Please help me adjust these settings

[DRM812]

I recently purchased a Langmuir Pro with a Razorweld45 plasma to cut some sign work out of 1/8" material. I am at the point of extreme frustration. I have played with cutting parameters for the last month trying to get the cut quality that I want to achieve. I have tried suggestion from individuals on the Langmuir forum but only a couple people have responded with input.

If any one out there could give me some advise or input would be greatly appreciated.
I have already cut a full sheet of 1/8" with not much to show. Around here a sheet of 1/8" is $107.02. Not to start a sob story but at this point I am reluctant to even attempt to cut due to the fact that I feel like I am wasting money.

I have tried to cut some words with the cut quality pretty good but I can not seem to get rid of the divot in the beginning of the cut. I have played with pierce delay, lead in and lead out with no success. Either the divot worse or just a little better.

I started cutting bottle openers as tests plates. I figure even if they are not perfect they can still be used to relieve my pain

While my air compressor is on the lower end of recommended, I have a steady 75 psi at all times. The air dried by a drier on the back of the plasma. I did find it strange that my nozzle was not perfectly centered. I was beginning to think is was my set up but the rest of the cut is fine. Maybe I'm over thinking it now....

[adbuch]

I have sent you a PM.
David

[53hudson]

Hello Just a couple of things I would consider. Increasing the lead in and out to .25 to ensure that the initial pierce is not compromising the shape as it transitions to follow your cut line. Also I would do a test run on a piece to see what your actual kerf is then adjust sheetcam seting to actual measured value. the other thing I would check is for any physical mechanical play in x and Y axis.

[53hudson]

Sorry forgot tick the reverse cut direction for inside cuts

[acourtjester]

A few things that may help, one is to attach an ink pen to the torch (post here about how to do that use the search function in upper right), this will show if a movement problem if you have one. Two the torch tip offset may be a problem as if may affect the plasma path, this can be checked by cutting squares and looking at the shape of the edges in both directions of travel. Another is to add some overcut in the plan so the lead in and lead out are not at the same place, and as 55Hudson said longer lead in/outs too to clear the wanted cut shape.

[DRM812]

I just tried to attach a pen to the torch holder. For some reason you cant do a dry run with Langmuir's Firecontrol.

After playing with the kerf I was able to get the divot down slightly. .085 kerf seems large when Langmuir recommends a .06 for the Razorweld 45.

To me the problem definitely seems to be happening in the beginning. It seems like the torch is traveling to far into the piece before it starts the actual cut.

[acourtjester]

here is something to try change the pierce height to .150"
Are you using SheetCam to plans your cuts you can adjust the kerf width in the tool setup. if you have true width selected it will show the change as you create the operation portion in the plan.
You may suggest to Langmuir to add the dry run ability to their software it is a great help with odd problems.
This will also help if you want to add a diamond scribe or drag knife to expand you usefulness of the table

[robertspark]

To me the problem definitely seems to be happening in the beginning. It seems like the torch is traveling to far into the piece before it starts the actual cut.

Can you explain?

Can you post the gcode / *.tap file please to have a look at.

Divot chasing can be a time consuming process....

Normally there is a divot at the start of the cut, or at least where the pierce takes place..... hence the reason why there is a leadin.....

Jim Colt recommends and hypertherm too recommend a perpendicular leadin .... but it does require a high acceleration to change direction (I have not been successful with them and leave mine as an arc.

There can also be a divot at the start of the cut because at the end of the cut the kerf will cross into the previously cut zone.

You can "try" to turn your torch off before the end of the cut..... because the plasma cutting of steel is an exothermic reaction which will continue briefly once the plasma is stopped for a short while (must admit struggling to cite an article on this with a low cost air plasma cutter using free oxygen in the air at 20 odd %).... however..... M03 (plasma torch fire) and M05 (plasma torch extinguish / turn off) are not [generally (I've never seen / heard on one yet.... but my experience is limited to what I've used)] synchronous motion M-codes..... i.e. when a non-synchronous M-code is issued within the gcode, motion will decelerate to a stop to implement the gcode and then accelerate back up according to the next motion movement (G0, G1, G2 & G3)

Although this is nothing to do with what you are doing, it is a good explanatory short video clip of the principle of cutting a hole and avoiding a divot at the end / start of the cut where the kerf crosses (this is applicable to high end machines where you can vary the cutting current on the fly and they use synchronous m-codes to control the torch (turn it off + ramp the current)

[DRM812]

This was the last one I cut earlier today. I was playing around with the kerf hoping that was my issue.




