[urbanimports02]

ok, starting a new thread about motor tuning and mach3 setup. as far as my stall goes, it is only in one line of code at the end of a rapid. it doesnt seem to be a tight spot in the gibs as it only does it on this one line, and there are 5 or 6 other rapid codes to this same position. this line of code has the y axis moving like .020" and the x axis moving oh 1.5" the other rapids to this location only move the x axis, could be the simultaneous rapid of both axis, but there are several more of those in the code as well, and they all sound nice and smooth with a nice smooth stop before doing next move. i changed that rapid movement to a G1 move and ran the program successfully with rapids at 120ipm, so this has me perplexed. as far as pulse tuning, not quite sure what those numbers do, but mine were at 2 and 2 and i tried a couple different combos to no avail. some wierd things about my setup.
none of the novakon xml files will work on my machine. one they sent me shows external e stop requested and wont reset, another looks very similar to the one i modified to get to work, but none of the axis will move. my limit switches do not seem to do anything, and the settings are all the same as the xml that novakon sent me and the one they just posted a couple weeks ago on the site. i tried changing a couple of the settings, but nothing seems to work. would someone be willing to send me a known good xml file? ihavenofish?

[urbanimports02]

ok, got homing and limit switches to work and implemented soft limits. i seem to have everything pretty well tuned, so i tried that program again with 120ipm rapids, and again it fails at that spot. tried moving the table to totally different location and resetting offsets and rerunning again(testing for tight spot in gibs), and still, same line of code. so i turned rapids down to 50ipm and tried again. still same buzz sound and stall, just more quiet, but still there at the same spot just shy of the rapids destination. this is about the most bazzar thing i have ever seen. all other rapids seem to work fine. i can work around this, but it is really irritating!!

[joeybagadonuts]

Urbanimports02,

Are you sure you have your license file "Mach1Lic.dat" installed in mach3?

Mach3 runs in Demo Mode without it and will only run 500 lines of code.

JoeyB

[urbanimports02]

yes, it is installed, or atleast i see it in the mach3 folder. it doesnt fail as in just stops, it misses some steps at the end of that rapid and keeps on going, just now its out of place and machining the part offset by however many steps it misses, if that makes sense?? all other rapids leading upto and after that rapid are nice and smooth. i killed the breaker for the 240v to the back of the machine so the fan and spindle were not running or anything so i could really hear what is going on. i ran the program in air with 50ipm rapids, and can still hear it buzz, not as bad, but buzz lightly, at the end of that one rapid, all other rapids even when i turned it back up to 120ipm are smooth as silk. when at 120ipm it misses bad and is off like 1/4", when at 70ipm, it is only off .007", but still off. its wierd that it misses at the end of that rapid, where as normally if it is going miss from being too fast, or tight gib or something it misses as soon as it attempts the rapid. it is also a long enough rapid that it is at full speed for some time before stalling, so its not an accel problem. i am thinking it has something to do with the rapid calling for such a slight move in the Y direction while making a larger move in X. it has plenty of rapids that are simultaneous and has no problems with those, just maybe its one of those weird glitches. i changed that rapid to a G1 move with feed at 15ipm, and ran it with no problems, and at least got my part done.

[urbanimports02]

the other wierd thing is the first time it did the big stall and was off by 1/4", the x axis went offline at the exact same time. i stopped the program and jogged up out of the part and tried jogging the x axis and nothing, so i shut down mach and restarted, and still no x axis, so i restarted the cd100, and the x axis was back. could this be a faulty gecko drive or stepper or something?? i figure i have had all sorts of other little problems with my machine that others dont have, so why not?? i also know that my machine was delayed on shipment for like 3 extra weeks and khai said they were having problems or something, but never said what it was, maybe i just need a whole new cd100. i really dont have the time or patients to chase this down.

[joeybagadonuts]

Urbanimports02,

What is the acceleration set at?

Please post the xml

JoeyB

[urbanimports02]

acceleration was set at 40, as per ihavenofish settings. but tried 20 and even 10 and even 10 at 50ipm, and still hear and measure a slight miss at those settings. i will try yo get some time to post the xml tonight. i cant emphasize enough that it only does it on this one line of code at the same spot at the end of the rapid, all other rapids, including several to these exact coordinates are smooth. i am sure you would find plenty wrong with my xml, as i have been creating everything from scratch as none of the xml files i have received from novakon will work for my machine. when i pop them in, nothing will move at all. mine seems pretty good now as i finally have homing and soft limits working well. another thing to note, is the machine will not travel 15" in y like they claim, there is no way. it hits mechanical limits at both ends with only 13.5" in between. one end puts the table against the collumn, and the other is clearly at its limit, and just about touching the front splash shield. i set it to 13.35 for soft limits, and it works great.

[jalessi]

What stepper driver is your mill fitted with?

You may have a resonance issue, do the steppers have harmonic dampers attached?

