Pokeys PlasmaSens and PlasmaSensOut THC issues

[Ossy169169]

Hello all,

I really need some help!

I have just built my first table 4.5" by 5.5" with Pokeys57CNC as the controller, PoStep60's as the X, Y, Z and A axis drivers all with Nema23's. I am using PokeysSens and PlasmaSensOut for the THC, this is where my problem lies!

I am getting a large slope on my cut (cut slopes from top of the plate inwards as the cut moves down) which is indicative of the torch being too high above the work from what i understand.

It appears that the THC is working as I am getting the signals for Ark ok, THC up and THC down on both the PlasmaSensOut and Mach4 GUI.

I am setting the target voltage for around 115v as per the instructions for the Hypertherm 45xp cutter. The machine will probe and then retract to the pierce height (taking into account the z-switch offset correctly - I have measured this so can confirm this is correct).

The cut will then start and is always around the correct 110-120v region for around 1-2 seconds before it drops down to about 85-90v, never getting above 95 for the rest of the cut, this makes little sense to me as i have set the THC for a target of 115v!

Could someone who is using Mach4 and the PlasmaSensOut unit share their settings as i am sure that this is software related.

Thank you for any help or advise that you can offer, the cut quality is really good with the exception of the bevel.

Thanks in advance.

Mike

[weldguy]

Hey Mike

I am not at all familiar with your equipment, hopefully someone here who is familiar can help you further with the numbers being off. I am curious though...if you were to pause the machine mid cut after the THC has turned on and torch has stabilized and then measure the distance between the torch and workpiece what is the measurement. .060" is what it should be in most cases and because of your excessive bevel it must be much higher despite your voltage being too low. If you pause the machine and your higher than .060" try reducing your arc voltage to bring the torch down. What voltage do you get to in order to achieve the .060" working height?

[acourtjester]

I have not use that electronics but have other types. it could be you have the polarity of the arc voltage reversed. If the THC is working correctly you should see it slightly move up and down as it cuts, not going in one direction only. There can be small value changes between one set and other as to actual arc voltage, but 25 volts is way off. Another thing is the voltage divider setting I think most Hypertherm come set at 50:1 you electronics/software could be looking for a different value. Here is a video of what weldguy is talking about.

[Ossy169169]

Hello weldguy and acourtjester,

Thank you both for your replies.

To start with Weldguy's questions - I have measured the distance once the probing routine has been completed and the torch height is correct at 1.5mm (0.059") from the material. I have not checked the distance from the tip to the plate once the cut has been running (i think that this is what you are asking). I am not at home at the moment but should be back on Friday so will test this and let you know the outcome. I have tried reducing the value on the THC (both on the Mach4 screen and on the THC itself to 85v and this does not make any difference at all. I have attached a video link to this post showing the THC unit during a small part cut, you can see that the THC tries to increase the cut height due to the voltage being below the 115v setpoint, but it is incredibly slow and the cut is complete before it has made it anywhere near.

acourtjester - when you say the polarity of the arc voltage reversed, i assume that you mean the cables with the 50:1 voltage from the Plasma cutter to the THC unit the wrong way around? i did check this and i believe that they are correct based on the Hypertherm manual, but could confirm this with a voltmeter. The THC does try to move up and down (mainly trying up as the voltage drops so low) but it is very slow (see the attached video link to see how slowly the voltage increases), it takes so long to increase the voltage that the cut is finished before it is anywhere near the setpoint. It also jumps like crazy from like 85v to 1 I need to speed up the adjustment rate i think but i am not sure how to do this. The cut is perfect from where it starts but gets very beveled after around 10mm or so.



You also mention that "Another thing is the voltage divider setting I think most Hypertherm come set at 50:1 you electronics/software could be looking for a different value." how do i check this? is it possible to check this in Mach4 for example?

What takes priority, the voltage set on the THC unit or the voltage set in the Mach4 screen set? i think it is the THC but i want to confirm.

Thanks for your help guys , it is really appreciated.

Mike

[acourtjester]

The video shows the THC is slow to correct as you stated, and the up and down arrows on the unit stay on to long. after the start it has about 85 and the up arrow stays on then it makes a sudden voltage change and the arrow is down.????
So you know how it should work is after the pierse there is a delay for the arc voltage to stabilize about 1 to 1.5 seconds. This is not the pierce delay on the chart, that is for the plasma to blow through the metal before moving. The delay I am talking about is after the torch starts to move. This is when the THC starts to take control of the Z axis motions. Now let's say you set the arc voltage to 100 after the start of movement any changes to that voltage you should see the torch move up or down to keep the 100 volts constant very quickly basically the up/down arrows should just blink. You should have a tolerance for the amount of change is allowed before the THC signals up or down to the controller, 1 to 2 volts.
This is an old video I made, in the middle it shows the lead screw for the Z moving. As the torch follows the slope of the metal it moves up and down as the voltage changes. But along with that you can see small reverse movements as the arc voltage changes and the torch makes slight up/down movements to maintain the set arc voltage.

[acourtjester]

I seem to remember that another guy had a similar setup and had problems. With one of the tables I built I used a different electronic controller which was not to be used for Plasma. It did not have enough signal inputs. if you are using home/limit switches you need 5 inputs. One for home/limits, one for a floating head switch or ohmic sensor, and 3 for THC (arc Ok, up and down).
I don't think you arc voltage is wired wrong, or that the divider ratio is wrong, that would make the arc voltage way off

[Ossy169169]

I am fairly sure that this is a software/motor tuning related issue and that it is not arc voltage wiring, i am not sure about the divider ration though.

Do you use Mach4? I would be interested to see what sort of numbers you are using for the settings in the THC config page such as adjust rate, XY blend velocity and anti dive rate. The stepper motor that i have driving the floating head is working fine but is very slow when jogging / probing, i did this intentionally to reduce the risk of bending light material when touching off and also to reduce the likelihood of damage due to the head diving too quickly into the material. What sort of IPM or MMPPM of movement do you use for the Z axis?

As you note, the THC did try to increase the voltage but it was very slow, leading me to believe that i have the settings set too low/slow.

Thanks again for your help, we apear to be narrowing down the potential issue

[weldguy]

It does sound like your on track to narrowing it down as it appears to be doing what it should but too slowly for it to be effective. THC's can be finicky and hopefully someone else can chime in here and get your z correction speed up, there has to be a setting somewhere to increase the Z motor speed. Best of luck.

[acourtjester]

Sorry I did not use Mach 4, I moved to UCCNC and then to Linuxcnc. Is there tech support for the PlasmaSensOut, they should have the answers, if this is a new unit it may be defective.

[Ossy169169]

Thanks guys, i will be back home tomorrow so can have a bit more of a play with the table, its just frustrating that i have everything else working but this

[Specnaz]

Thanks guys, i will be back home tomorrow so can have a bit more of a play with the table, its just frustrating that i have everything else working but this

Hey have you ever come to the bottom of the issue?