(SheetCam - FireControl Post v1.5)
G90 G94
G17
G20 (Units: Inches)
H0
G0 X4.3478 Y1.1028

G92 Z0.
G38.2 Z-5.0 F100.0
G92 Z0.0
G0 Z0.2
G38.2 Z-5.0 F10.0
G92 Z0.0
G0 Z0.02 (IHS Backlash)
G92 Z0.0
G0 Z0.15 (Pierce Height)
M3
G4 P1
G1 Z0.06 F50.0 (Cut Height)
H1
G3 X4.306 Y1.0026 I0.0292 J-0.071 F70.0
G2 X4.3248 Y0.9457 I-0.395 J-0.1624
G2 X4.1719 Y0.4236 I-0.4687 J-0.1461
G1 X4.1776 Y0.4133
G1 X4.1781 Y0.4124
G3 X4.6885 Y0.1776 I0.4072 J0.213
G3 X4.8781 Y0.5054 I-0.1576 J0.3098
G2 X4.878 Y0.5079 I0.0449 J0.0023
G1 X4.8797 Y1.0954
G3 X4.8427 Y1.2552 I-0.4231 J-0.0139
G2 X4.8416 Y1.2579 I0.041 J0.0185
G3 X4.6196 Y1.4242 I-0.2411 J-0.0906
G3 X4.1758 Y1.1752 I-0.0231 J-0.4789
G2 X4.306 Y1.0026 I-0.2648 J-0.335
G3 X4.3561 Y0.9817 I0.0355 J0.0146
H0
M5
G0 Z1.0
G0 X5.4556 Y1.1006

G92 Z0.
G38.2 Z-5.0 F100.0
G92 Z0.0
G0 Z0.2
G38.2 Z-5.0 F10.0
G92 Z0.0
G0 Z0.02 (IHS Backlash)
G92 Z0.0
G0 Z0.15 (Pierce Height)
M3
G4 P1
G1 Z0.06 F50.0 (Cut Height)
H1
G2 X5.3677 Y1.1643 I-0.0121 J0.0758 F100.0
G1
G3 X4.9379 Y1.6338 I-0.4527 J0.0171
G3 X4.9356 Y1.6339 I-0.0023 J-0.0449
G1 X1.0327
G3 X0.9995 Y1.5585 I0.0 J-0.045
G1 X1.3413 Y1.1843
G2 X1.3413 Y1.0946 I-0.0513 J-0.0449
G1 X1.2047 Y0.9451
G1 X1.2041 Y0.9444
G2 X1.135 Y0.9444 I-0.0346 J0.0303
G1 X1.1343 Y0.9451
G1 X0.5185 Y1.6192
G3 X0.4853 Y1.6339 I-0.0332 J-0.0304
G1 X0.3875
G3 X0.3855 Y1.6338 I0.0 J-0.045
G3 X-0.0446 Y1.1642 I0.0197 J-0.4498
G1 Y0.4246
G3 X0.3852 Y-0.0449 I0.4527 J-0.0171
G3 X0.3875 Y-0.045 I0.0023 J0.0449
G1 X4.9356
G3 X4.9376 Y-0.045 I0.0 J0.045
G3 X5.3677 Y0.4247 I-0.0197 J0.4498
G1 Y1.1643
G2 X5.4075 Y1.2012 I0.0384 J-0.0014
H0
M5
G0 Z1.0
M5 M30
(PS100)

[robertspark]

You have a 1 second pierce delay

Code: Select all

G4 P1

does the fire control software use an ArcOK / Transfer signal from the plasma cutter?

it wont be the kerf, this is just the offset of the cut line / gcode relative to the actual cupout you want

[robertspark]

you are also cutting at different feedrates .... any reason why?

note: F70.0

Code: Select all

G3 X4.306 Y1.0026 I0.0292 J-0.071 F70.0

then note F100.0

Code: Select all

G2 X5.3677 Y1.1643 I-0.0121 J0.0758 F100.0

[DRM812]

I thought inside cuts should be a little slower?

I originally was cutting at 75ipm and 34amps for inside and outside offset. I started playing with different feed rates and amps to see if I could remedy the problem.

