Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: dougie329 on June 11, 2007, 04:26:33 PM
-
A problem with the Z axis is driving me totally mad. I have a 3 axis cnc router and after 5 or 10 minutes of running a program the Z axis stalls when it tries to make a +ve move and just makes a buzzing noise ie no steps. I let it be for half an hour it works and then repeats the problem... I have tried everything (no real help from HEIZ cnc in Germany) and even put a bigger motor on to no avail. It is worse in hot weather??
I have it running through a pci parallel port on my cpu and after checking the pins directly out of the computer port for voltage here is what I get:
X axis pin 2 direction 5v
X axis pin 3 step 5v
Y axis pin 4 direction 5v
Y axis pin 5 step 5v
Z axis pin 6 direction 5v
Z axis pin 7 step 0v ??
I cant understand why I get 0v on step and the motor still goes, is this the problem? pin 1 is spare and is 5v but I am not sure how to swap everything.
Any help would be much appreciated :)
-
It is probably because you have the step low active checked for X and Y axis that you are seeing 5V on these pins. I think your problem is likely to be the drive for the Z Axis. Are the drives for each axis seperate drives? if yes then would be a good idea to swap two drives around and see if the axis problems follow the drive.
Hood
-
x axis is active low, other 2 arent.
I just swapped the y and z axis drives around in the control box and it doesnt do it anymore? problem shifts.. Could it be wired incorectly in the control box for some reason?
It does have seperately wired drives inside the controller, 2 for the x motors and 1 each for the Z. Funny thing is the x and y are wired the same colours in order but the z is not wired in the same order. I dont want to blow the thing up as this is how it came!!
On output signals in diagnostics enable 1 has a red light on, not sure if this is significant..
:)
-
x axis is active low, other 2 arent.
That is why you see 5v on two but not the other. If you look closely you will see 5V on them when stationary but if you Jog the axis you will actually see it falling slightly. Because it is a pulse and your meter will not be able to keep up you dont actually see the voltage dropping to 0V. Same goes for the other axis, stationary it will be 0v but as you Jog it will go to 5V on each pulse but again your meter will not be able to keep up so you will see just above 0V on it. If you put a scope on these pins you would see what is really happening.
I just swapped the y and z axis drives around in the control box and it doesnt do it anymore? problem shifts.. Could it be wired incorectly in the control box for some reason?
Not really sure what you are meaning. Have you physically moved the wiring from one drive to another? If yes did the problem move from the previously bad axis to another?
It does have seperately wired drives inside the controller, 2 for the x motors and 1 each for the Z. Funny thing is the x and y are wired the same colours in order but the z is not wired in the same order. I dont want to blow the thing up as this is how it came!!
Dont mess with the wiring unless you can work out where each wire is from and where it should go. This is especially true if the axis worked properly before.
On output signals in diagnostics enable 1 has a red light on, not sure if this is significant..
Thats normal
Hood
-
I took the y drive plug out of the controller and swapped it with the Z plug on the front of the control box and then jogged z using what would normally be the y quick keys. problem didnt occur in z but does now appear in y. I didnt physically shift any wires inside ???
can I tell what definately needs to be where using a voltmeter?
http://www.cnc-router-routers.com/
Controller for S-400, S-720 and S-1000:
Appropriate 1/8 Microstep Controller 4 channel:
4 channel controller
Power up to 2,1A per Phase
4 connections per motor, 1 connection for. 4 end or emergencu switches
connection to PC-parallelport
Microstep controller 1/8 Step
„Sleep-Function“ for energy saving in standby mode
Safety device for high overvoltage points
Power pack
ALU cooling device / 2 ventilators, 80 and 50 .
Option to connect 5 different end or reference switches, and emergency switch
All connections for motors and endswitch - 9 Pol sub D
-
OK so does the wire you shifted go to the motor? Next thing is to keep the wire shifted and go into config menu, Ports and pins then motor outputs and swap the pin numbers around for the axis you have changed. Which axis faults now?
Hood
-
If it works fine for 1/2 hour and then gives you trouble, and the problem is worse when hot weather,
it sounds like something is overheating. A bigger motor would not help if it is the driver overheating.
