Cut height problems with shielded Tecmo PTM-80 torch

[jazzkramer]

Hi, i read many posts here and I leart many things, THANKS YOU ALL!!

It's the first time I write, i built a (diy) plasma CNC driven by LinuxCNC, mesa 7i96, THCAD10 (for THC), cutter Lotos 5500D-CNC. (Ok, strange thing I'm temporarily using the 1.5m X axes turned 90 degrees so the Z axes move horizontal like a 14cm Y, and I mounted a small Z (6cm) on top of the Z to work as the real Z.... It works!)

Recently I switched to a vertical torch (PTM60) and I bought shielded consumables because I wanted to add a Ohmic sensor to get precise sensing of the materials height position.

First problem/question: the Lotos cutter has a recomended height cut of 1.5mm and at that height it performs good vertical cuts, but the shielded torch has the consumable 1.02mm inside the shield... so, I have to set the cut height from the material at 0.48mm.

Is it enough clearence?

What actually happens is that the torch moves up and down (THC) during the cut and it often touch the material so I have to stop the cut not to damage the torch.

What cutting distance do you have on your shielded torch?
Does it means I can not use the shielded consumables with this Lotos because the cutting 1.5mm is too near for this?

I don't have the Ohmic sensor yet, so,
a possible interpretation of the problem could be:
1) The torch goes down before the cut, to sense the material position (by the floating switch).
2) The material flex a little (or maybe being not perfectly plane on the supporting grid because has some dross and cuts from previous cuts) and the torch thinks it's lower than it actually is.
3) The torch goes to the "correct" piercing height BUT actually it's lower.
4) LinuxCNC starts the torch ON and lower the torch to the cutting height (too low) maybe already touching the material.
5) After 0.X (did not remember.. 0.2 ... 0.3 seconds) the auto voltage of the THC decide that's the correct voltage for the cut... but it measured it with the torch too low on the material, so it wants to
keep the height on this voltage
6) eventually the THC corrects the height crashing in the material AND I have to stop the CNC to
protect the torch.

Do you agree with the above interpretation?

Do you think that adding the ohmic sensor to the shielded consumables, will solve the problem and that the clearence is 0.48mm only will be enough?

Do you maybe have other suggestion (a part from adding a torch collision detector which I'm doing soon)?

Do you think unchecking the "Auto Volts" on Linux CNC and set a correct voltage manually could
tell the THC to go to the 0.48 correct height?




SECOND "Problem":
This is not a real problem (I hope so), it's more a question:
Because of the fact that I was in a hurry to cut some 2mm thik parts, I decided to use the unshielded 60A nozzle that came with the torch at 1.5mm height (instead of crashing an possibly damaging the torch on the material).
I cutted the parts with the machine at 55A (it's maximum even if an online guy says it measured it and it's really 48A on the Lotos), 75 PSI and they came VERY VERY good. No Dross, almost perfect cut, a part for a small miniZ axis vibration.
So the question is: is it normal? A material that in theory wants to be cutted with 35A, cutted with 55A came so good?
I mean: can I cut every material at the maximum and I will get always this so good result?
So why bother with lowering the amps??

I'm a bit confused about all that.

I apologize for my bad english, and the "many questions" in this post, but I'm really new in Plasma cutting world and I did not find an answer to these questions.
Thanks for every answer you can give me,
have a nice day, happy cutting!
Alberto.

[acourtjester]

Hi Alberto
What I think will help you with cutting height is to not use the THC until you know at what height is correct for your torch.
This will mean making test cuts maybe about 140 MM long and do an edge start (place torch at the very edge of the metal) so you don't have to have the pierce height involved. Also do not change the torch height at the end of the cut, when you have good cuts measure the distance from the tip to the metal surface. This will give you the correct cut height for your torch. Now most torches will cut 2 MM steel at about 1.6 MM if you measure and find a difference this will tell you how far off your torch is working. You may be overthinking the design of the torch's inside and trying to correct for it. Actual cutting distance is more important.
With the MiniZ axis could it be your THC could be causing these movements.
Cutting with to high amps will make a wider kerf and will put to much heat in the metal and can cause warping as you cut.
Your English is not a problem The info is there for us to help

[adbuch]

Tom gives good advise. I agree to do some testing without our thc to see what specific torch to material cutting distance works best for your particular metal thickness and cutting speed.

David

[jazzkramer]

Thank you. I will do some more tests,
Alberto.