I belive on that g code I was cutting at 45 amps with 100ipm for outside that resulted in no dross and 70ipm for the inside with some noticeable dross but very easy to knock off with a file.

[DRM812]

You have a 1 second pierce delay

Code: Select all

G4 P1

does the fire control software use an ArcOK / Transfer signal from the plasma cutter?

it wont be the kerf, this is just the offset of the cut line / gcode relative to the actual cupout you want

I am learning as I am going along here so please excuse me if I do not have all the answers.

The recommended pierce delay from Langmuir's fourm was 1sec for the Razorweld 45.

[robertspark]

Do you have a list of gcodes for the firecontrol??

I am struggling to find a list to stop a programme and not rewind it

is M0 valid with the firecontrol?

(trying to write you a simple programme to test for the correct delay)

[robertspark]

does the fire control software use an ArcOK / Transfer signal from the plasma cutter?

Answered my own question..... that would be a no, so your G4 Pxxx time will need to "guess" that the arc has transferred and then allow for the material can be pierced....

https://www.langmuirsystems.com/thc/gui ... wiring-hyp

(not a very good system sorry that is just a bit £$%$!! for a torch height controller + cnc plasma cutter).

suggest you play around with the pierce delay time and see if this can improve your cut quality

I see you have a Razorweld45.... sorry I know absolutly nothing about them......

Do they have some cut charts that provide the pierce delay expected?
Edit: found this whilst looking for a list of G and M Codes: https://forum.langmuirsystems.com/t/raz ... chart/1232

This is the one that I use for my hypertherm 45 with a duramax torch (same as the 45XP, 65, 85 etc cut charts, kerf etc)

https://www.hypertherm.com/Download?fil ... &zip=False

for 1/8" mild steel at 45A (10g) it is 0.4 sec pierce delay AFTER the arcOK / transfer signal is recieved..... (I've not timed it but I think from reading elsewhere that the transfer signal could be 100mSec to 200mSec so the delay would thern be about 0.6 Sec if I was to guess it without using an ArcOK /transfer signal..... hence 1 second is probably a bit long....... but only you can work that out....

try a series of pierces with different delays to find something ballpark that works for you.....

eg:

Jog the machine to a location..... and run the following gcode (copied from yours below largely)

Code: Select all

G90 G94
G17
G20 (Units: Inches)
H0 (THC Off)
G92 X0 Y0 Z0 (set work co-ordinates to zero)

G38.2 Z-5.0 F100.0 (Initial touchoff)
G92 Z0.0
G0 Z0.2
G38.2 Z-5.0 F10.0
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.1 (100mSec delay)
M5

G0 Z0.2 (retract torch a bit)

G0 X0.5 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.2 (200mSec delay)
M5

G0 Z0.2 (retract torch a bit)

G0 X1 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.3 (300mSec delay)
M5

G0 Z0.2 (retract torch a bit)

G0 X1.5 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.4 (400mSec delay)
M5

G0 Z0.2 (retract torch a bit)

G0 X2 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.5 (500mSec delay)
M5

G0 Z0.2 (retract torch a bit)

G0 Y0.5 X0 (start next row)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.6 (600mSec delay)
M5

G0 X0.50 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.7 (700mSec delay)
M5

G0 X1.0 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.8 (800mSec delay)
M5

G0 X1.5 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P0.9 (900mSec delay)
M5

G0 X2 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P1.0 (1Sec delay)
M5

M0 (PAUSE)

(PRESS START TO CONTINUE IF YOU)
(HAVE NOT FOUND THE RIGHT PIERCE TIME)

G0 Y1 X0 (start next row)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P1.1 (1.1Sec delay)
M5

G0 X0.50 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P1.2 (1.2Sec delay)
M5

G0 X1.0 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P1.3 (1.3Sec delay)
M5

G0 X1.5 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P1.4 (1.4Sec delay)
M5

G0 X2 (move 1/2 inch)

G38.2 Z-5.0 F10.0 (perform quick touchoff)
G92 Z-0.02 (IHS Backlash switch offset)
G0 Z0.15 (Pierce Height)

M3
G4 P1.5 (1.5Sec delay)
M5

M30 (stop & rewind)

This should work in a small 1.5" x 2" square
It should not take too long to do.

Once you find the RIGHT pierce delay....

