Troubleshooting Beveled/Tapered Cuts

I’m working out the kinks in my machine, and having a blast knocking out projects. I’ve got a problem I can’t figure out, and am looking for some help. Perhaps this will serve as a reference for others.

The right (X+) and top (Y+) sides of any parts I cut have a bevel, of between 10-15*. What’s odd to me is that the X- and Y- sides of all parts all appear nearly perfectly straight. I’ve tried putting the pierce on a X+ side of a part, and the pierce appears to start at this bevel, then continue there, until the part turns a corner.

Where do I start? I followed the book in putting the table together as far as X/Y gantry adjustments go. I could wrap my brain around a consistent bevel of 10-15* on all surfaces, but this one is stumping me.

Pics to show what I mean. On this particular part, the edges with the white line/arrow–>T (taper) is the edge that has a bevel, and the other edges are all square.

Hypertherm Powermax 45xp
.250 hot rolled mild steel
Ground clamp on the material
Air pressure filtered and fresh dessicant dried, does not drop below 80psi
Yes, air pressure is not supposed to drop below 90, but the taper happens even if the cut occurs with >100psi

FWIW, all holes are beveled/tapered in this manner as well.

Pierce was top right in this example, and the zero of this program was bottom-right.

Thanks in advance.

Example of bevel on another part:

Sounds like your torch isn’t squared up 100%. May need to do some fine tuning on the adjustments to get just right.

1 Like

If it wasn’t squared, then wouldn’t it cut a similar bevel/taper on both sides of a part?

I ruled torch squareness out because of that.

Edit: I am running the machine torch mount, so there very little / no room for getting the torch out of square.

Have you swapped in new nozzle? Bevel like that is almost always caused by nick in the nozzle orifice. It’s unlikely that the torch is THAT far out of square since it would be extremely apparent to the eye.


Again, I would expect the torch being out of square to show the same bevel on both x+ and X- sides of the part, or both y+ and Y- sides of the part. Is that expectation incorrect?

Similar thinking with a nick in the nozzle. If one side of it was damaged, shouldn’t both sides be affected?

Rotating the consumables is something to try at least.

what direction is your GCODE cutting? when i export in Sheetcam, i run reverse cut path.

years ago, when I ran a larger torchmate table, and i got a taper like that, i would decrease the torch height.

it could be a number of things though.
poor ground
consumables gone bad
torch height
cut path
heavy millscale from the sheet. (why i only buy P&O plate, no millscale)

1 Like

Cut goes clockwise around the exterior of the part. Cutter comp is set to the left, and I think that’s why the direction is that way. It would be interesting if the tapers changed just by changing the direction of the toolpath though, will give that a shot before changing consumables.

Torch height is controlled by the THC. Can’t do much in that regard while that’s enabled as I understand it.

Will swap out consumables, check my ground again, and give it another go. Glad the first thing to check isn’t the machine, because I figured I built it incorrectly somehow!

Things to try for now:

  1. swap cutpath direction
  2. swap consumables

Thanks for the ideas!

Swapping consumables fixed it. Crazy, considering how little I feel like I’ve cut, but will have to keep an eye out for that I guess! Thanks for the replies :smiley:


I just started using my Pro for production. Previously used two xfires since batch 1. I cut around 1500lb + of steel a month. I never had much taper issue even with shimming up my warped sheets and prolly cutting a bit too high for most of the time. My consumables would get me on average about 30 sheets that had almost 100 pierces per sheet. So Im pretty in tune with knowing about the taper/bevel.

With the Pro I seem to have it worse than I ever had. In fact I just changed consumables and reset my machine torch. BTW my Hyperthem machine torch would hit the Knob on the torch holder from the machine. So I ground down one of the ends so now the machine torch will push back farther and not hit that knob. I’m also using the machine torch holder by Langmuir.

I have the Hyp xp 65, been using it for months. Used the 45xp bf that. The settings my book calls for is a pierce of .15 and a cut height of .06. This is valid for all the thicknesses Im cutting, 10g up to 1/4".
Changed consumables several times Im getting worse bevels than on my crossfire.

Iv checked my torch multiple times. I had some slop in the Z axis and have tightened it up. But still just very odd bevels on pretty much everything. Using very similar cut settings to when I was on the reg crossfire.

