PDA

View Full Version : Z axis moves below set zero



whensparksfly
28-05-2016, 03:59 PM
After I set my z axis to zero in a g code run the Z axis moves below set zero.What am I missing?

Sent from my SM-G361F using Tapatalk

magicniner
28-05-2016, 06:46 PM
The first thing to eliminate is the code itself, if it has a G54....G59 the code is applying an offset which will override your zero.
Best to at least post code to the point where Z drops below your zero,

- Nick

whensparksfly
28-05-2016, 06:52 PM
It dose it in a few codes I've tried to you have any code I can try

Sent from my SM-G361F using Tapatalk

routercnc
28-05-2016, 07:31 PM
In my CAM software I define the top surface of the work as work coordinates Z Zero. Any cuts down into the work are then in the -ve Z direction. So it goes below zero when cutting but that is normal. Is that what you mean?
You can also define the bottom of the workpiece as zero in CAM and then set work Z zero at the bottom of the work (eg the spoil board). Then all cutting is +ve Z

Can you write down each step you are taking as this might help understand your problem? Explain at which point it does something unexpected. Video and screenshots may help if it is difficult to describe

lucan07
28-05-2016, 08:20 PM
With a plasma cutter I believe sheetcam is a popular product to use with Mach3 and looks simple to generate plasma code with

whensparksfly
28-05-2016, 08:21 PM
There seem to be no g54 or g59 how do you set up the z

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:29 PM
What dose the red light at the z meanhttp://uploads.tapatalk-cdn.com/20160528/13fb58081b16e9540fcfd96c1959d9dd.jpg

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:39 PM
http://uploads.tapatalk-cdn.com/20160528/802f237ec5fe912cddb6f0b37216644a.jpgA video

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:40 PM
Somebody delet above sorry

Sent from my SM-G361F using Tapatalk

lucan07
28-05-2016, 08:44 PM
Somebody delet above sorry

Sent from my SM-G361F using Tapatalk

you should be able to edit and delete

magicniner
28-05-2016, 08:50 PM
It dose it in a few codes I've tried to you have any code I can try

Sent from my SM-G361F using Tapatalk

You should post some code here with which the problem occurs

whensparksfly
28-05-2016, 08:51 PM
I tried to load a video the z just keeps moving down past 0 and keeps going on down levels out eventually

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:52 PM
http://uploads.tapatalk-cdn.com/20160528/802f237ec5fe912cddb6f0b37216644a.jpgA video

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:52 PM
http://uploads.tapatalk-cdn.com/20160528/802f237ec5fe912cddb6f0b37216644a.jpgA video

Sent from my SM-G361F using Tapatalk


Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:54 PM
Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:54 PM
[IMG]http://uploads.tapatalk-cdn.com/20160528/802f237ec5fe912cddb6f0b3721664

Sent from my SM-G361F using Tapatalk



Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:55 PM
Hk


Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 08:56 PM
Woops tried to delet made it worse

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 09:08 PM
N0010 (Filename: nigel first flange final.tap)
N0020 (Post processor: Mach3.scpost)
N0030 (Date:24/05/2016 Time:01:13:16)
N0040 G21 (Units: Metric)
N0050 G40 G90 G91.1
N0060 F1
N0070 (Part: nigel first flange)
N0080 (Operation: Outside Offset, 0, T1: Plasma, 0 mm kerf)
N0090 (Plasma, 0 mm kerf)
N0100 T1 M06
N0110* G43 H1
N0120 G00 Z20.0000
N0130 X95.8839 Y95.8838
N0140 Z10.0000
N0150 G01 Z1.500 F100.0
N0160 F440.0
N0170 X94.1162 Y94.1161
N0180 G00 Z20.0000
N0190 Y95.8839
N0200 Z10.0000
N0210 G01 Z1.500 F100.0
N0220 F440.0
N0230 X95.8839 Y94.1162
N0240 G00 Z20.0000
N0250 X95.0000 Y20.0001
N0260 Z10.0000
N0270 G01 Z1.500 F100.0
N0280 G03 X95.0000 Y20.0001 Z1.5000 I-0.0000 J74.9999 F440.0
N0290 G00 Z20.0000
N0300 X94.6339 Y4.9339
N0310 Z10.0000
N0320 G01 Z1.500 F100.0
N0330 G03 X94.6339 Y4.9339 Z1.5000 I0.0000 J4.6999 F440.0
N0340 G00 Z20.0000
N0350 X155.9880 Y29.3121
N0360 Z10.0000
N0370 G01 Z1.500 F100.0
N0380 G03 X155.9880 Y29.3121 Z1.5000 I0.0000 J4.6999 F440.0
N0390 G00 Z20.0000
N0400 X182.1339 Y89.9339
N0410 Z10.0000
N0420 G01 Z1.500 F100.0
N0430 G03 X182.1339 Y89.9339 Z1.5000 I0.0000 J4.6999 F440.0
N0440 G00 Z20.0000
N0450 X157.7557 Y151.2880
N0460 Z10.0000
N0470 G01 Z1.500 F100.0
N0480 G03 X157.7557 Y151.2880 Z1.5000 I0.0000 J4.7000 F440.0
N0490 G00 Z20.0000
N0500 X97.1339 Y177.4339
N0510 Z10.0000
N0520 G01 Z1.500 F100.0
N0530 G03 X97.1339 Y177.4339 Z1.5000 I0.0000 J4.7000 F440.0
N0540 G00 Z20.0000
N0550 X35.7798 Y153.0558
N0560 Z10.0000
N0570 G01 Z1.500 F100.0
N0580 G03 X35.7798 Y153.0558 Z1.5000 I0.0000 J4.7000 F440.0
N0590 G00 Z20.0000
N0600 X9.6339 Y92.4339
N0610 Z10.0000
N0620 G01 Z1.500 F100.0
N0630 G03 X9.6339 Y92.4339 Z1.5000 I0.0000 J4.6999 F440.0
N0640 G00 Z20.0000
N0650 X34.0120 Y31.0799
N0660 Z10.0000
N0670 G01 Z1.500 F100.0
N0680 G03 X34.0120 Y31.0799 Z1.5000 I0.0000 J4.6999 F440.0
N0690 G00 Z20.0000
N0700 X95.0000 Y-0.0000
N0710 Z10.0000
N0720 G01 Z1.500 F100.0
N0730 G02 X95.0000 Y-0.0000 Z1.5000 I-0.0000 J95.0001 F440.0
N0740 G00 Z20.0000
N0750 M05 M30