How is the ventilation in the driver box? At least temporarily hook up a fan blowing on the drivers and see if it helps.
Bertho
-
still having big problems with this >:( so if anyone has anymore ideas I can try as the machine is pretty worthless as a production item operating like this..
ok heres what I have done since:
*Got a higher rated Z axis motor-didnt fix the problem.
*swapped z and y drives over and still does it.
*contacted manufacturer and they reckon its some software config issue????? any ideas. The standard software is PCNC which is not very good to me so I havnt tried it but they reckon its perfect with that... IM STUMPED
:) ooh and I did try taking putting a cooling fan on drivers and no luck!!!
-
what are you using for a power supply for your motors. Used together, the motors can take 7.5 amps, and if you are putting bigger ones on, then even more. A power unit to supply 24 volts at 7.5 amps is quite a big ****, and after a while, if it is not up to the job it will give in, or the voltage outpput will fall and cause you steppers to miss.
To try it out get the machine up and running a test the voltage out of the power supply when running a G0 x1 y1 z1 or something similar that tests all three motors running at once.
-
this is the controller info belowfor the cnc machine. I dont run a seperate power supply just through the CPU. The problem is always the Z axis and usually happens when it tries to retract upwards at the end of a cut so not necassarily when using all 3 axis at same time. I have 2 controllers and they are both the same so it cant be a faulty one.
Controller for S-400, S-720 and S-1000:
Appropriate 1/8 Microstep Controller 4 channel:
4 channel controller
Power up to 2,1A per Phase
4 connections per motor, 1 connection for. 4 end or emergencu switches
connection to PC-parallelport
Microstep controller 1/8 Step
„Sleep-Function“ for energy saving in standby mode
Safety device for high overvoltage points
Power pack
ALU cooling device / 2 ventilators, 80 and 50 .
Option to connect 5 different end or reference switches, and emergency switch
All connections for motors and endswitch - 9 Pol sub D.
-
Hi,
I think you should halve the velocity and double the acceleration figures for all the axis and let it run a very long program, see if it faults, it may be right on the limit.
Graham.
-
tried playing with the velocity and accelaration, even with the Z axis on 20mm/min and slow accel it still does it.
Weird thing is I let it be switched off for half an hour or so then I get 5 or 10 mins use and it buggers up again so unfortunately thats how I have to use it at the moment, its really driving me nuts!!! :P
X and Y axis never miss a beat and I can run them whatever speed I want..
-
Have you had the Z axis in bits to see if there is a jamming condition, if I remember correctly these machines use plain bush slides, is one contaminated?
Graham.
-
I took the y drive plug out of the controller and swapped it with the Z plug on the front of the control box and then jogged z using what would normally be the y quick keys. problem didnt occur in z but does now appear in y. I didnt physically shift any wires inside ???
This proves that the motors were working correctly for both axes, so no need to upgrade those.
The problem occurs between the computer and the controller output.
Based on this it could be the controller board, PC port, or the software (including a software configuration error).
The fact that they state their preferred software works is significant - they probably also provide a config file which sets all the correct software config states up for their machine to operate
x axis is active low, other 2 arent.
X axis pin 2 direction 5v
X axis pin 3 step 5v
Y axis pin 4 direction 5v
Y axis pin 5 step 5v
Z axis pin 6 direction 5v
Z axis pin 7 step 0v ??
Very confusing.
I am wondering if you have mistyped something, and that x & y are active low, and z isn't.
That would make more sense.
If that is the case, try changing the sense for the z axis step control.
Other thing from left field, is the PC doing something after 5 or 10 mins?
eg maybe a memory cleanup 'garbage collection' cycle or swap file flush etc.
I had a similar problem when I first set up my machine.
Trying to mill a test piece it would skip on the z axis at about the same place everytime (the 3D face test file, doing the fine cut about 5 mm into the side about level with the nose)
I finally fixed this by removing the ACPI chip driver and replacing it with the 'Standard PC' driver.
I presume some long delay interrupt triggers or similar, and windows gets sidetracked too long to drive the motors properly.
Now I have replaced the drivers, the machine has never skipped since.
-
I have had the Z axis apart and its smoooooth on the slides :) the only resistance comes from the antibacklash drive nut, its quite tight but then I guess it needs to be...
