[fullbore]

When I push the torch button in Mach for manual torch on. The torch dives right down to the plate ????...When I do the manual test cut as stated in the manual from Tom at candcnc the torch dives right to the plate and starts to do the move...anyone have any suggestions. I have went back through all the setups like three times with the same outcome.

Thanks for any and all info

[Tensaiteki]

What plasma power supply/torch are you using?

When you do the manual torch on, does the torch fire before the dive starts?

Does it still dive even when you turn off the THC?

Are you sure that the top of your material is zeroed and that all the G-code moves for the Z-axis are to positive Z points?

[fullbore]

Hypertherm85 with machine torch.
Torch fires before the Z moves
Even with THC off
Yes all G code is positive.

I single blocked through the first part of a program...

G00 Z.5................rapid plane Z .5 on readout
rapid to start position
G28.1 Z 0.12.........Takes z down to -.135
G92 Z0.................Setting Z "0"
G00 Z .150............Moves Z to .150 on read out
G92 Z0..................Setting Z 0
G00 Z .2...............Moves Z to .200 on readout
M03.....................Moves Z up to 1.0348
G04 P0.2.....pierce dwell
G02 X____Y___Z .08..Feeds X Y Z actual position of Z is .9148

These Z moves are all over the place

[Tensaiteki]

Interesting, the code looks right to me.

What are your "Cut Profile" settings? (i.e. target voltage, delay, upper limit, lower limit, span gap, fault options)

What is the torch voltage displayed in mach when the torch is moving up?

[fullbore]

Target Voltage..........118
Delay.......................1
Up Limit....................200
Low limit...................50
Span Gp....................4
No fault option


Voltage moving up is 118..

all these z moves and resetting of z 0 is okay?

[Tensaiteki]

all these z moves and resetting of z 0 is okay?

Yes, that is, assuming that you are using a "floating head" type torch mount and that from the distance between the torch tip touching the material and the switch getting activated is 0.150" on your machine (mine is closer to 0.125"). Also, that the switch on the Z-axis is connected to mach as the Z home switch.

So far, to me, it doesn't look like a problem with sheetcam, so now we're trying to figure out if the problem is mach or the THC hardware.

When the Z is moving up (even though it shows it at the correct voltage), is the UP "light" in mach on/flashing?

I would try running the code with the PM85 turned off, watch mach and see if it still thinks it's getting an Arc OK signal (it shouldn't).

I should've asked this earlier: has this machine ever worked correctly before? Is this a new build that has never run or did this problem start after everything was previously working?

[Cburt]

The problem is most likely your post processor file in sheetcam. It has a value for backlash that might need to be edited. I had the same problem but when I hit the Torch button my torch moved up not down. Also check the THC moves box on the screen when you hit the torch button. Does it give a value when hit. If it does then that is the distance that the post processor is moving it. Hope this helps some.

[fullbore]

Thanks guys for the responses. so with the PM85 off I run the program when it gets to the M03 The "light up" is on, .82 in the THC box, and arc ok is on???....remember the plasma unit is OFF.

I get an error: THC tip volts outside range...well the unit is off so I should get that error...right?

When I turn off the torch in Mach everything works perfect.

So It sounds like its either the THC or my torch right?

This is a new build. Could it be the ground?
Its obviously not getting the tip volts of something.

[Tensaiteki]

... arc ok is on???....remember the plasma unit is OFF.

Sadly, mine does that too (PM85 with CandCNC THC). Kinda makes you wonder why we even bother hooking up the Arc OK signal to the plasma if the friggin' THC is just going to ignore it anyway. But this doesn't really help, so I digress.

How is your grounding set up currently? The table and plasma should be grounded to an actual ground rod (or metal water pipe) no more than 10 feet away. The computer/CandCNC hardware are not connected to the same ground, they use the ground wire from the mains.

What material and consumables are you testing with?

What happens when you set the THC delay to something large like 10 or 15 seconds?

