Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: andythepandy93 on November 26, 2015, 11:27:42 AM
-
Hi all,
I came across an odd problem with my machine today. When I'm both running some Gcode or jogging, the steppers sound a bit dodgy and it causes the whole machine to vibrate and make a high pitch almost frictiony sound. However as soon as I rotate the toolpath it stops and the steppers and machine sound very smooth and happy again.
Has anyone got any ideas on what could be causing this?
(Also this happens at whatever maximum speed I set for the motors in motor tuning)
Thanks,
Andy
Machine details:
2.2Nm Nema23 steppers. Drivers set to 5 mircosteps
Running Mach3 on
Windows XP, Intel Pentium 3Ghz, 2GB RAM
-
can you post your xml file, you will need to call it something unique like andythepandy93.xml...
-
Sure, I'm pretty new to all this so don't be too surprised if there's a stupid mistake in there... ;D
Also forgot to say, using the 2010 screenset
Thanks
-
Couple of questions (probably not creating your issue... but worth a consideration / change or curious question....)
Config >>> Ports & Pins >>> Inputs tab....
Why have yo got Limits Over-Ride input enabled... without a pin ref, with a hotkey
(what are you trying to do as there may be another way?)
Config >>> Ports & Pins >>>Spindle Setup tab (Left Hand column)
Why have you got the CW spindle setup as Output 2.... and CCW spindle as Output 1..... but ou have defined only one spindle output?
Config >>> Motor Tuning >>> All Axis
Why have you set the acceleration at 150mm/s/sec .... and the limiting velocity at 1000mm/s/min (16.667mm/s)
Have you tried to lower the acceleration ... maybe say start at something like 50mm/s/s and then increase the machine speed up to 3000mm/min (50mm/sec)
You are running 5mm pitch ballscrews I presume? depending upon your setup voltage it can "probably" handle 50mm/sec max velocity..... but that acceleration seems a little quick to me (but I don't know your setup enough! [moving mass / weight etc of gantry]).
Config >>> General Config >>>
change the lookahead to 200 (2nd column from left)
un-check the watchdog checkbox (1st column from left)
I would suggest changing your job increments (given your max resolution at 200 steps / mm = 1/200 or 0.005mm/step... hence something like
10, 5, 1, 0.5, 0.1. would probably be more workable (or whatever else you choose... all your choice!)
I would suggest checking a few of the checkboxes within the Right hand "Pgm End, M30 or Rewind Box"
(personal preference this one) I would suggest changing the initialisation string (centre top of tab / screen) to something like "G80, F50" and checking the checkbox to enabled the initialisation string on all resets.... the reasoning being is the "F50" will set the feedrate to 50mm/sec... and not the Mach3 std..... note the one consequence of this is if you try to restart a gcode it will have reset the feedrate to the G0 feedrate..... (but your jog will work quicker....)
What voltage are you running your stepper drivers at?
Rob
-
Rob,
Thanks for such a detailed reply.
Why have yo got Limits Over-Ride input enabled... without a pin ref, with a hotkey
(what are you trying to do as there may be another way?)
My understanding of the limit override is that when the machine hits the limit switch (embarrassingly happens a lot at the moment!) when you press this switch it will disable teh limit switches so you can jog the machine back (is this correct?). I don't have a physical switch to do this so I have assigned the key next to the number 1 key to do this job instead.
Why have you got the CW spindle setup as Output 2.... and CCW spindle as Output 1..... but ou have defined only one spindle output?
I do not use spindle relays as I have a VFD which is controlled over serial. I believe I need the 'Disable Spindle Relays' tickbox unchecked though for my spindle to work. I don't recall changing the Output #'s for the CW and CCW.
Why have you set the acceleration at 150mm/s/sec .... and the limiting velocity at 1000mm/s/min (16.667mm/s)
Have you tried to lower the acceleration ... maybe say start at something like 50mm/s/s and then increase the machine speed up to 3000mm/min (50mm/sec)
You are running 5mm pitch ballscrews I presume? depending upon your setup voltage it can "probably" handle 50mm/sec max velocity..... but that acceleration seems a little quick to me (but I don't know your setup enough! [moving mass / weight etc of gantry]).
That is a very good question. I guess I was stupidly plugging in numbers and not really thinking about their actual meaning. Yes it's a 5mm pitch ballscrew. I'll try the numbers you suggest now
change the lookahead to 200 (2nd column from left)
un-check the watchdog checkbox (1st column from left)
I would suggest changing your job increments (given your max resolution at 200 steps / mm = 1/200 or 0.005mm/step... hence something like
10, 5, 1, 0.5, 0.1. would probably be more workable (or whatever else you choose... all your choice!)
Okay will do, could you briefly explain what these settings do?
The drivers are up to 50v but I have a 36v power supply hooked up.
After changing all those and restarting it still sounds the same. It's weird because if I hadn't noticed that the problem went away when I rotated the toolpath display, I would totally be blaming the bearings as it really sounds like they're making graunching noises but as soon as I move the toolpath it sounds perfect... ???
Many thanks
-
Have you been through the wiring to check its all wired up correctly?
What sort of power supply is it using (switched mode or transformer)... Some smps require a minimum current draw before they regulate corrrctly
What stepper drives are you using?
Do you have a model number for the stepper motors?
Are they correctly wired as 4, 6 or 8 wire steppers?
Is the peak current setting correct on the drives (not RMS!)
