Messages

561

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 08, 2022, 09:40:47 PM »

Hi,

re-locate the home switches and this local fellow will help us get on the standard track. Thank you so much for all the valuable info. It does make more sense now.

Kool. You do not not have to change the Home switches though, they can stay where they are.

Lets say you leave the switches alone and you retain the fact that the machine drives to the lower right hand corner of its travel. When it gets there you'd normally
say 'You are Home', ie the machine coordinates would be set to 0,0,0. But if you want to call the lower left hand corner Home then you'd say 'You are at 0,450,0' and
the machine would set the machine coordinates to 0,450,0. (I am assuming 450mm is the Y axis extent) So if you now said <GoTo Work Zero> it would drive 450mm
to the left and stop at your nominal Home location, ie the lower left hand corner despite it being homed to the lower righthand corner.

The Home switches define one particular location in your machines envelope, and so long as that location is precisely and repeatably defined it does not matter where it is.
You can declare home to be so many mm in that direction and so many mm in that direction from that unique location, and call that Home. The settings to make it so are
the Homing Offsets on the Homing tab of the Control Plugin.

Craig

562

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 08, 2022, 08:01:30 PM »

Hi,

the attached pic is of the right hand rule and is in the Mach4 Mill Programming Guide in Help Docs.

Craig

563

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 08, 2022, 07:52:43 PM »

Hi,

he Digital Read Out in Mach 4 counts into the minus numbers for x an y. You're saying it should be positive? Is our "Home" in the typical spot?

I would say no, the typical spot for home is the bottom left corner ie first quadrant, however anywhere is acceptable. In your case you have drawn your table in the second quadrant.
You may remember drawing graphs at school, the first quadrant is normal, but the second is OK to.

Still that means that the Y axis should count up, ie positive, from your home, the X axis would count negative from your home.

Craig

564

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 08, 2022, 06:49:47 PM »

Hi,

So this set up is wrong? See attached.

By my reckoning what you have nominated as  -Y should be +Y.

Craig

565

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 08, 2022, 05:56:59 PM »

Hi,
the most likely explanation is that your original machine/controller had one axis reversed by comparison to the industrial standard, and you became used to it.
I did the same many years ago with Mach3 and a parallel port. It was not until I started using CAM (MasterCAM in particular) did I discover that I had one axis back to front
relative to the standard. I changed my machine. It did mean that some of the Gcode that had previously worked no longer did, at least without manipulation, but it meant that
any standard Gcode would work on my machine.

Use the little diagram I posted earlier to determine the 'right hand' directions and stick with it.

That may mean that you have to prepend a single line of code to your existing Gcode files, namely a g51 line to have them work with your new control combination.
Another alternative is to use a Lua program the re-write your Gcode files by reversing the sign of all the Y coordinates. That starts to look complex if you have g3 and/or
g4 moves in your files. Do you?


If you need to change the direction of an axis use the ESS Pin Config tab to do so.

Craig

566

General Mach Discussion / Re: CNC work with robot arm?

« on: August 08, 2022, 05:33:07 PM »

Hi,
in Mach the Gcode describes the position and movement of the controlled point. Thus a move like:

g0 x100 y100 z100

will cause the machine to move in a linearly interpolated fashion to 100,100,100. Mach will produce numeric data to the motion controller in millisecond time
slices, and the motion controller will generate three pulse streams, one to each of the X, Y and Z motors.


But what happens if you don't have separate X,Y and Z motors? Say a robot arm will usually have a rotating joint at the bottom followed by two 'elbow' joints,
followed by another rotating joint. So how would Mach deal with that?. It just produces numeric data describing where the controlled point should be, it does not
know how to calculate how to get there with so many complex joints and the ranges of movement. That's where you need a motion control that can calculate on the fly
how to move the individual joints but in coordinated fashion that the controlled point moves  the required path, that's called kinematics.

In Machs case you might say that it has kinematics of a particular type, 'Cartesian Kinematics', but it does not have kinematics to do robot arms.

Mach3 will never do that, all development ceased seven years ago, it is what it is now forever. Could Mach4 end up with kinematics (other than Cartesian)?
I suppose its possible but I rather doubt it.

Craig

567

Mach4 General Discussion / Re: Probe calibration error

« on: August 08, 2022, 02:08:17 AM »

Hi,

They should figure out a better way forwards because disabling the big money features like probing because you can't find common ground is a bad customer experience. PERIOD.

But all those features work perfectly, all that is required is to check/uncheck a single box to have the ESS behavior conform to your wishes.

I for instance use Autoleveller and for that purpose I WANT the ESS to record a 'Probe Failure' and stop the routine so I can inspect and detect
the fault. I've been using Autoleveller for eight years, the last three commercially.

Note also that Autoleveller predates McSuraface  and Mach's Probing routines by at least several years. So if you want to talk about
'standard ESS behavior' then the ESS default behavior is it. Its only since McSurface and Machs Probing came out that people have started having issues,
and if they actually read the manual they'd find it had been explained and documented for nearly four years now.

Craig

568

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 07, 2022, 11:50:57 PM »

Hi,
yes that solves the problem but does not describe why it happened in the first place.

The real test is when you come to generate Gcode from a CAM program like Fusion. It assumes that you have a right hand machine, and if you don't it will be glaringly
obvious in short order.

Craig

569

Mach4 General Discussion / Re: Mach 4 and/or ESS has my Y AXIS running backward.

« on: August 07, 2022, 09:45:40 PM »

Hi,

Comparing DRO readouts to observed machine motion was the critical part.

100% there. Despite doing this now for years I can still bamboozle myself and set something up the wrong way.

I have often, with beginners, used a printed diagram on the bed of the machine to indicate the directions....it can help with the confusion.

Craig

570

Mach4 General Discussion / Re: Probe calibration error

« on: August 07, 2022, 09:28:23 PM »

Hi,

It is midsummer in the country where I live now, and it is very hot.
  Hi joeaverage, how are you doing?

Well I woke up this morning to snow on the hills behind my home, so pretty damn chilly!!

I would like Art Soft and Warp9 to discuss and resolve this issue.
Many people struggle with this issue.

I don't see that there's any real need, the default behavior of the ESS is to record a 'Probing Failure" if a g31 move executes and reaches its terminal location WITHOUT a probe event being detected.
If that does not suit, for instance you want to use Mach4 probing routines then you change it, it only requires a change in one check box, hardly an imposition.

Craig