You will need to find the initial delay which is added for the very first pierce.... or it may not go all the way though looking at the sheetcam post processor this is set as 0.0 as default..... you may need to add a 0.1 or 0.2 delay just for the very FIRST pierce.

You can also use the above gcode test file to find what setting you would like to use to dimple for a centre punch mark for drilling...... if you pierce all the way through the metal can become hardened and be tough on HSS drill bits making them blunt a bit quicker than necessary ..... if you dimple the drill marks this is less likely to be an issue.

[robertspark]

I thought inside cuts should be a little slower?

I originally was cutting at 75ipm and 34amps for inside and outside offset. I started playing with different feed rates and amps to see if I could remedy the problem.

I belive on that g code I was cutting at 45 amps with 100ipm for outside that resulted in no dross and 70ipm for the inside with some noticeable dross but very easy to knock off with a file.

Don't normally mess around with the amperage unless you are cutting on thin material and using finecut consumables (hypertherm) [whatever the book says]

if I dimple (see my earlier post) then I will lower the amperage to minimum for this only.

There is normally a recommended feedrate from the manufacturer (like the cut chart I posted earlier) which says for 10G (1/8") MS with standard consumables with a hypertherm 45 with a duramax torch the LINEAR feedrate is 90 ipm at 130V for best quality (this could be increased to 134 ipm at 128V for production)..... so there is a range.....

For arcs and small circles the recommendation is to slow down to approx 60% of the LINEAR feedrate (or there abouts) and turn off THC.... this will allow for the cut to have less bevel / taper given the swirling plasma will cut a taper so by slowing down it allows the cut to be more square..... HOWEVER..... the THC must be off, given the kerf will get a little wider.... and the voltage will rise and the torch would dive to try to lower the voltage which would lower the cut height... and you will get more bottom surface dross forming.... given the torch is moving slower than optimum

useful guide on cut direction etc
https://www.hypertherm.com/Download?fil ... &zip=False

for 60% feedrate for circles (or arcs less than 1" dia see page 6/16 here:
https://forum.linuxcnc.org/media/kunena ... mholes.pdf

you can use "cutting rules" in sheetcam (in the tools section) to allow for these adjustments..... make sure you turn off THC with H0 and turn it back on with H1 with the firecontrol software.

[greg9504]

I just tried to attach a pen to the torch holder. For some reason you cant do a dry run with Langmuir's Firecontrol.

To do a dry run in Firecontrol:
- Turn off our plasma
- Turn off THC in Firecontrol
- Have sheet on table.
- run

[DRM812]

I just tried to attach a pen to the torch holder. For some reason you cant do a dry run with Langmuir's Firecontrol.

To do a dry run in Firecontrol:
- Turn off our plasma
- Turn off THC in Firecontrol
- Have sheet on table.
- run

I will try that now.

I am still having a difficult time. I played with pierce delays and was able to make the divot disappear slightly.

I just picked up a small sheet of 16g to try. Still having the same issues.

Here is a picture of .50 pierce delay, .2 lead in and .2 lead out.

I guess this is what it is? I understand this is not a water jet but I definitely feel like I should be able to make that divot disappear more.

[robertspark]

With the fircontrol can you see the settings such as acceleration and maximum velocity?

If so, what is it?

I will edit the sheetcam post processor as I have an idea

[adbuch]

Don,
I think your results are pretty typical - you could probably get rid of most of that pierce divot with a belt sander with #100 grit paper. I doubt that editing the post processor is going to be of much help - if any. Perhaps I will be proved wrong. As you say, this is not cnc laser cutting.
David

Here is a picture of .50 pierce delay, .2 lead in and .2 lead out.

I guess this is what it is? I understand this is not a water jet but I definitely feel like I should be able to make that divot disappear more.

20200603_200732.jpg (66.03 KiB) Viewed 3480 times

20200603_200732.jpg (66.03 KiB) Viewed 3480 times

[DRM812]

As discussed with some individuals on here, the goal is to cut signs with the intent to sell. I cut 2 small signs out of 16g yesterday and they cleaned up very nicely. I just order a sheet of 14g so I hope the outcome is the same.

I'm still learning here guys!

But I want to say thank you to everyone that shared some knowledge with me and was more than willing to go out of their way to help me out.

[robertspark]

.
I'm still learning here guys!

Me too, ...as always never stop learning something new