[Mr.Chips]

First time i ever saw it

found mine and all lines say empty, and all 0.0000

What now?

could you run the TXT file a couple of lines up, and see what happens? i explained what happens when i run it.
thanks Gary

[Mr.Chips]

i ran this TXT file and it runs perfectly.???

[machinehop5]

G49 - cancels tool length compensation missing? H1 would still be Active at m30.

[GME]

First time i ever saw it

found mine and all lines say empty, and all 0.0000

What now?

could you run the TXT file a couple of lines up, and see what happens? i explained what happens when i run it.
thanks Gary

HH

Sorry, but I have no way to run your code. I took apart my CNC and am in the process of building a new one. I'm also building a new work table (think oversized Festool MFT). Both CNC and work table are 80/20 extrusion based. I've been milling 80/20 extrusions to get the ends square and getting same size parts all the same length. Sprinkled in the process is a remodel of the master bedroom, master bath, sitting room and sun room. I also have 1-1/4 acres worth of yard work (4 trailer loads of yard waste so far this spring, and still counting). Fortunately, our dump also makes mulch and takes all yard waste without charge.

I don't understand people who are bored in retirement. I sure don't have a problem keeping busy.

Gary

[Mr.Chips]

Uhhhhhhhh lost in translation, sorry don’t know what your trying to say. Is that related to the above code?

This is nin response to Machinehop

[Mr.Chips]

No problem Gary

HH
No problem. But I really would like to see what another machine does..
thanks

Sorry, but I have no way to run your code. I took apart my CNC and am in the process of building a new one. I'm also building a new work table (think oversized Festool MFT). Both CNC and work table are 80/20 extrusion based. I've been milling 80/20 extrusions to get the ends square and getting same size parts all the same length. Sprinkled in the process is a remodel of the master bedroom, master bath, sitting room and sun room. I also have 1-1/4 acres worth of yard work (4 trailer loads of yard waste so far this spring, and still counting). Fortunately, our dump also makes mulch and takes all yard waste without charge.

I don't understand people who are bored in retirement. I sure don't have a problem keeping busy.

Gary

[machinehop5]

Looking at your Programs...and do not see where you have cancelled tool length compensation. T1 or H1(offset) would still be Active from Line N140G00G43Z0.8000H1 right?

[Mr.Chips]

Looking at your Programs...and do not see where you have cancelled tool length compensation. T1 or H1(offset) would still be Active from Line N140G00G43Z0.8000H1 right?

Sorry I’m not familiar with tool length compensation, if that’s one of the Mach tables they are all 0.0000 and empty.
Not using any offsets.
Sorry I’m no help here.
Did you run a small file and see how it reacts with you. There is one in one of the above posts.

Thanks I appreciate you response

[GME]

Looking at your Programs...and do not see where you have cancelled tool length compensation. T1 or H1(offset) would still be Active from Line N140G00G43Z0.8000H1 right?

Wouldn't tool length compensation trigger a setting in the Mach3 tool table, assuming there was one? Since nothing is set up in the Tool Table, wouldn't T1 wouldn't point to a valid tool, and there would be no offset. So, aren't they just meaningless commands that, in this case, do nothing?


Gary

[Mr.Chips]

I posted on the VcarvePro forum and getting some input.
They are mentioning H1 also talking about the “registry “. I asked where it was, and told them that the “tool table” maybe same as registry, it was all 000 and empty.
Seems they are busy with it now.

Thanks for sticking with me I’ll let you know how it comes out.
Hager

[RCaffin]

material zeroed at surface.
it cuts deep and when it finishes the run i use MDI and say z0 and the vbit goes 0.200" into the surface.

My comments may have been already considered - if so my apologies.
I will assume you are using Mach3 - if not, my apologies.

That sounds as though you are getting extra pulses on the Z axis as you work. If this is not noise, and it may not be, then try inverting the Active state in Mach3 (under Ports & Pins I think).

Explanation: Most motor drivers require a gap between a change in the DIR state and the Step pulse. This works fine IF the rest or INactive state for the Step pulse matches the inactive state on the motor driver. However, if you have one of them reversed, you can lose a single pulse every time you change direction. Or you can gain one! Note that changing the Active state for the Step pulse will NOT change your calibration or anything else.

Cheers
Roger

[Mr.Chips]

material zeroed at surface.
it cuts deep and when it finishes the run i use MDI and say z0 and the vbit goes 0.200" into the surface.

My comments may have been already considered - if so my apologies.
I will assume you are using Mach3 - if not, my apologies.

That sounds as though you are getting extra pulses on the Z axis as you work. If this is not noise, and it may not be, then try inverting the Active state in Mach3 (under Ports & Pins I think).

