Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: ChuckAustralia on July 26, 2019, 04:08:42 AM
-
Hi guys,
I'm trying to machine a relief designed with Fusion 360 derived from a jpeg, what I'm finding is the tool continually goes deeper into the job, where it should be carving at a maximum of 3mm its cutting down to more like 17mm and not much of a profile to be seen.
I'm using a small Chinese cnc machine and I don't have any trouble engraving shapes and text.
I recently bought a mach1 license hoping that would make a difference but that only allowed more lines to be processed.
I don't know if its the CNC or mach3 settings at fault
I have attached what I think is the config xml, the Confcheck text and the item .tap file
Thanks for looking at this for me
Regards Chuck
-
Hi Chuck,
Welcome.
Your Gcode and Mach3 set-up seems to run just fine here with a maximum Z depth of -5.848 mm in your test Gcode file. If your cutter is going deeper than that then two possibilities spring to mind...
1) Your cutter could be loose in the collet and it is moving during the cutting operation.
2) You could be loosing steps on the +Z axis moves - try reducing the Z axis Velocity and Acceleration (in Motor Tuning) by 50% and save the changes - does it still cut too deep ?
Tweakie.
Another thought comes to mind - is your spindle motor tight in it's clamp ?
-
Thanks Tweakie,
Both the cutter and Spindle are tight, I can actually watch the Z axis (spindle) move down while in operation, it almost bottomed out onto the work piece.
I will try reducing the Velocity and Acceleration and let you know how it went.
Cheers Chuck
-
Finally got a chance to reduce Z axis Velocity and Acceleration, but still cutting too deep, the cnc seems to be cutting at an ever increasing depth each row, starting at approx. 2mm to 13mm halfway through the job where I cancel.
Could it be the control board for the cnc not understanding the commands, although I can engrave and cut shapes without any problems.
Could there be any differences between .tap .nc files that could be causing this?
There are no problems when run through a simulator.
Thanks Chuck
-
what does the z-axis DRO Show, allways the same value or is it also increasing?
-
Does you Z axis have weight - ie. could a counterbalance or gas strut reduce the mass when it is travelling in an upward direction ?
-
what does the z-axis DRO Show, allways the same value or is it also increasing?
It shows the same values as the gcode
-
Does you Z axis have weight - ie. could a counterbalance or gas strut reduce the mass when it is travelling in an upward direction ?
I'm at a loss with this Tweakie, I would have thought to remove weight if that would be the answer, there is only the weight of the spindle and minimal weight from the Z axis casting.
-
so if the Z-DRO Shows correct value, there are IMHO following possibilities:
-z-axis machanical loose
-z-axis loosing step's during Lifting
-additional steps are indicated by electric noise
for a test you can run following simple GCode
M98 P123 L100
M30
O123
G90 G1 Z-5 F250
G1 Z0
M99
%
it will only run Z-axis a hundred times to -5 and 0 (you can stop it any time).
here you will see wether your z-axis moving down or not when it is running "standalone".
if not i would run the same test with spindle turned on (electrical noise).
and then with X/Y movement added.
-
Thanks TPS,
I ran the code with and without the spindle running and it didn't go any further than 5mm (not up on programming X/Y movement)
Do you think it could be due to computer power 2ghz processor, 3gb memory, Vista operating system.
I don't have any problems with simple tasks such as engraving and cutting shapes
-
maybe it can be Computer power, you can try to run the program and turn toolpath off (diagnostic page).
-
Does you Z axis have weight - ie. could a counterbalance or gas strut reduce the mass when it is travelling in an upward direction ?
Hi Tweakie,
Sorry I didn't understand what you were saying at first, I do now and will look into it, I may be able to attach a spring somewhere
Thanks Chuck
-
Hi Chuck,
Stepper motors torque reduces with speed so heavy loads are best lifted slowly. If they loose steps on the way up then, on the downward travel, the axis will move to a lower position and this is the problem I think you are experiencing.
Reducing Velocity and Acceleration is usually the answer but if you can attach a spring (or better still a gas strut) to reduce the weight the stepper has to lift then it may help.
Tweakie.
-
I tried fitting an extension spring to the Z axis with about 5 kg strength, unfortunately it didn't make any difference, this was using a profile that only cut .2 to .5 mm. it still cut at an ever increasing depth.
I have read that someone having the same experience applied more power to the stepper motor and that fixed the problem, what would your thoughts be on that?
Thanks
Chuck
-
have you tryed to run Programm with toolpath turned off (see below)
-
have you tryed to run Programm with toolpath turned off (see below)
No not yet, I will try tomorrow
One thing I did find though was when checking motor tuning settings, the pulse settings were at 0 for all axis so I set them at 5 and saved, after closing that window and reopening, they were all at 0 again, does that sound normal?
-
I have read that someone having the same experience applied more power to the stepper motor and that fixed the problem, what would your thoughts be on that?
Hi Chuck
Stepper motors should run hot to the touch (but below 60deg.C) if your Z axis stepper is running cool then you could try increasing the current by one increment.
Ideally, stepper motors have their rated current printed on a label and that is the current that should be set at the driver - a stepper motors rated current should not be exceeded.
Tweakie.
-
have you tryed to run Programm with toolpath turned off (see below)
I tried running with toolpath turned off, unfortunately no change
-
Im using a small Chinese cnc machine and I don't have any trouble engraving shapes and text.
May be true, but when doing 3 axis moves with numerous small axis moves, you really need
to have steppers that can provide the accel necessary. The stepper can skip and you would
not even know it, the dro will show correctly but the actual postition will be wrong. the skipping
may be repeatable or ramdom depending on the gcode your doing.
From reply #8, you ran the following code and there was no Z problem.
You tested for a Z move of 5 mm ( .197")
M98 P123 L100
M30
O123
G90 G1 Z-5 F250
G1 Z0
M99
%
A look at the gcode you posted shows small Z moves approx range from
say .003mm to .050 ( .00011" to .002" ) for example.
I would suggest you see what the machine can do at the feedrate your calling,BUT,
test at Z distances of 4,3,2,1,.5,.1 mm's etc. You may want to just jog those distances and see
what a indicator shows. You will not move .0001" accurately as that is theoretical resolution
of your axis. If you have backlash, and you do, machining of small moves accurately just dosen't work well!
Test with only the Z, down only, then down and up and down movement and if that test is good,
add X & Y moves and test again. BTW, may want to add a pause using G4 so you can do a measurement
post move. Let us know how things go.
RICH
-
Hi guys,
Not having much luck with settings etc. so I thought I would upgrade stepper drives to TB6600 seeing that one of my originals died.
After doing that I face the dilemma of getting them to work, I had no problems getting the "A" Axis to turn, but the others only buzz.
The originals dips where all set the same so I set all the new ones the same, but no that wasn't the answer, I tried all available combinations with no luck.
I tried changing the motor profiles to that of the "A" Axis in Mach3, but no change.
Any Ideas?
I have attached pics of the original and TB6600 drivers to give an idea of how Ive wired them up.
Thanks Chuck
-
Found the problem, the cagey buggers swapped motor wires, black and blue, that's why the "A" axis was working but the others not and I had wired all according to the "A" axis.
A trap for "young" players I guess, all working now, will do a test cut later.
Cheers Chuck
-
Hi guy's,
Carried out a test cut and all is good, thanks for everyone's advice, but it seems the problem was down to inferior stepper drivers.
Cheers Chuck