Thought it might be a voltage issue from parallel port so got a PCI card, no different. Even tried it on two other computers and still the same!!
-
Hi, dougie329
A couple more things to try.
Some stepper controls are sensitive to leading and trailing edge pulses, set your z axis Dir., Step "LowActive" settings in config, Ports & pins, Motor Outputs
to the same as your X, Y. (It's a hardware change)
In Config, Home/Limits, You can change the Z direction if needed in Reversed coll-um for your axis.(It's a software change)
If that doesn't help in Config, Motor Tuning, Increase the Step, Dir Pulse's to 5. (This sends a longer pulse to your axis controls)
Hope this Helps, Chip
-
RE YOUR Z AXIS - What is it that you are trying to drive. - My Z axis or my milling set up is a very large milling head. As it was it was sprung UP and I couldn't get the stepper motor to drive it down. Coming the other way it was fine. Yours is probably the opposite way - gravity helps it going down - but works against it going up (although I think you said you had tried a bigger motor).
I ended up gearing my milling head stepper down 5 to 1. I didn't need the speed, but I needed the power to shift this big lump of metal. Bear in mind if the motor is working hard, then the driver board will be working hard - and cut out if it overloads. There is also a warning on my stepper driver notes that if the voltage on the steppers falls below 12 volts, there is a danger the card will over heat - have you checked you driver voltages - you cannot be running them off the 5 volt computer supply.
You have proved (as far as I can see) by swapping the drivers round, that the fault lies on the pysical side of the equation, ie. the mechanics of the Z axis, not on the electrical (computer) side, so the fault will lie somewhere in this area.
JIm.
-
Dougie what do you mean by "tight" on the Backlash nut? There should be very little resistance to turn the screw. If it is too tighten it may get worse as the friction heats up the components. Try losing the nut until the screw runs free turning by hand. Test the machine, retune the nut if needed. (;-) TP
-
Is the Z axis head counterbalanced?
My X3 has a gas strut to help balance the up and down loads
Large machines have a counterbalance.
It goes down OK, but has trouble going up.
With it all turned off try and measure how different the torque required to turn the Z screw up and down.
As stepper motors get hot, their resistance goes up, and if you don't have a contant current drive circuit, the torque will decrease as the temperature rises. Actually reducing the current limit on the stepper drive may keep it cooler. Z often gets a punishing because it does not move much.
If you do a lot of work where z does not need to move, consider locking the z and turning off the z drive unless you need it to move.
Best still balance the load.
-
Dougie said that when he swapped the y and z axis cables from the controller to the stepper motors, the problem shifted to the y-axis.
That implies the driver electronics or software config - pulse widths, pulse polarity and edge detection.
Or did I mis-read the thread somewhere?
If not there seems little point investigating the mechanics of his machine Z axis when now the movement of the Y-axis begins exhibiting the same problems. Any problems with tight leadscrews/nuts or excessive weight of the Z axis machine head etc are not relevant if the Y axis suddenly starts doing it too.
He also said it did it when jogging around, so the heating affect ought to be minimal.
-
ok heres what I have done since:
*Got a higher rated Z axis motor-didnt fix the problem.
*swapped z and y drives over and still does it.
*contacted manufacturer and they reckon its some software config issue????? any ideas. The standard software is PCNC which is not very good to me so I havnt tried it but they reckon its perfect with that... IM STUMPED
:) ooh and I did try taking putting a cooling fan on drivers and no luck!!!
Seems that he swapped the drives and the problem still exists?????
Don't believe it is software config as it does work when it is cooled down, Software is not heat sinsitive(;-)
Doug please clarify that when you swapped the drives the problem was still on the Zaxis or did it move to the other axis. If it moved to the other axis then I believe you have a drive that fails when it heats up.
(;-) TP
-
Dougie said that when he swapped the y and z axis cables from the controller to the stepper motors, the problem shifted to the x-axis.
That implies the driver electronics or software config - pulse widths, pulse polarity and edge detection.
Or did I mis-read the thread somewhere?
Glad you picked up on that DaveC, the thread seemed to move away from that near the beginning so I assumed I was missing something ;)
Hood