Sent from my SM-G361F using Tapatalk

whensparksfly
28-05-2016, 09:10 PM
A mental speeds is too high warring just came up

Sent from my SM-G361F using Tapatalk

whensparksfly
30-05-2016, 01:09 AM
When I toggle the z with the arrow keys it moves In inchs and the z and y move in mm

Sent from my SM-G361F using Tapatalk

Clive S
30-05-2016, 08:11 AM
When I toggle the z with the arrow keys it moves In inchs and the z and y move in mm

Sent from my SM-G361F using TapatalkHow do you know it moves in inches? Does the X move in inches as well?

You need to give more info about the settings on your machine etc.

lucan07
30-05-2016, 08:45 AM
If Z is moving around 25 times further than required then the motor tuning for the Z axis is probably incorrect, it is most likely doing exactly what it has been told to do.

Make sure tuning is correct may be totally different to X & Y and be sure to save axis settings every time changes are made, you solved the motor tuning problem with auto calibration in this or previous thread, how are you measuring that movement accurately with much smaller Z axis movement any errors may be larger.

whensparksfly
30-05-2016, 09:51 AM
What setting do you mean I'm new to this

Sent from my SM-G361F using Tapatalk

lucan07
30-05-2016, 10:26 AM
The motor steps per unit in motor tuning.

whensparksfly
30-05-2016, 10:48 AM
I done the automatic motor configuration in the setting menu measured it movement it was correct.
When I toggle the z and y with the arow keys it will work in mm I can be seen that the movment matchs the fro but when z is moved the movement in the dro and a axis will move in inchs

Sent from my SM-G361F using Tapatalk

magicniner
30-05-2016, 11:12 AM
When I toggle the z and y with the arow keys it will work in mm I can be seen that the movment matchs the fro but when z is moved the movement in the dro and a axis will move in inchs

Sent from my SM-G361F using Tapatalk

The only way that can happen is if you've got your motor tuning wrong.
Motor tuning units are set for all axes at once so you can't have a mix.
If tuning units are set to mm and X and Y work fine you will get really close if you take the number for "steps per" currently set for Z and multiply it by 25.4, Mach3 can't tell when you instruct it to do stupid things, it simply does what it's told.

- Nick

lucan07
30-05-2016, 11:35 AM
When I stated in your previous thread to check wiring you responded I was definitely wrong then that you had found the broken wire.
When I stated the code was wrong and would not run in a different thread, you replied I was wrong and later PM'd me to get the edit to correct that code.
Again I try all axis are using same measurement scales they cannot do otherwise, therefore your tuning steps per unit are incorrect they are for mm's if axis is moving in mm and for Inches or out by a factor of 2540% if axis is moving in inches instead of mm.
There is no alternative computers just do what they are told, you put rubbish in you will only get rubbish out its the way it is.

whensparksfly
30-05-2016, 12:32 PM
I started changing the motor tuning rather than the running in the setting.
But when I run the code after setting zero the z just go's down past the materal down to the bed maybe I'm not setting the zero right.

