0.0015 the DRO‘s do not reset to zero, but reset a + or - 0.0009..
I am trying to look beyond the macro solution.
When I tried the calibration axis feature the circles would cut as ovals. Motor tuning are set to spec’s. I checked the machine for possible mechanical issues and was not able to find any. When I cut a hole for a bearing on a project it is very accurate. The problem only occurs after movement to the next cut location.
The dial indicator I use for testing measures to 0.0005. The control box came with the Zenbot, if it was missing pulses would it travel less and stop short.?
I really appreciate your efforts and will continue to implement your suggestion to try to find a solution. Thanks again and best regards.
-
Motor tuning are set to spec’s.
You did lower rapid and acceleration as I have suggested about three times, haven't you? :)
Let me know what settings are spec and what lower settings you have tried.
When I cut a hole for a bearing on a project it is very accurate. The problem only occurs after movement to the next cut location.
This is a strong indication that lowering the rapid and acceleration or "tuning them" will make a difference.
While it is cutting the hole it is travelling at cut speed, as set in gcode.
Then when it rapids to the next location is when it is losing its position.
Greg
-
Hi, I posted a jpeg of the tuning page. If this is not the info you are requesting please let me know where I can find and set it.
I did try lower setting.
-
Try setting Velocity to 100 and acceleration to 10 and try your cutting job.
Do that setting for each axis and save axis settings each time.
Let us know if that improved position error.
Greg
-
Hi, I tried the suggestion and there is still a accumulating positive shift of 0.0015 .
-
OK set velocity to 30 and acceleration to 5 and try again.
You still have not told us your driver, power supply and break out board setup.
Explain how you are measuring the 1.5 thou discrepancy?
Post your XML.
Greg
-
Hi,
“OK set velocity to 30 and acceleration to 5 and try again.”
I did the suggested resets and there is still a problem.
“You still have not told us your driver, power supply and break out board setup.”
The control box and power supply came with the Zenbot.
“Explain how you are measuring the 1.5 thou discrepancy?”
Method for measurement:
1. I secure a heavy metal plate to the tooling bed.
2. Attach dial indicator with magnetic base to plate.
3. Home all axis and zero display.
4. jog to dial indicator G0 X2,Y2,Z-2 and zero the indicator
5. Home axis G28
6. Jog back to dial indicator G0 X2,Y2,Z-2
7. This is where dial indicator starts showing the accumulating positive shift of +0.0015
8 repeat steps 5 and 6 the error increases a constant +0.0015 after each repeat and the error does not show up on the DRO’s
If I use the same method but replace G28 with G28.1 X-0.0015 Y-0.0015 Z0.0015 then no error occurs.
I did notice that when I use G28.1 the DRO’s do not reset to zero but reset to + or - 0.0009. They should reset to zero after a G28.1 command. That is why I had to reset the DRO’s to zero so many times within the macro.
“Post your XML.”
I posted the XML and I am continuing to review the config Mach3 pdf . Thanks again for efforts and best regards.
-
You have posted the wrong XML. That one is blank. No settings.
Post the XML that you are using.
Greg
-
Hope this is the correct one
-
"The control box and power supply came with the Zenbot."
Need a better answer than that. :) How about some details?
I think it is very unlikely that your steps per unit is correct.
The chance of it being a nice round number like 1112 when using a belt dive is, to my mind, almost impossible.
That would easily explain a 0.0015" discrepancy.
Greg
-
I do not see how a G28.1 can fix your problem, all it is doing is homing the machine via an intermediate point X0.0015" etc.
If you change the X value to X-2. you will see what I am saying. The X axis will move 2" away from home and then home.
Have a look in your fixture offset table and see what is in G59.253 and above.
Graham
-
Graham
Lokey does not have home switches so in effect he is simply adjusting the machine coordinates by the 0.0015 amount each time.
Moves to the intermediate point and the zeros the machine coordinates without any further movement.