[weldguy]

Tom does have great advice and is very helpful. Once thing he may have missed is that you already know the recommended cutting height which was specified from the manufacturer as 1.5mm. With a shielded cap on you are correct you still want to maintain the same arc length of 1.5mm but to do that the shield is very close to the material as you stated. This should not be a problem with an accurate and fast reacting torch height controller but yours may not be sufficient to accomplish this I don't know.
There is no doubt that your cursed from the beginning with the material flexing down and throwing off the initial height sensing and with the shield installed this only makes this issue worse and this problem should be addressed before anything. Really all you can do is hold the material down with your hand or something during the initial height sensing to solve that. You can also increase your piercing height, many plasma cutters pierce at a higher distance then they cut at, this will give you a little more distance to make things easier and may also extend consumable life giving the molten metal some airspace to evacuate itself from until the arc pierces through the material.
You really should be able to achieve exactly what you want to simply increasing your pierce height, holding down the material for initial sensing, and then you dial in your torch height controller to achieve the 1.5mm arc length and determine if your height controller is working fast and accurate enough to maintain the correct height when material warps during cutting. Hope this helps.

[jazzkramer]

Hi,
it's time I don't write here, it's time i don't use my diy CNC (mesa board.. linuxCNC..)

My cutter is the cheap LOTOS LTP5500D-CNC and last year I bought a vertical torch: PTM-80 tecmo (the original one, not the chines copy).

There is something I do not understand with it, maybe you can help: I see the recommended cut height by LOTOS (maybe it's with the IPT60 manual torch included with the machine?) is 1.5mm from the sheet.

Thinking "maybe the cut height is different for the PTM-60", I tested various height and I should go seriously closer to have near vertical cuts, BUT, let's say 1.5mm distance from the metal to cut was correct: now I have this problem:

The tip inside the shielded cup of the PTM-60 torch is 0.88mm recessed (inside the shielded cup), this way, if the tip must be a 1.5mm from the sheet I'm cutting, the remaining clearance from the shielding cup to the sheet is 0.62mm.

So, even if (i think) my THC now is working well (2mm sheet Volts varies between 92 and 98), if the torch is cutting near a previously cut part, which has some little "dross" on it's side that is higher than 0.62mm, it's crashing in the material and the cut is stopped.

Now my questions are:

Is this because my cheap LOTOS cutter wants to be too close to the sheet to cut?
Are more expensive cutters capable of leaving more clearance even with shielded cup?

Is a variation of 6V in the THC reading too big? Should I fix this or it's a normal variation?

If you are using the PTM-60 with shielded cup, do you experiment this same problem?
And, which height are you cutting at?

Thank you, I'm just trying to start a little production to sell something I invented and I'm not able to cut reliably, so I'm almost stuck here. (but this is the money I had... I had to start with what I had, looking to upgrade as soon as possible).

Alberto.

[weldguy]

Hey Alberto

Unless you have to use the shielded cap for ohmic sensing or some other reason I would eliminate it and give yourself the extra clearance you need to clear the top dross. Do you need to use the shielded cap for any particular reason?

[acourtjester]

I would suggest you do some testing to find the best setting for your setup. This is to make test cuts in a straight line without the THC and then with the THC. You first need to find the correct pierce and then cutting height to get good cutting action. Along with that is the arc voltage when you have that info to then use it for setting your THC voltage for each cutting operations. Then you work on setting the THC delay (after pierce height) then the voltage tolerance for the THC and last the time for corrections. If the correction time is to fast your THC will bob up and down to much, to slow and the torch crashes. Here is a video I did sometime back showing some of these things. It is for a different THC but they are similar.

[jazzkramer]

Thank you all,
So I know that a small clearance like about 0.5mm can be used and works... but in my case:

I just realized, looking at the video (and the post), I really need to fix the sheet bed to the CNC...
Now the CNC mechanism seems to be really rigid per se, but actually it's not fixed/solded/bolted to the bed, which in effect support it... so taking such a precise measure is not really simple.
The bed and the "CNC thing" are not squared...
(This is because of.. really small space... and other location/organization reasons)

Playing with the PID settings on the THC I'm not really able to prevent about 5V variations even when the THC is not locked (in corners), so this mean 0.5mm distance variation... so that's a problem too... maybe because the bed and the CNC not in one piece? Or maybe there's some backlash in Z axis or... maybe it's not reactive (slow PC? Yes I was at the limit on jitter too in linuxCNC... maybe...).

So, I have to investigate all this things.