Do you have a way to take and post a video?

[Torchhead]

[QUOTE=Tensaiteki;981279]Sadly, mine does that too (PM85 with CandCNC THC). Kinda makes you wonder why we even bother hooking up the Arc OK signal to the plasma if the friggin' THC is just going to ignore it anyway. But this doesn't really help, so I digress.

Open the Cut Profile. Find the ARC OK Current setting. IF it is set to 0 then it will fire the ARC OK all the time (anything is greater than 0). Set it to something like 20. Save the ARC OK on all the time will stop.

TOM Caudle
www.CandCNC.com

[Torchhead]

Target Voltage..........118
Delay.......................1
Up Limit....................200
Low limit...................50
Span Gp....................4
No fault option


Voltage moving up is 118..

all these z moves and resetting of z 0 is okay?

Go through the step by step setup and testing section in the DTHC II manual. There it asks you to do the self tests on the modules (the DTHCII and the THC Sensor PWM). If those pass then the next (and important step) is to do a manual cut (THC turned off in MACH and the Auto THC On in the cut profile disabled). The manual cut should be generated using the SheetCAM POST and the settings and the setup moves should place the tip of the torch the right height (.063) for the initial cut. If the cut does not start at the right height then something is wrong with the touch-off distance (switch offset) or the inital cut height (SheetCAM setting) or the Z is not properly calibrated (steps per unit) If the torch is cutting at the right gap, Then you watch the TORCH VOLTS readout in the screen as it cuts with the THC off in MACH. That will establish if your voltage is close to being correct. It should be pretty close to what the 85 chart calls for. If it is not and you set the target at 118 then it will climb until it makes the Torch volts 118.

If the readout is showing lower than it should, make sure the jumper setting in the THC Sensor PWM matches the voltage divider setting in the 85.

Without some feedback as to what the voltage is doing during a manual cut it's not going to be easy to diagnose the problem.

If you are using the 50:1 divider setting ANY ground loop voltage (especially AC leakage) of more than 500MV will skew the reading by more than 20%

If the leakage is bad enough it will even cause the 20:1 range to skew some. We have a 65 and 85 in house now and both hooked up and ran perfect but we have very low impedance grounds on the table. We are working on setting up a test bed with a ground loop (on purpose) to work out a way to measure and ignore it.

If you can give me sufficient information (it's fine to describe what it's doing but it's like watching a video of an airplane crash and tying to figure out what caused it) The feedback signals (Torch voltage, and the UP and DOWN indicators ) are the true clues to solve WHY it is raising up. The TIP SAVER (anti dive) will lock on if the voltage gets too high but you can turn that feature off. The THC Fault Delay in the Cut Profiles will delay any faults from happening (voltage out of range etc) so during the test you can set it to a long time period. That does not stop the UP and DOWN signals but if you have the THC logic off in MACH (button off) it will simply ignore them.

Also see the previous post on turning up the ARC OK Current trip point.

The whole thing is to realize that the Torch Volts is supposed to reflect the ARC GAP (distance from the the tip to the material and is directly proportional so the longer (higher) the gap the higher the voltage. It will read that (whatever it is) and will compare it to the preset you have set and will raise or lower the torch to make them match.

If the feedback voltage is wrong then it will still try to make the voltages equal even if that means lifting the torch up to the ceiling.

If the actual height of the torch always agrees with the Z DRO then you have the Z tuning and the THC Rate set ok. If the UP LED comes on and the torch goes up then the hardware and software is doing what it should. If the feedback is wrong then it does not know. If the torch moves UP or DOWN without the THC Button on (after the pierce and plunge to initial cut height) THEN there are problems in the motion electronics.

It is KEY to know what the indicators are telling you especially during a manual cut at the right height. There is a lot to watch so it may take a few straight long cuts with the THC disabled to gather the information. Make notes of the voltage readings as it cuts.

