Printable View
Hi All
Got back to playing with my cnc mill. Tried to do some engraving but as the engraving proceeds accross the x axis the cutting depth gets greater so I assume the z axis is loosing steps.
How is one supposed to test for this, I think I should try a dial gauge.
Peter
That sounds to me like the surface isn't parallel to the X-axis. Does the cutting depth get less when moving the other way around (just an obvious thing to check!)
You could also try locking the Z-axis somehow (so it won't move).
Problem is if you lock the z-axis you wont be able to do engraving!
peter
Yes, but you can check if it moves up and down - it's only for testing.
Reduce the acceleration what is happening is that the Z axis is losing steps on the way up so when it lowers it goes deeper into the work.
May pay to also reduce the microstepping as more steps, less torque.
.
There is a difference between up and down.
Up can be the start of two consecutive G0's with a 180 degree turn in the middle.
A sort of worst case scenario for losing steps if you're accelerations aren't quite up to scratch.
so what should the acceleration be!
I thought more steps was better, obviously not.
Peter
Unfortunately no one can tell you as it's all machine dependant.
You need to alter the settings and try it under power.
Replace the tool for something with a flat end [ reverse the cutter ? ] and then get something like a 1/2" roller or just a piece of short scrap.
Put that under the 'cutter' and zero onto it. Remove roller and move head up and down a few times and see if the roller will roll back under when it's in the down position.
This will give you a bit of leeway for it to move without clouting into any work.
.
Got back to my X1 mill and tried JohnS suggestion to reduce the acceleration. Tried a search on the web, not a lot of help so just winged it. Also tried JohnS suggestion about the spaced under the z-axis and it definately showed it losing steps on the up direction.
Thought some more and have now tried a dial gauge, and old one I have had for some time and never found a use for in that it works by pulling up the plunger. So I fixed the plunger rod to a chuck in the z-axis and proceeded to raise and lower the head, could not detect that much change but it did alter over 10 cycles, ie gets lower.
I then thought about the gibs and tried to adjust them, this has made it worse.
I also noticed another strange thing in that if I jog up or down using the slowest jog rate the dial gauge shows the head to move in the same direction for a few steps after I have changed direction?
I have read on the web that some have had problems with the head gibs saying that are difficult to set due the weight of it. I may now have to consider a counter weight system as for what I want the mill for a poor z-axis is next to useless.
Peter
Peter,
How are you driving the Z-axis and what is the drive to the quill motion, a worm and rack or pinion and rack?
Putting in a counterbalance arrangement will aid in supporting the weight of the quill etc, but at the expense of adding more mass to move = lower acceleration for the same motor torque
Direct drive to ball screw moving the whole head.
Peter
Well ballscrews work both ways because of their efficiency.. so the inertia of the head is driving the motor and overcoming the motor torque. Its not the weight per se (although obviously the inertia is related to the mass of the head) but the overall inertia of the system - head, ballscrew and motor rotor that is the problem.
As a first indication what happens if you program some G-code to move the head down... stop for a while then move it up again... as opposed to changing direction on the fly... does it actually stop on cue, or overrun... if the latter there may be a solution in increasing the deceleration time or reducing the z-speed... which might suggest gearing the motor to the ballscrew to reduce the reflected inertia on the motor...
Not sure if overrun is a problem but the problem is more apparent if the program I wrote is just run, in step mode the error is small.
I have often wondered if the controller will allow for the steps to be carried out.
All in all I am surprised that this all works as in to get the max power from the steppers you have to step slowly and therefore max moving speed will suffer
Peter
yes, steppers lose power the faster they go, but this is countered by using higher volts. but z-motion rarely needs high speed, so can afford to gear down and get better results that way
Have you tried turning the screw and stalling the motor by hand?
That tells you what you're asking it to do and what it has to offer.
The problem may become instantly apparent :whistling:
Going to try a more powerful stepper a 3Nm one against the ones I have fitted 1.8 Nm.
Looking for others I found a nice little one on Arc euro a 3.5Nm but noticed something a bit odd in the datasheet, they are quoting 3.5Nm holding for parallel and 5Nm for series most other datasheets show the same rating for both. Is this correct or a miss print.
peter
Finally got it to work without lost steps on the Z-axis.
Changed the motor on the Z for a 3Nm and after some adjustments to acceleration etc it now does not lose steps.
I am now not sure if I have its settings correct as it does not get as hot as the X and Y ones do.
This cnc lark is like waiding through treacle, just as you think you understand it another comment is made that makes you doubt it will ever be understood!
I have read that the normal working temp of steppers is around 80c but not sure!
Peter
bigger motor, more thermal mass, working less hard - likely to be cooler...
sorry to say but most reliable answer to this prob is to get a bigger stepping mottor for that axis
which is what Peter did. however bigger motor isnt always the solution to losing steps... could just have been too high an acceleration rate, or he could have supported the motor with a spring or pulley/mass system to balance up and down movements
As Irving said I did change for a more powerful stepper and also changed to parallel drive.
I did try all the options for lower acceleration but they all made no difference that I could detect.
I may in the future try and balance the head but at the moment I am thinking more in the way off removing some of the flex in the column, I am also trying to work out the options of changeing the way the Y-axis works by siting the stepper to the rear and fixing in place instead of moving with the Y-axis. aka KX1.
My biggest headache with the rear siting is the head tilting mechanism I may get rid if it altogether and make a new support for the Z-axis arm, just need a great lump of metal!
Peter
No real substitute for a great lump of metal :naughty:
...but I am seriously considering filling my round column with concrete.
Rather a lot of concrete reinforcement bars have mysteriously appeared on my steel racks, probably enough to fill it. Hey, it was cheap :heehee:
Wonder what the optimum concrete to iron ratio would be for stiffening a round colum :whistling:
I have also been pondering on adding threaded rod to add tension to the column but I have not seen this done anywhere or at least not reported on the web.
I really must test this to see if it has any effect, I know it has effects upon brick walls and concrete makes them stiffer!
Peter
I helped a friend about a year ago with a similar problem, we installed some constant force springs to counterbalance the Z-Axis, we did it very "empirically" and weighed the Z Axis and installed two springs, but only connected one for 2 Kg. as that is what it weighed, but as he wants to eventually change to a bigger router, we just left the other one unconnected for the moment!!
Now his Z-Axis is just about neutral we believe, or fairly close.....
The springs were wound onto ball bearings with "side cheeks", sorry I do not have fotos and he has since sold the machgine, but it worked just fine after that.....
The constant force springs (I am told) always exert the same force, they do not increase as they get longer as normal springs do, which is a big advantage,,,,,
You can see them and read a good explanation here:-
http://www.leespring.com/int_learn_c...ce_springs.asp
originally I wanted to use a counterbalance, but the springs were simpler and added far less mass......