Brian and I worked probably 4 hours last night testing and adjusting and just making general observations as to why the X will not home each time. We set up a test cut that makes the X go back and fourth in short 2 inch movements so we can observe it's movement. It starts out ok but progressively gets farther and farther past the marker I put on the pully and a static point on the Phlatprinter top. The program comes to an end and it consistantly stops just past the marker about an 1/8 to a 1/4 ". I even isolated the X to work by itself thinking it was starving for voltage and amperage but it does the exact same thing. This is the most frustrating thing that has ever happened. Everything is glued tight and absolutely no slippage. Anybody at all have an idea? I am at my wits end here

 

Forgot to say I even reinstalled Mach

 

try changing out the steppers

 

Hey Jeff. That is the next move. I just bought that one couple weeks ago and it is the exact same one as on my Y axis. That stinks if it is a bad stepper or control board.

 

There have been more than a couple threads in the last few weeks where one axis is overshooting by 1/8" - 1/4". Have you guys done any windows updates lately that might be causing a problem on the parallel port? What version of Mach are you using?

 

Don't go changing your steppers out just yet!! I hit this one with the bare board and the steppers on thier own. I think if you check the frequency that Mach is running at you'll cure your problem.

I had it set up to run at a higher core frequency because I though the PC was up to it... turns out that when you do set it like that it freaks and does this overshoot thing. There was something about this over on the Mach forum but I can't find it at the moment.

Try setting it back to the default setting.

Best Regards,
Julian (Fozzy The Bear)

 

Mark, I remember you had a creeping Z axis once, when you were doing a live carving demo. You thought it was a setting in Mach3, that you just changed. Did you figure out what that was?

Fozzy, Randy (rjarois) also had the same problem after increasing the kernal frequency:
viewtopic.php?f=105&t=635
But I am most amazed that you checked repeatability with bare motors.

-Kwok

 

What can I say... I was waiting for the rest of the parts to turn up. and Yeah! I had flags on them and checked them to specific numbers of rotations by running really low feed rates and counting.... What can I say apart from I'm a bit obsessive... LOL I wanted to see what effect changing the kernel frequency had on the output. I was also doing torque tests to see what happens when running at full steps as opposed to 1/8th steps. see what I mean about obsessive!

Anyway! as I was saying.... if he isn't running at the base kernel speed it's worth turning it down to the minimum, because I observed exactly this effect when I turned it up.

Best Regards,
Julian (Fozzy The Bear)

 

This is what I got when I tested in the mach3 tester. I also went ahead and grabbed the bigger 10 amp board from easy-cnc. I really like that idle feature on it. I tested the x motor on the y and it worked perfect. I really think I might have messed up the X controller on the board. Just odd that Y and Z work perfect no matter what I mess with.

 

This is what I got with my desktop and dell sx260. :



Attached files

 

hmmm wonder where all that noise is coming from on mine?

 

Randy, I don't think that's line noise. It should be the delay of the pc in sending pulses to the parallel port? Both my pc don't have the controller connected, if that is making any difference. Frank can disconnect his parallel cable, to see if there is a change.

I was trying to understand the meaning of the data, so looked in the Mach3 pdf for info. And the picture they have of the test screen is totally different. The interesting thing is they now have a rating, so the meaning is clear. So I installed the latest version 3.042.029 in a different folder, to try the newer test. I don't know if you can have two versions installed, but it seems to be ok. Or if anyone wants to try the new "DriverText.exe", I will upload it. Use it at you own risk. Be sure to save your config file, "Mach3Mill.xml", just in case it screws up something and you need to reinstall. Here is a picture of the new test, running:



Attached files DriverTest (try at you own risk).zip (149.2 KB)Â

 

Kwok - I tried your new DriverTest file, and nothing happens, meaning it never starts or burps up an error page or anything. What gives?
I see the "description" of the file says "Mach1 OCX and Driver loader" - Does that mean it's for a MAC?

 

Dorsal, I am running winxp, so you can disregard the description. Since you are having problems, I just checked if it would work on my dell sx260. The sx260 only has Mach3 v2.6 installed, instead of the 2 version of Mach on my main desktop pc. And it worked fine. It just opens up the window, when you double click it. Since it's generally run from the Mach3 folder, I also checked if it can be run from the window's desktop and yes it works that way, too. So I got no idea why it's not working for you? Lets see what happens, when someone else tries it. Here is the results ,just now, from the sx260:

Attached files

 

Try changing the parallel port in the bio to something else, i forget all the different types you can try at least in my bios. maybe also try a parallel port card that is cheap to buy (if they still make them)

 

I have the Intel Core 2 duo CPU
E6750 @ 2.66 Ghz
2,67 GHz and 2 gigs of ram

Windows XP home ver 2002 SP3

 

It is a home built system. I do a lot of Flight sim flying and need a lot of power and the video card is the Geforce 880 512

 

I don't have a mechanical problem it is definitly a controller issue. I ordered a new board by the way from easy-cnc and I got an email saying they are completely out of stock so i got a refund. Going to look at some other boards and see what I can come up with. argh.... Ok I just ordered the hobbycnc 3 axis Pro board kit. Lot cheaper than the 10 amp unit from easy-cnc