I'm very thankful for all your help don't think I'm not the pm was a personal thank you

Sent from my SM-G361F using Tapatalk

magicniner
30-05-2016, 01:42 PM
out by a factor of 2540%

Can you show your working for the maths that give you this result? ;-)

lucan07
30-05-2016, 01:52 PM
Can you show your working for the maths that give you this result? ;-)

1 inch approx 25.4 mm so if moving 25.4 to 1 = 2540%

JAZZCNC
30-05-2016, 02:44 PM
Your trying to run before can walk.??

First job before trying to run any G-code is to make sure the machine is setup correctly and know how mach3 works. Ie It's fundimental thing to know why Dro is Red or Green. Means the Axis as been homed or pushed ref all home if no home switches.
I suggets you read the manual and learn Mach basics before trying to run G-code.

It's impossible to Mix Inch and MM in same setup without using a scaling factor. So if this is definatley happening then highly likely your Steps per setting is wrong.

To help you further you need to give more detailed info about your system. Your one line cryptic replies are pretty useless.

Tell us the Micro steps set on drives and the pitch of the ballscrew and if any ratio using timing belts etc and will give you the exact steps per setting. This will remove any guess work on your part.

When the Steps Per setting is correct then any error is either from the G-code or Mechanical. Mach3 just does exactly what it's told.

JAZZCNC
30-05-2016, 02:48 PM
1 inch approx 25.4 mm so if moving 25.4 to 1 = 2540%

Mach doesn't scale by percent enter 2540 into the scale Dro and see what happens.!!!

lucan07
30-05-2016, 02:50 PM
Mach doesn't scale by percent enter 2540 into the scale Dro and see what happens.!!!


I know how it works it was the factor of the tuning error required to produce results claimed.

JAZZCNC
30-05-2016, 02:56 PM
Sparks. . . .Is the PC for machine connected to the internet and do you have fast internet connection.? If so then I can Remote into your system and check whats going on.
Will still need the info I asked for thou. PM me if want to do this.

JAZZCNC
30-05-2016, 02:58 PM
I know how it works it was the factor of the tuning error required to produce results claimed.

It's impossible to produce what was claimed. Can only be one or the other can't work in both units without axis scaling.!

lucan07
30-05-2016, 03:07 PM
It's impossible to produce what was claimed. Can only be one or the other can't work in both units without axis scaling.!


With the tuning steps per unit entered at 2540% of required value would be the only way it could occur, which is what I was saying because it cannot work in both units.

JAZZCNC
30-05-2016, 03:26 PM
With the tuning steps per unit entered at 2540% of required value would be the only way it could occur, which is what I was saying because it cannot work in both units.

Yes understand but not very likely that two axis would be correct for MM and one exactly 2540% different.!! Plus if sparks used Auto tuning correctly then Mach would take care of Steps per. Axis scaling is the only way it can happen if machine is setup correctly.

lucan07
30-05-2016, 03:40 PM
Having helped him on other threads, I know its unlikely that it is in inches, I was pointing out how far it would have to out in order to be doing what is claimed. I was tempted and perhaps should have given the only sensible response RTFM.

magicniner
30-05-2016, 05:13 PM
1 inch approx 25.4 mm so if moving 25.4 to 1 = 2540%

Sorry, yes. I didn't read carefully enough.
What confused me was that I'd already given a direct solution by telling the OP to multiply his Z axis steps per by 25.4 rather than spewing out proportions or percentages for the OP to work out, which might confuse him as easily as me ;-)

lucan07
30-05-2016, 05:55 PM
I doubt its moving that far I think it may be out but he obviously has not RTFM and in other threads has claimed similar which is where we got to auto tuning as simpler than trying to get information required to do calculations for him. Over several different threads its obvious he doesn't quite understand how Mach works when told its definitely tuned right I tried to point out margin of error that would be required so used 2540% to emphasize enormity rather than 25.4x to attain the results claimed.
I would imagine set z Axis zero to top of table rather than top of material as his sole piece of G-code never goes negative or below 1.5mm, just difficult to get information to help would guess all being done from his phone with the minimal responses.

magicniner
30-05-2016, 06:25 PM
Once motor tuning is complete, if you don't confirm moves with a dial gauge you're on a hiding to nothing.
I'm always surprised by how many take the "Red to red and black to black, throw the switch and stand well back" approach to engineering fault finding, I bet half of them can't count as far as 10 with their shoes and socks on ;-)

Neale
30-05-2016, 09:50 PM
Once motor tuning is complete, if you don't confirm moves with a dial gauge you're on a hiding to nothing.


