Torch Height Problem

[SHAPEDMETALS]

Hello. I've been searching everywhere for the answer to my problem and I'm sure it's something simple and I'm just not phrasing it correctly to find the desired answer.

I have a Premier 5x10 plasma table with a hypertherm 85 sync running sheetcam and mach 3.

I'm honestly not even sure if this is a sheetcam issue or something I need to start researching on the mach3 side of things.

I'm having an issue with my tip being too low but only when it doesn't do the touch off the material.

I'm not sure what the parameters or qualifications for this scenario are, but from what I can deduct it's when the next hole is within a certain proximity to the hole it just cut. It won't touch off the material but just move and come down without hitting the material. The issue is, that when it does that, it's way too close to the material and I'm roasting my tips. However, when it does the touch off the pierce height and everything is perfect.

I did come across a couple posts mentioning refdistance but I'm unsure if that's what I'm on the hunt for and before I just went changing things I figured I'd come ask before I make anything worse

If anyone can shed some insight into this I'd greatly appreciate it.

[bLouChip]

My research on Premier Plasma systems is they use Proma THC SD THCs. Is that what you are using ?
If so, then you are correct in understanding the problem, since this THC is operates autonomous to Mach3, and can/will leave the torch height out of position relative to where Mach3 thinks it is, where Mach3 last left it.

I don't know Mach3, but its likely the proximity logic for next cut to avoid a torch probe cycle is in the scpost. Please upload it the scpost you are using and I can take a look at it.

[bLouChip]

I found "PremierPlasma CNC Floating Head.scpost" in the sheetcam post lib. If this is what you are using, then...
It certainly has code to skip torch probe cycles, but its not a proximity of next cut, rather its an accumulated distance of all motion by XY on straight line measurements between end points of motion. Anyway, its easy to adjust such that a torch probe cycle is performed on every cut, as it should be with the Proma THC SD unit in play.

change this statement in OnInit() function as such,
from:

Code: Select all

refdistance = 130 * scale

to:

Code: Select all

refdistance = -1 * scale

[djreiswig]

Proximity probing is an interesting concept. I guess it wouldn't be hard to implement. Simply note the x & y of a probe position and then calculate if the new pierce is outside of a circle centered on the previous probe position.
Probably makes more sense that cut distance.

[SHAPEDMETALS]

I appreciate the replies. I'll implement that change and report back. Thanks!

[bLouChip]

@djreiswig- agreed.

[djreiswig]

Doesn't look the the formula is terribly complicated. Care to give it a try?

[bLouChip]

Proximity Distance to skip a Torch Probe Cycle before pierce...

Code: Select all

in OnInit()
   ...
   LdcProximityLastX = 0
   LdcProximityLastY = 0
   LdcProximityDist = 100  -- mm,  0 = always run Torch Probe Cycle

in OnPenDown()
   ...
   if math.hypot( LdcProximityLastX - currentX, LdcProximityLastY - currentY ) <  LdcProximityDist then
      post.Text("; Torch Probe Cycle skipped due to pierce being inside the proximity radius of a previous pierce \n")
      return
   end
   LdcProximityLastX = currentX
   LdcProximityLastY = currentY

   ... do your normal OnPenDown() code...

[djreiswig]

Cool.

[bLouChip]

Correction, rather than 'return' from OnPenDown() if inside proximity distance of a previous pierce, you'd still want to perform other key OnPenDown() functions such as firing the torch and pierce delay. Also, the initial value of the ProximityLastX, and Y, should be a large value such as 1e+17, so that the first pierce is sure to have the Torch Probe Cycle run.

[djreiswig]

Yeah, I was just looking at your formula. Looks like it should work. I think that is a better idea than the cut distance at determining when to probe.

[finyuk2005]

I found "PremierPlasma CNC Floating Head.scpost" in the sheetcam post lib. If this is what you are using, then...
It certainly has code to skip torch probe cycles, but its not a proximity of next cut, rather its an accumulated distance of all motion by XY on straight line measurements between end points of motion. Anyway, its easy to adjust such that a torch probe cycle is performed on every cut, as it should be with the Proma THC SD unit in play.

change this statement in OnInit() function as such,
from:

Code: Select all

refdistance = 130 * scale

to:

Code: Select all

refdistance = -1 * scale

Just had a quick question regarding the refdistance thing here, i have tried using it in my post, with distance set at 250mm, but when i check the G code i only see one instance of G31 probing, which i thought I should see much more often, (this is on a full sheet of 10mm mild steel 2.5mx1.25m, 470 plus pierces)
So i wondered if the post processor is calculating the accumulated distance or is it calculated as the program is running?

thanks very much!

[djreiswig]

The probe moves should be inserted as the gcode is created by the post processor.

[finyuk2005]

Ok cheers
I will need to check over my post code,
I had simply copy and pasted what i thought was the relevant parts from another processor into the one i was using with my THC, but i must have missed some part of it!

I now using a Proma 150 THC and have an ohmic sensing probe input, so i might just look for a more suitable post

cheers!

[finyuk2005]

Tried a different post, and now i am seeing plenty of probings in my G code,(that sounds a bit rude! ) so i am happier with that,
thanks for the info!!