Just another question: when it's all setup correctly, a good THC really can get the voltage FIXED to the desired voltage? Or how large is the variation/oscillation during cuts?

Thanks again,
Alberto.

[jazzkramer]

and yes... I mistook the torch name as well It's a Tecmo PTM-60 not 80

[acourtjester]

Are you using the THC that uses the Mesa card THCAD-5 or another THC unit?? I have moved to the LinuxCNC with QTPlasmaC GUI, Mesa 7I76E, THCAD-5 (THC) with Ohmic sensing and it works very well. Are you using the Plasma units work cable directly connected to the metal you are cutting?

ground clamp (1).jpg

[jazzkramer]

=> Are you using the THC that uses the Mesa card THCAD-5 or another THC unit??

For THC I'm using a THCAD-5 (or THCAD-10 not sure at the moment) and some resistors (I think I put them in series) to lower the voltage because my cutter output the actual voltage. I check some LinuxCNC docs to calculate these resistors.

I'm using LinuxCNC version 2.8.0 with Axis based on a Mesa 7i96S.

I would like to upgrade to QTPlasmaC, but I'm waiting for a new hard drive to leave the current installation untouched, and I'm also a little worried about the settings: I had to put some lines in the config files to get the CNC to start get arc ok and some other things working. I don't remember...
With unshielded consumables my CNC works decent (I just ordered some unshielded tips)...

I was trying to use the shielded one to add an ohmic sensor like you by another THCAD board... but then I got stuck in the "crashing into sheet thing" while cutting with shielded, and I'm trying to solve that... but.. I need to have some parts cut and I'm temporarily going back to unshielded tips now.

=> Are you using the Plasma units work cable directly connected to the metal you are cutting?

I'm using the "ground clamp" that was included in the cutter, when possible I clump directly to the sheet I'm cutting.
Do you think a more robust clamp like the one you show here could help to solve the THC "voltage" fluctuation? I did not understand how to work well with so small clearance... Maybe it's only my imprecise CNC setup... I'm trying to understand.

Now it's coming something to my mind: The 2mm sheet I'm cutting now is a galvanized one: does this make a difference? Could this be part of the problem?

Thank you,
Alberto

[acourtjester]

I was mistaken with the THCAD-5, it was the -10 for the THC, I did need to change the divider output from the Hypertherm to the 20:1 for it to work well. The circuit I used was the one posted by the LinuxCNC group.
With the Galvanizes metal, depending on the thickness of the Zinc coating it may cause problems. You also should watching the fumes from the cutting as it may not be good for your health.
If you are connecting the work cable to the metal that came on the plasma unit, that should be fine. What I made was just so it was smaller so the it was out of the way of the torch.

[jazzkramer]

So probably mine is the THC-10 as well.
I don't remember the ratio I used in the divider.. could be I was too conservative, maybe, to be sure I do not exceed the THC-10 voltage, I divided too much (I will check that too, more resolution could help my THC precision). When I built the circuit I was worried about Voltage spikes. I think I calculate that for spikes till to 500V... is it too much? I did the calibration (do not remember the procedure) related to a number that was written to the back of the board, and put the voltage scale and voltage offset in linuxCNC.

I'm using FFP3 masks (98% filtering) when cutting galvanized sheet, I just don't know effects on THC/voltage/or "little points/bubbles/hills of material" near cuts, in which sometimes the shielded torch with a reduced clearance is crashing, stopping the work.

Ok I see about the clamp, you're right, many times the big clamp coming with the machine is way too big to be clamped directly to the sheet.. but usually find a way to. Yours is better, I will think about building one like yours.

Thanks, Alberto.

[acourtjester]

The write up in the LinuxCNC/Mesa talks about the value to put in the software for the divider ratio. Most THC units use 50:1 and many plasma unit come with that set in the divider circuit. I think the reason I change to 20:1 was it did not seem to lock on the voltage being that low with the 50:1 output. I have not been doing much Plasma work with my table as I have been switching over to Servos and Proximity sensors. And my projects have been toward CNC router things. I have completed that and will put the water table and torch holder back on the table soon.

[jazzkramer]

Thanks, I have to better look in this thing, maybe I can reduce the ratio, like you did, to improve my THC.
At the moment I have about 5V oscillation during cuts, which corresponds to about 0.5mm in height, so... with just about 0.5mm of clearance with the shielded torch... it crashes. After I finish the project I'm working on, I will take precise measures to see the actual V before and after the divider.