I'm a bit surprised that you said that. I would have thought that, given that you should know to the limit of manufacturing accuracy everything you need to calculate the "steps per unit" figure, calculating is the best way to get the right number to put into the configuration page. Yes, by all means check it afterwards, but even on my crumby old MDF machine, a quick check with a rule was enough to make sure that I had, for example, set the right micro-step value on the drivers. Any error will leap out at you - you are likely to be a factor of two or more out, not some small number of per cent (and assuming that you double-check your sums and data entry, of course). Any error so small that you could only measure it with a dial gauge is looking for design or build errors - too much backlash, not properly tightening the locking nuts on a bearing holder, or something like that. And in this case, you are checking that the dial gauge returns to the same value when you have moved off zero and then back again, not measuring travel. You will probably even check for missed steps by a "return to zero" test, rather than measuring travel. The idea mentioned elsewhere in this thread that you get the "steps per" value by measuring it suggests that the builder/operator doesn't actually understand how the machine works, and that is a worrying idea.

But I might have misunderstood, and if so I'm happy to be corrected. Always willing to learn!

On a different point, unfortunately, this thread has been riddled with typos (from the OP, it has to be said) that make many of the questions really difficult to understand. Please, please, read through any post before hitting "post" and make sure that it reads correctly! Predictive spell checkers really don't understand CNC terms and concepts and can't be trusted to get it right...

magicniner
30-05-2016, 10:28 PM
My point is that so many half-informed attempts are made without a clue that double checking is always a good idea, even if you think you got your maths right, before assuming your software or hardware is broke because things don't do what you want.
I would never run a mill without checking some moves with a pointer on a rule then with a dial gauge and blocks, some might trust that the pitch of their screw is perfectly as advertised, I like to check actual movement.

magicniner
30-05-2016, 10:35 PM
this thread has been riddled with typos (from the OP, it has to be said) that make many of the questions really difficult to understand. Please, please, read through any post before hitting "post" and make sure that it reads correctly! Predictive spell checkers really don't understand CNC terms and concepts and can't be trusted to get it right...

At times I've wondered............

JAZZCNC
30-05-2016, 10:50 PM
I would never run a mill without checking some moves with a pointer on a rule then with a dial gauge and blocks, some might trust that the pitch of their screw is perfectly as advertised, I like to check actual movement.

Couldn't agree more on checking. Can't tell you how many times I've seen cheap drives with wrong Dip switch markings.
However to get the OP issue they would have to be massively wrong. Plus Mach's Auto Steps per option does a very good job of getting right provided accurately measured.

What his problem is I've no idea because is replies are much use fart in space suite.!

whensparksfly
30-05-2016, 10:53 PM
I have ajuster the motor tuning its not perfect but very close.now when I move the dro it's movement matchs the z axis.I load g code move the torch to the top of the materal zero all the axis press reg move up the z from the materal the run the g code .the torch will move down to materal then carry on down toward s the table then eventually will start to cut

Sent from my SM-G361F using Tapatalk

JAZZCNC
30-05-2016, 11:09 PM
I have ajuster the motor tuning its not perfect but very close.now when I move the dro it's movement matchs the z axis.

Told you this before The steps per is not something that you tune or tweak. It's Calculated and Fixed number based off mechanical and Drive parameters.
Motor tuning and Steps per are completely different things. Motor tuning won't make the machine move wrong distance.! (well it will if over tuned but that would be missed steps and very small amounts)

Your replies are not helping us help you.!! . . Be more clear please. Comments like " zero all the axis press reg move up the z from the materal" don't help tell whats happening or not.

magicniner
30-05-2016, 11:20 PM
press reg move up the z from the materal

Turn off predictive text and read before sending!

whensparksfly
30-05-2016, 11:25 PM
Just forget it just because my spelling and I'm not great at explaining things thanks for making me feel stupid I not be on this forum again

Sent from my SM-G361F using Tapatalk

magicniner
31-05-2016, 01:47 PM
When you decide to stop learning you may as well bin your router, move to the woods, build a rudimentary toilet and become a full time spoon whittler.

- Nick

routercnc
31-05-2016, 01:48 PM
If you are still there (!) I’ve noticed that -
.
1) You are running Mach 3 from a PC
2) You always use your phone to post on the forum
Does this mean you do not have easy access to the internet on your PC therefore cannot read the Mach3 manual? It is quite a large pdf document and one which would not be fun to read on your phone.
.
If this is the case, I’m sure someone could paste out the motor tuning part of the manual as text into a reply post. This would show you exactly what you need to do. I’m away from my home PC so cannot do it right now.

JAZZCNC
31-05-2016, 11:24 PM
Just forget it just because my spelling and I'm not great at explaining things thanks for making me feel stupid I not be on this forum again

Sent from my SM-G361F using Tapatalk

Oh pick the Teddy up and stop being Pratt.! . . . Everyone here was trying to help you but you are not helping us to help you which is all anyone has been saying.

If you took the time to give Info we asked for and gave better replies this would have been fixed ages ago.

Now if you want to try again then go back read some of the posts and give us the info we need.