THC problem

[_Duke_]

I decided to try the ol dog & pony show of cutting some corrugated steel, and all I got was crash city.
I've had some issues with the THC diving the torch into flat material in the past, but now it's pretty apparent that it's not working at all. I'm using a 45xp, so the book values for voltage and travel speed are pretty spot on.

The material I was trying it on was a section of a shipping container, 1.5mm corten steel with pretty shallow slopes, maybe 25-30 degrees. Slowing the feedrate didn't have any effect.

I checked out the voltage on the pid config graph and ran the "run program" whatever that does, but I have no idea how to interpret that. Photo at the end of my build album

I also noticed a vague note in the user's manual on page 56 stating to rerun the thc config program every time you make a change to the axes, but no other reference to it in the entire manual. So not quite sure what that is either?

[_Duke_]

Ok, I guess I was being a little dramatic earlier, there is something working with my THC, at least I haven't fried it.
I saw a very helpful video Joey shared on another post about how to interpret the plots:

I ran a 4" line test on some 16ga and it was almost identical to the one in the video. So that's a good start.
I then ran the same 4" cut on the corrugated. The pierce started on top of a plateau about 2" from a down bend, and I let it burn across the edge of the bend to see what would happen. Just like my previous corrugated tests, the THC response was imperceptible. But according to the plot (attached) it was trying to do something until the THC control enable cuts off for some reason. Is that what the purple line THC Dive is for? to disable the THC when it detects too steep of a change? I tired cranking my Anti Dive Threshold up, but the cut was the same.

I feel like the answer to cutting corrugations lies in changing the THC settings from stock outlined on p40 of the manual, but if there's anybody that understands all that better than I do, feel free to chime in while I try to wrap my feeble brain around it.

[Joey]

Your making progress.

The purple is the THC anti dive. This is controlled by the Anti Dive Hysteresis in the THC config.

If the voltage rises more than 7 percent of the target voltage then anti dive will kick in. Default settings is 7 percent o believe

I would increase the anti dive from the stock 7 percent number to something higher like 15-20 percent.

The anti dive velocity will prevent THC if the speed gets below 90 percent over the target velocity. I would change this setting to about 80 percent.


Cant you get any faster X and Y accel/decel rates? The plot makes it look like the accel/decel rates is pretty big. Maybe even the default .5 setting.

I'm not a PC to check out your report atm

[_Duke_]

Hi Joey, that helped a bunch. I see big words like hysteresis in the settings and my brain kinda fogs over!
I've got the anti dive threshold up to 25%, and I tried running the 16ga corrugated on the 12ga profile to slow down the feed rate, and that seemed to help too.

I'm not sure if I just dial back the feed rate override on the 16ga profile, will that not allow the THC to enable since it's below 80%? Guess I could try that and find out.

I was able to get a successful albeit drossy cut if I start the pierce on top of a corrugation and go downhill

But if I start the cut down in a valley and then try to have it ramp up a corrugation, it still can't get out of the way fast enough and crashes

Is it weird that the THCoffsetCnts line is going in the positive direction on both uphill and downhill cuts?
I also changed the axis accel/decel rates to 0.2 seconds, you're right they were still at 0.5
I created a new report too since so many things have changed since the last one

[Joey]

For reliable THC operation always run a job with Feed Rate over ride set to 100%.

I would set set the Cut height slightly higher to give the Z more time to react if possible.

Or dial back the Feedrate. Try to cut at the slowest Feedrate and amperage as possible to get the Z more time to react.

Tune the Z config settings for a max rate close to 500ipm and the accel/decell rate as low as possible.

I would make a custom profile.

Find out what the lowest amperage you can cut with.
When you find a desired Feedrate to cut at the low amperage setting then set the Target voltage to what the THC is reading when cutting at the new Feedrate/ cut height settings.

Depending on the slope of the corrugate the torch tip will make contact at the default cut height of around .06 so a slightly higher cut height may be needed.

I think on the up hill cut the THC started raising the torch but towards the end of the plot the torch tip made contact with the material and started burning a bigger kerf width causing the voltage to climb and hence the Torch started lowering.

Smoothing should also be used when cutting corrugated to keep the speed up with any X/ Y direction change.

I can see the Pierce delay looks a bit long in the plots and the THC delay could be shortened slightly.