Remember Arc Lost position

[Robertspark]

I think terry is implying intergrate thc within the uc300.... problem is.... analogue input not fast enough or not high enough bit resolution....

What about spi interface with the MCU for thc add-on board with higher analogue sampling (2 channel, 16 bit)....

[beefy]

Keith I had same thought about lost arc position and doing a touch off .... then I thought.... do you need to do a touch off? Given the torch should have been at the right cutting height when the flameout occurred.... then the torch is at the right height for an on the fly start....

Problem is... a 2m retract to start of line could be another issue.... unless you used that clever pcb plugin to correct for bed height (autoleveler).... and did a few touch offs

Good point about touching off at the flameout point BUT (always a but) if the plate has any warpage/lifting due to heat or it's lying on built up slag on the slats, etc, etc it's even possible that at the position which is the start of the gcode block the torch could end up dragging on the plate. Exactly the same the other direction too (touch off at start of gcode block) but worse, you would have the correct cut height at the start of the gcode block but you could end up with the torch nozzle dragging on the plate at the refire position.

I've seen a few photos of tables with LOTS of built up slag on the slats so I can see the potential for this to be quite real with a long gcode block.

Keith.

[Robertspark]

I think terry is implying intergrate thc within the uc300.... problem is.... analogue input not fast enough or not high enough bit resolution....

What about spi interface with the MCU for thc add-on board with higher analogue sampling (2 channel, 16 bit)....

Sorry parallel discussions....

There was a post from Balazs on 6thoct

About the analogue inputs on the uc300... hence my suggestion about spi, and external high speed and resolution spi (monitoring the isolated 50:1 side... )

viewtopic.php?f=2&t=27&hilit=Analogue&start=30

[Robertspark]

Good point about touching off at the flameout point BUT (always a but) if the plate has any warpage/lifting due to heat or it's lying on built up slag on the slats, etc, etc it's even possible that at the position which is the start of the gcode block the torch could end up dragging on the plate. Exactly the same the other direction too (touch off at start of gcode block) but worse, you would have the correct cut height at the start of the gcode block but you could end up with the torch nozzle dragging on the plate at the refire position.

I've seen a few photos of tables with LOTS of built up slag on the slats so I can see the potential for this to be quite real with a long gcode block.

Keith.

Yeah, I've got a load of slag on mine, (15mm flatbar on edge...cut over length so it slots in and has a natural curve...). Cheap and quick to replace and move the slats around

The problem is... starting on the fly ... you don't know the height over the run-up distance... but with the autoleveler plugin.... that would set it up for you

[shad]

I think we have to provide several methods for CutRecovery. For example "On the fly" or "SkipIHS" or "SkipPierce" or all in one.
Operator have to decide which way to use in the current situation. It is simply impossible to foresee all.

[Robertspark]

-"On the fly"
-"SkipIHS"
-"SkipPierce"
-all in one.

Good point, at least we have the starting point with the very accurate arc lost position... the rest is macro of plugin work

[cncdrive]

I think terry is implying intergrate thc within the uc300.... problem is.... analogue input not fast enough or not high enough bit resolution....

What about spi interface with the MCU for thc add-on board with higher analogue sampling (2 channel, 16 bit)....

Sorry parallel discussions....

There was a post from Balazs on 6thoct

About the analogue inputs on the uc300... hence my suggestion about spi, and external high speed and resolution spi (monitoring the isolated 50:1 side... )

viewtopic.php?f=2&t=27&hilit=Analogue&start=30

OK, but why to involve analog inputs when a dual $0.5 comparator or a window comparator can do this job?
The rest of things like the delay after arcOK signal is implemented inside the software, so no need for complex logic at all.
One comparator for the ArcOK and another for the setpoint.
I don't see a difference in comparion as if the analog inputs where used other than that the setpoint could be then controlled by software,
which could be still implemented in a digital way, just it requires a few more external stuff then.

[Robertspark]

The advantage is being able to set the setpoint and any hysterysis internal within the uccnc software.....
(could do setpoint via modbus if external hardware.... but just sounding out what is possible?)

then there is stuff like gap detect (rate of voltage change)

I'm sure Terry, Keith and Andrew could list a whole load of other functions.

[cncdrive]

Rob,

Sure, if the external hardware which sets the setpoint talks modbus then you could send that information via modbus.
You could add a DRO and make a modbus function which sends the values and make a macroloop which fills those modbus registers which are sent by the modbus function.
Or you could build a plugin and do the same from the plugin instead of macroloop.

[Robertspark]

viewtopic.php?f=4&t=168