THCAD-300 and Hypertherm (65)

So I finally have my table running, and I even tackled my first job last weekend. I experienced a few issues that I need to work through, but before I burn up a bunch of material, I though it might be prudent to ask here first.

1. I regularly get an error that the torch flamed out before the M5 command was reached. I'm 99% sure this is because of my lead-outs, which in my ignorance I made the same as my lead ins (basically .25" radius arcs). I'm wondering what you guys recommend here for a lead out? It seems to me I have two options: A. figure out the lead outs, or B. put the hypertherm in corrugated mode where it basically locks the torch status to on, and let it get to the M5. Option 2 seems like it will consume electrodes faster though.

2. I cant seem to get the THC to keep up with the X Y movements. I am almost positive it's a settings thing, because at the beginning of the cut it hits the voltage I input (134v in this case) nearly perfectly. But once it gets to a part of the material I knew wasnt flat, when I looked down it had gone above 160v, and the "VOID LOCK" was lit up. This happened on 18" long Y axis cuts, and the 36" long X axis cuts. I'll take any advice I can find here. Is there standard settings for a MESA THC and a Powermax 65?

One thought I had was maybe the movement per volt is off (.004" per volt is the standard I did not mess with), but I can't find anything to correlate to this in the Hypertherm manual.

Basically all of the THC settings are at default right now, the exception being the scale and offset of course.

Any thoughts or help would be appreciated!

-Greg

Are you using PlasmaC in mode 0 or mode 1?

Void sensing is not very good yet, I have not done any real work on it to improve it. It is probably best to run with Void Sense Enable disabled.

The THC motion should run at the maximum available Z velocity.

Can you post a screenshot of you Run Panel and your Config Panel plus your ini file.

The .004" is a rule of thumb setting from Jim Colt of Hypertherm

I am running in Mode 1.

That's good to know. I was using it in conjunction with issue #1 in my Original Post. In those areas where it flames out, it would start to dive.

Current Z velocity is 110in/min for me (I think), it will be higher at some point in the future as I plan to change to a ball screw over a acme thread.

I attached the .ini and requested screen shots.

Thank you for the help!

That's good to know. I was using it in conjunction with issue #1 in my Original Post. In those areas where it flames out, it would start to dive.

If it flames out and your Hypertherm is in normal cut mode then it should lose Arc OK and then PlasmaC will pause in an error state

There are a few guys using Hypertherms, maybe one of them will chime in with some hints.

1: No or very short lead out, that causes the torch to chase the falling small parts and subsequently loses the arc prompting the machine to strike the arc again........
2: Disable void lock, and set the Z axis to move faster, much faster in your case.
After you do that, you can try disabling the auto volts and find a voltage that works good for you.

tommylight wrote:

1: No or very short lead out, that causes the torch to chase the falling small parts and subsequently loses the arc prompting the machine to strike the arc again........
2: Disable void lock, and set the Z axis to move faster, much faster in your case.
After you do that, you can try disabling the auto volts and find a voltage that works good for you.

So, I am currently 110 in/min in the .ini. What is a good rule of thumb here? Also, how do I keep it from diving on kerf crossings, etc?

For the record, I do normally disable auto volts, but I have a glitch going on where saving it turns it back on.

I did forget to mention that the Height Per Volt setting is only used by Void Lock which doesn't work too well so it won't affect anything else.

rodw wrote:

I would disable the leadouts altogether. They just got me in to trouble.

I would increase your THC delay to at least 1 second (possibly 1.5 seconds). The voltage spikes on piercing and takes up to 1.5 seconds to settle down to a stable voltage. See: linuxcnc.org/docs/2.8/html/plasma/plasma...imer.html#_thc_delay
So what could be happening is the sample of cutting volts is a lot higher than it should be before it settles to a stable voltage so your torch is lifted to increase the volts. This means you are cutting higher than expected. Then there is less leeway with loosing arcs... because you are further away from the material. (0.5 seconds is cutting it pretty fine based on my tests)

Jim Colt's rule of thumb for inches per volt should be OK. It is based on 10 volts per mm. (0.004" or 0.1mm per volt) In my tests of 16000 readings with a 50 amp Everlast I measured 7.53 volts per mm with 99% confidence. See this graph for more info linuxcnc.org/docs/2.8/html/plasma/plasma...mer.html#_cnc_plasma
This is really the only source of info about volts per mm I've seen on the web and its based on my research!

I don't think THC delay or the volts per mm varies significantly between machines as there was not much difference in Arc characteristics between my Everlast and my current Thermal Dynamics 120 amp machine.

Thanks for the tip, and the links! That is definitely helpful.

Sounds like between that, and having a faster overall Z axis, gives me things to try for sure.

Are there any more documents that cover the rest of the THC settings in the config tab? threshold, PID, VAD, Void override, etc?

phillc54 wrote:

I did forget to mention that the Height Per Volt setting is only used by Void Lock which doesn't work too well so it won't affect anything else.

This is really good to know. I did ask Rod here as well, but are there any more documents that cover the rest of the THC settings in the config tab? threshold, PID, VAD, Void override, etc?