If you find the Torch volts is lower than it should be try changing the Preset to the lower value and see if the torch will lock in at the right height.


TOM caudle
www.CandCNC.com

[fullbore]

All self tests are good. I dont have the newer pickup module but went through the set up in the manual that came with mine and was all okay.

Did a test cut with PM85 OFF. program worked perfect. Tip at the right height and everything.
Did the same test cut with the PM85 ON it moves to position lowers torch to material sets Z "o" moves up to start of pierce height the turns on torch moves down to plate. Auto arc is disabled and THC is off in mach.

there is a -.79 in the THC moves box, Tip saver is on, arc ok is on. voltage when its right on the plate is 152 its set at 118 in the cut profile. Seems like its sensing the bottom of my water table for the voltage. I can turn on torch manually and do a move manually and the voltage is right on what the cut profile is set to. I double checked the switch offset and steps on motor and all good.

[I Lean]

voltage when its right on the plate is 152 its set at 118 in the cut profile.

This is it. Right or wrong, the measured voltage is higher than the preset voltage, so it will move the torch down until they match. (or it crashes...)

[fullbore]

Yea I got that that is the prob. its trying to close the gap to get to the set 118 volts right?...now how do i go about fixing it ? It must be something really simple because I have passed it 20 times now without actually dialing that in.

Thanks

[I Lean]

Set your preset voltage to whatever value lets the system maintain the correct torch-to-work distance. If 152 volts is what it takes to keep it cutting at .06, then put in 152. Trial-and-error your way to the correct number.

And write it down! I have a little "cheat sheet" next to my table that tells me where I need to be with my settings most of the time.

[Torchhead]

Yea I got that that is the prob. its trying to close the gap to get to the set 118 volts right?...now how do i go about fixing it ? It must be something really simple because I have passed it 20 times now without actually dialing that in.

Thanks

What is confusing about your post is that you state when it cuts with the THC button off at the correct height you see the 118V you are expecting. If that statement is accurate then you should be able to click the THC button on while it is cutting and it will start to track.

I suspect you are setting too long a pierce or you have the plunge rate really slow so it sits over a spot too long and burns a big hole and the voltage spikes. At the same time the delays in MACH (THC DELAY) and the delay in the cut profile (THC Fault delay) are short and start locking things out and sending faults if the pierce sequence takes too long.

The 152 volts is voltage that indicates you are either WAY too high for WAY too long or you are cutting "air". A plasma cutter does not throw that kind of voltage if it is cutting metal.

I keep coming back to the statements you are seeing the right voltage during a manual cut as it cuts along the path. It then points to a pierce sequence that is incorrect and the DTHC is reacting to the voltage it sees (way out of spec).

First you need to make sure your entire pierce cycle is short. Pierce dealy is not JUST the delay you program in SheetCAM. It is also the total time the torch sits over one spot so it includes the plunge from pierce to initial cut height time (plunge rate) and any delay the arc OK introduces. This gets more critical the thinner the material

Second you need to set the two delay times (Mach - THC delay) and (Cut Profile THC Fault Delay) so that the torch control does not react to the voltage spikes at the beginning. Once the torch has made the pierce and you have started the cut (cutting new metal) the voltage should start to come down. Assuming the move sequence is correct and the cut starts with the tip at the recommended cut height (intial cut height setting) THEN the voltage should stabalize as new metalis cut. Then and only then should the delays allow the DTHC to start controlling the Z. The THC Delay in MACH will let MACH ignore the frantic UP or DOWN commands from the DTHC and the THC Fault delay will keep the DTHC from triggering faults like the TIP SAVER and the other faults.

Start out with a 5 second delay in MACH and a 6 second in the Cut profile (they run concurrently and start as soon as the torch fires NOT when it starts XY motion. Once you get it to stabalize and cut (and you have the pierce cycle under control) then step the delay times down.

TOM caudle
www.CandCNC.com