Is this via a PC or laptop (some laptops apparently don't output the correct TTL 5v signals on the parallel port)
Just guessing at what it may be now
-
Switched mode PSU (found here https://www.cnc4you.co.uk/400W-PSU-36Volt-11Amp?search=400W36V%2F11A)
These stepper drivers: https://www.cnc4you.co.uk/Stepper-Motor-Driver-4.5A,-50V-CNC-Microstepping-CW5045?search=CW5045
Stepper motors are Nema23s 2.2Nm, wired up as 4 wire.
The drives have a 8 DIP switch which configures the current, half current or full current and microstepping. I have them set to 4.2A, full current, and 5 microsteps.
It's a PC.
-
Lookahead and various other bits are covered in this info
https://www.google.co.uk/url?sa=t&source=web&rct=j&url=http://www.machsupport.com/wp-content/uploads/2013/02/Mach3_CVSettings_v2.pdf&ved=0ahUKEwiA9pbGk7HJAhVrp3IKHfp8CHIQFggbMAA&usg=AFQjCNGWrcUnJ8turvOQaSeQmA9w-aREBA&sig2=LGeyXwTdLLFcd26O98MkMQ
-
As for the wiring, I followed the guide provided by cnc4you and have checked and it matches that.
Didn't say but this happens on all three axes' steppers
-
Nothing wrong with their stuff (I have their steppers...)
Can I suggest trying them on half current setting
I'll be back... On a phone... Sorry drip feed
-
Wiring guide I followed here: http://cnc4you.co.uk/resources/CW5045%20Wiring%20Diagram.pdf
-
Okay just tried on half current, didn't seem to make a difference at all
-
I took this video to demonstrate. You can't hear all the bad noises but you can tell that something doesn't sound smooth. The first two runs are without rotating the toolpath and the second two are while rotating the toolpath (when I put the camera down)
https://vid.me/6Y0f
-
A few more shot in the dark suggestions...
What version of mach3 are you using... The recommended one is ....0.62... To replace your existing one id suggest copying your profile XML file to your desktop, deleting the mach3 folder and installing 0.62, then copying your XML file back into the mach3 folder and everything should be back as it was fingers crossed without the problem...
0.62 is available in the m3 ftp site
Have you tried another PC (laptop) or anything... I was going to look into printer ports and the BIOS as I thought that at one time you could select different printer port types in the BIOS... Not sure what mach3 prefers...
Another PC should split the system in two... Same problem... Then its a bob onwards hardware fault... Problem gone it's a PC fault...
Rob
-
ftp://anonymous:guest@machsupport.com/Mach3/
-
One to consider, you are not alone (mach4 though)
http://www.machsupport.com/forum/index.php/topic,31357.0.html
-
Okay great, I'll look in the morning when I'm back with the machine.
Cheers,
Andy
-
I believe Mach prefers the EPP setting in the BIOS.
-
I'm on 0.66 version. The link you posted doesn't work but I'll have a look for another now.
Unfortunately I don't have another computer with a parallel port so testing with another PC is not really possible.
I'll have a look into the EPP Bios settings
-
Try this one... Need to go to the ftp downloads tab on the RHS, mid screen, then navigate to mach3
http://www.machsupport.com/software/downloads-updates/
-
Okay I have:
Reinstalled to 0.62 - no change.
Changed the parallel port to EPP in BIOS - no change
I'll keep an eye on that other thread
Really banging my head against the wall with this!
Thanks for the help
-
No problem sorry I could not help.
If you have a look in the mach3 installation and setup manual, it details a programme within the mach3 folder that is used to test the speed of a machine, it may be worth running it just to see what comes back.
It doesn't do any loop tests or anything on the parallel port but it may be worth a check.
Presume you've cleaned out all your temp files with something like ccleaner so that is done.
-
Are you referring to the DriverTest.exe? It is in the Mach3 folder.
-
Think that's the one its suppose to test the capabilities of the PC and printer port for clock cycle / loop load.
(Never used it... I run a smooth stepper, if you just had one basic machine I'd recommend a uc100)
-
When I run the DriverTest.exe file I get a "System Excellent" with only ever 5 spikes on the line max.
-
Oddly enough I always get a system excellent result. Even on a computer that make my steppers stall.
I just posted a thread in regards to that test and what to make of it.
Sent from my iPhone using Tapatalk
-
Hi andythepandy93,
I can only think of one explanation of the problem you report. The odd/poor sounding of the steppers must come from losing steps or
velocity/accelerations set higher than the steppers can physically handle, maybe heavy load on the table or similar.
When you rotate the axes now both steppers are operating and the load is shared between them. If you rotated by 45 degrees they would
share equally. Ten steps in X and ten steps in Y results in 14.4 'steps' along the new axis...
Could you try an axis rotation of 90 degrees so effectively x becomes y and y becomes x. If my theory about load sharing is correct then
your reported problem should recur. Would give some proof without to much shagging around.
Craig
-
When I say rotate the axes, I mean rotate the 3D view of the gcode that is displayed in the program, not actually changing the coordinate axes of the model I'm cutting.
So when I jog in just the X axes, it makes a rough sound until I leftclick with the mouse and rotate the toolpath when it becomes smooths. As soon as I let go it's rough again
-
How much ram have you got?
-
2x1GB sticks
-
This problem seems very backwards, if the steppers are missing steps then surely when I rotate the toolpath there is even less time given to the drivers and surely should get worse and not better?
One thing I have noticed is that when you are rotating, the DROs don't update. In fact when the DROs are updating the steppers are noisy and when they aren't updating they are fine. Could this be anything to do with it?