Jeff...

[urbanimports02]

i believe the stepper driver is a cnc4pc driver, i am not too good at this side of the stepper system stuff. i do not believe the novakon nm-200 has dampers installed on the axis, i am not even sure what that is? my understanding was that the motors were direct coupled to the ball screws. again i am not sure why that would have anything to do with a stall at only one line in one program code regaurdless of where i set the work offset zeros. i have tried multiple different coordinates(different spots on the table) and it always does it at this one spot in the code. the stepper driver or gecko drive or something inside the controler i guess could be to bkame, but i am not saavy on whats all going on inside there, thus i bought a turnkey machine. i will probably understand this stuff more in the near future then i ever wanted to!! any help on this side of things will be greatly appreciated! thanks for the help.
jesse

[urbanimports02]

oops, i guess you were probably refering to the gecko drive, and i am not too sure what exact gecko they are using currently, but i am sure there are plenty of people here that know which one it is. how do i upload the xml? i tried attaching it, it says invalid file type. i tried draging and dropping it in the body of the message, and it just opens it up in a new window.

[jalessi]

The file needs to be zipped.

[jalessi]

Link to Zip Utility

http://sourceforge.net/projects/seve...5.exe/download

[urbanimports02]

testing to see if this worked. attached should be the xml that novakon just posted that they say should work. when i look at the settings, i do not see any thing that should stop it from working, but when i open mach with this xml, nothing will mov or do anything. if this attaches, i will go pull the current xml i have working and post it as well. does someone else have a known good novakon nm200 xml they care to share?

[urbanimports02]

here is the xml i am currenly running. let me know if anything looks off. thanks for the help.
Jesse

[jalessi]

The port on the factory xml was not the same as yours, I adjusted it for you.

Try the attached xml and let me know if it helps.

Jeff...

[kvom]

As a former programmer, I always try to narrow down the possible causes of a failure.

One thing to try is to add a line such as a M01 before the problem one. If the problem goes away then it's likely a Mach3 or PC timing issue. If not, then it's hardware or electronics.

The next thing I would try is to edit the program to split the G0 in two so that it moves one axis at a time. Or you could try splitting it into two 2-axis moves along the same path. That might tell you if the issue is at the start or end of the path.

[urbanimports02]

jeff, what is the port, and why would it be different? would the port in the factory have worked on any other nm200/cd100? i am just not sure why novakon would not have an xml that works, or is there just something screwy with my machine/controler? i will go give this a try, but i like to know why something does or does not work.

kvom, i will go try the m01 command, i already thought of splitting that line into one axis move at a time, but wasnt sure what that would teach me. i know the stall is at the end of the path in the x axis. it z's up out of a cut and is moving along for a few seconds and then right before it gets to its next destination, it makes a buz and then does its z- move to start next cut. one thing i noticed was that when it does a simultaneous rapid, the feed in mach shows 160ipm or more on the screen, is that normal?

i have never actually done an m01 command, what will happen? will the program stop and require i hit cycle start again like m00, or? i guess i will just do it and report back and then you can elaborate.i really appreciate all the help. i can work around this problem, but fear the next program i do may do the same thing, although so far all the others seem to be just fine. thanks
Jesse

[jalessi]

Jesse,

The port setting is dependent on the parallel port address of the specific computer.

The default xml parallel port address setting will be different depending on which lpt port the computer is assigned to.

Jeff...

[urbanimports02]

well, that xml works now, thanks. however it still has the stall problem on that line(have not tried the m01 or sepperate rapids yet) also, i have not put the rpm meter on it yet, but the spindle does not sound accurate. i hear rpm changes from 100(shouldn't go that low) to 1700, then a jump to 2200 and no further change in sound from there up to 4000. i tweaked mine until all rpm calls were within 10-20rpm at any given random rpm, and i tested lots of different rpm, and was very happy with them. i can mod this xml to have the same settings, or??? i have pwm base freq. at 6(6-9 seemed to be good) with minimum at 10% then in motor tuning i have steps set to 1000 and accel set to 700. they have pwm base at 1800, min at 10%, steps per at 1000 and accel at 1500. once i got mine close, the accel number change seemed to make the fine tune adjustment. i also noticed they now have the kernel speed at 45khz, which is where i have it. i tried 60khz and that seemed to work then i read in the mach tutorial that it should not be set higher then 45khz so i changed it to that. they also have y soft limits set to like 7.6"?? kinda weird for a machine that can go almost double?? i guess they figure you will tweak to your liking(which i have on my xml) anyway, i will keep plugging away, just wanted to post current findings.

[kvom]

A M01 is an optional stop. The program stops only if set in Mach3; otherwise it's a no-op. If it stops, the cycle-start restarts at the line after.

As for the feed rate, it's not that surprising that a 2-axis move would be faster. You're moving along the hypotenuse of a right triangle.