Greg
-
This is the machine...yes?
http://www.zenbotcnc.com/merchant2/merchant.mvc?Screen=PROD&Store_Code=&Product_Code=zenbotMini&Category_Code=BestSellers
RC
-
RC
Lokey is not online right now. I am presuming that is the one.
Greg
-
Hi, the machine purchased is located at
http://www.zenbotcnc.com/merchant2/merchant.mvc?Screen=PROD&Store_Code=zenbot&Product_Code=zenbot1216&Category_Code=Products
I do not have any other information on the control box.
When I tried the calibration axis feature to change the steps per inch, the circles would cut as ovals.
-
Hi, I down loaded a free version of KCAM 4 set the steps per inch to 1112 and ran G code:
G0 X1
G28
and repeated code many times and the dial indicator showed no error.
The problem is not with the Zenbot I am missing a setting some where in Mach 3 that is causing the accumulating positive shift of 0.0015 that is not showing up on DRO's. How do can tell what version Mach3 I have. Maybe I downloaded the wrong version. If anyone knows please let me know. Best regards
-
The problem is not with the Zenbot I am missing a setting some where in Mach 3 that is causing the accumulating positive shift of 0.0015 that is not showing up on DRO's.
No I don't know of any Mach setting that I have not already suggested.
Glad you have it working to your satisfaction now.
Good luck with KCAM.
Greg
-
Hi, KCAM is not the answer I am looking. It only shows that the the problem is not with steps / unit. I paid to use Mach3.
I am still looking for solution to the accumulating shift problem or is there away to get a refund?
-
Lokey
I am only a forum member and not an employee of Artsoft.
Just one of the thousands of happy Mach3 users.
I am sorry I am not able to help you.
I do wonder why Zenbot are not helping you with your new machine.
Suggest you send an email to Artsoft regarding your refund request.
Greg
-
Hi, thank you I greatly appreciate all the help that I received in trying to solve this problem.
Thanks again to all the members and best regards
-
Hi Lokey,
Just went through this thread, and it looks very familiar to me ;) I had this same problem a while ago. To sort it out, what you're seeing is lost steps upon reversals. An easy test can be made to prove it. You don't need to home your machine and read the dial indicator after that. Leave the same setup with the dial indicator and use incremental jogging in Mach3. Set the increment to, say 0.005", then jog one increment towards the dial indicator (remember the reading), then one increment backwards, then again towards. The dial indicator reading theoretically should now show the same as the first time you jogged towards the the clock. In your case I'd bet it would show 0.0015" off.
If this is the case then your loosing steps on reversals. Your stepper driver expects the direction change pulse to appear in a particular correspondence with the step pulse. Some drivers work on the leading edge of the pulse, but most, I think work on the trailing edge (don't quote me on that). For the last case, for example, it means that a pulse is registered be the driver only at the trailing edge. So depending on when the direction pulse occurs relative to the step pulse, you get an axis reverse in the correct moment or not. If the direction pulse is sent "late" then all the steps(actually one) sent till that moment would still move the axis in the previous direction.
Theoretically, changing the active state of the step pulse should solve the problem. But sometimes it doesn't work (as it was in my case). Sometimes, the break out board messes up things. To check this you'd have to bypass the beakout board and feed the signals form the PP directly to the driver. Assuming your PP output voltage is sufficiently high for the driver.
If you want to learn more on the subject and my path in solving this problem, follow the thread "HELP! Accumulative moves" on Mach3 Yahoo Group, at
http://groups.yahoo.com/group/mach1mach2cnc/message/102773
And also the thread "HELP! Accumulative moves - Good News" at
http://groups.yahoo.com/group/mach1mach2cnc/message/102937
Daniel
-
Hi Daniel, thanks I tried you’re your test procedure and it show that every time I moved the axis there is 0.0009 accumulating shift in the plus and minus direction that is not showing up on the DRO’s.
Using different control software and using the test procedure showed no problem.
I did notice that in ports pins config. under mill options that Mach3 is not allowing me to select No Home Switch. I do not have the Homes enabled, is there another way to enable the selection?
If anyone knows please let me know, thanks again and best regards