Explanation: Most motor drivers require a gap between a change in the DIR state and the Step pulse. This works fine IF the rest or INactive state for the Step pulse matches the inactive state on the motor driver. However, if you have one of them reversed, you can lose a single pulse every time you change direction. Or you can gain one! Note that changing the Active state for the Step pulse will NOT change your calibration or anything else.

Cheers
Roger

Thanks Roger I look at this tomorrow. I’m burnt out right now.

[GME]

material zeroed at surface.
it cuts deep and when it finishes the run i use MDI and say z0 and the vbit goes 0.200" into the surface.

My comments may have been already considered - if so my apologies.
I will assume you are using Mach3 - if not, my apologies.

That sounds as though you are getting extra pulses on the Z axis as you work. If this is not noise, and it may not be, then try inverting the Active state in Mach3 (under Ports & Pins I think).

Explanation: Most motor drivers require a gap between a change in the DIR state and the Step pulse. This works fine IF the rest or INactive state for the Step pulse matches the inactive state on the motor driver. However, if you have one of them reversed, you can lose a single pulse every time you change direction. Or you can gain one! Note that changing the Active state for the Step pulse will NOT change your calibration or anything else.

Cheers
Roger

Roger,

An interesting post. Thank you. I learned something.

One issue, though. This machine has been in operation for 10 years and there's never been an issue. According to the OP, this is the first time in those 10 years that anything like this has happened. I've wondered if something got corrupted in the G Code when V Carve Pro created it. Stranger things have happened. The OP has gone all though the mechanical aspect of the machine and found nothing. I suspected coupling slippage, which will cause the symptoms the OP described. It happened to me. Turns out to not be the case.

Gary

[RCaffin]

this is the first time in those 10 years that anything like this has happened.
Ah, maybe, but what else has happened in those 10 years? What completely innocent little thing which has been completely forgotten by all but the machine?

I could tell you stories aplenty from the archives of Computer Field Service Engineers of these things happening, as some of you can imagine. Some of them are quite funny.

Please note: this is NO criticism of the owner of the problem. It is just hard experience from nearly 50 years in the game.

Cheers
Roger

[ger21]

You keep focusing on the g-code, but the g-code is not the problem.

What's actually causing the problem is anyone's guess.

My guess would still be lost steps.
You've changed your speeds, but have you tried changing the Z axis acceleration?

If it's always off by 0.2, then a mechanical issue would be most likely. Lost steps would usually be more random.
But it's not clear that this is the case?

It's possible that your Mach3 .xml file is corrupt, or the Mach3 installation itself.

I'd first try replacing the .xml file with an older copy from the \Mach3\xmlbackups\ folder.

If that doesn't help, then I'd completely uninstall, then re-install Mach3.


You might want to try swapping the Z axis drive or motor with an X or Y axis drive and motor, and see if the issue changes axis?

[GME]

[QUOTE=ger21;2280180]

My guess would still be lost steps.
You've changed your speeds, but have you tried changing the Z axis acceleration?

If it's always off by 0.2, then a mechanical issue would be most likely. Lost steps would usually be more random.
But it's not clear that this is the case?

It's possible that your Mach3 .xml file is corrupt, or the Mach3 installation itself.

I'd first try replacing the .xml file with an older copy from the \Mach3\xmlbackups\ folder.

If that doesn't help, then I'd completely uninstall, then re-install Mach3.

My guess would still be lost steps.

I'm confused. If the Z axis is cutting deeper than commanded, wouldn't that indicate an excess of pulses, or gaining steps (assuming nothing mechanical, like a slipping coupling)? I can't get my head around how lost steps would fit in this scenario.

You might want to try swapping the Z axis drive or motor with an X or Y axis drive and motor, and see if the issue changes axis?

I believe he tried that. Problem remained.

[machinehop5]

What if you slow down all Z Rapid move's...say make them a G1 at Feedrate of 25-50 IPM See if the problem disappears.

[ger21]

I'm confused. If the Z axis is cutting deeper than commanded, wouldn't that indicate an excess of pulses, or gaining steps (assuming nothing mechanical, like a slipping coupling)? I can't get my head around how lost steps would fit in this scenario.

When the Z axis stalls while retracting, or lifting, it ends up lower than it's supposed to be. So the next move down will be deeper.

[Jfrahm]

Yes, I'm leaning toward the loosing steps direction I looked at the code and didn't see any thing weird. I thinking an optional stop (m1) at each z retract between each letter and measure the z and compare to the readout and ( like somebody said earlier ) see if its random or a consistent amount and if happens in one spot or grows worse as it runs.

[RCaffin]

This is a standard test.
Set up a Dial Indicator on the table and put a square-ended bit of metal in the spindle.
Lower the spindle until you have a reading on the DI.
Now cycle the Z axis up and down 50 times and see if and how the DI creeps.

If it suddenly jumps - um, interesting.
But if it just creeps steadily, then you are picking up or losing Z axis Step pulses on every direction change.

Cheers
Roger