I have tried a ton and still get bevels. Luckily nothing I make is critical. I thought it was just part of the process. But I feel like Im seeing this more around here lately. There has to be a “simple” solution

I recommend checking your cut height with some feeler gauges and then adjust your IHS settings from there. There’s nothing really inherently different about the PRO, still just turns the torch on and off and moves it around the sheet (albeit with better control of standoff height than the OG).


I did check it with the thick spacer from the crossfire. I paused my program as it did its standoff and it was not quite touching the thick shim/spacer. I thought about trying some programs with a lower setting, but honestly when I cut with the crossfire I’v had some serious off contact at times and bevels just weren’t an issue. The PRO looks like its cutting just fine, in fact looks like its cutting much lower which would be better. I also have the torch as level as can be. I’v noticed the bevel worse on the Y axis. I checked two identical parts from a run, one cut vertically and other horizontally and one had more bevel. I under stand speed of torch and direction factors. I cut clockwise and cut 125 IPM on 3/16 at 63 amps with the Hyp 65 using regular consumables and machine torch.

I was just wondering if people are having this problem because of the initial “bounce” of the material. So would it be better to off set the Pierce height? But even that theory doesn’t seem like it would work bc again when cutting with the crossfire vs Pro it always appeared like I was cutting higher up.

Another theory I have is Im not perfectly sq with my Z axis. Example when setting the bearings to the X axis rail, Im thinking they are pushed too far one way creating a tilt on the whole setup.

I’ll post up some pics, bc I know those really will help paint the picture. For me I really think my Z axis must not be centered up. Im noticing a more distinct bearing tack on the bottom of my X rail, which leads me to believe all the weight of my Z axis is pushing from the top and straining the bottom bearings.

what is your plasma? and do you have pics of the torch setup?

If the bevel angle you are getting is the same as the torches angular misalignment, then that would be the cause. However, I’d be surprised if your eye caught the bevel on your parts but does not catch the same angle of the torch. I am guessing that your torch is probably perpendicular within less than a degree.

I’d have to go dig mine out, but wasn’t the thick spacer for the original 1/8" (.125") and the thin spacer was the 1/16 or .06" spacer?

If you’re using the Crossfire shims you’ll want to check against the thin spacer, otherwise you’re too high.

For the pierce it calls for .15 and cut of .06. When I watch my Z axis as its running the numbers are spot on. Just wondering if this has been a common problem, like the bounce compensation or maybe the adjustment of the Z axis bearings. I have some video of the Play that was goin on, I tightened up those bearings that ride on the Z axis rails. That seemed to atleast eliminate some play in the torch.

I’ll be playing with it but Im pretty stumped. Other than that things are going good. Couple program quirks here and there but overall been able to cut for hours on end.

If you mean you were watching the DRO of the Z axis, it will not change when the THC makes adjustments. The two systems are independant. When THC is active it takes control of the Z axis and any changes in height are not reflected in Firecontrol other than the Up/Down lights. Hopefully at some point they can better integrate the unit with Firecontrol so the actual height is displayed, along with logging, so that the user can better diagnose THC problems when they occur.

Yeah I wasn’t meaning actual height of THC while cutting. What I am referring to is the initial pierce on pieces that have alot of bounce. Like today I had a sheet really bounce and just wasn’t sure how accurate it is with the bounce. So I had like a 5" area that had 5 holes and each time it cut a hole the sheet bounced. But it seemed to fine. I squared up my torch holder today. man this this is a booger to get square. I had one bearing loosen up several times during the day. If you over tighten one bearing it moves this piece out of sq. This is when I checked it and then I sq it up after this. Cuts today were a bit better, not perfect but much better.

Aside from adjusting cut height in post, is there a way to adjust IHS? I’ve found it necessary to specify a very low cut/pierce height in post to compensate for what seems to be incorrect IHS sensing/compensation.

Edit: I’ve looked at the simulation (fusion 360) for both pre and post output gcode simulation and both show a tool height of 0.00 on the cut path. I’m unsure what exactly the issue is but I assume its related to whatever is coded in post for IHS.

Please chime in.

I have leveled mine, taken slack out, squared up, and everything. I have stopped cut and measured cut height exactly at .060". It doesnt matter I still get bevel around my whole cut. Thicker material is better. But I cut mostly 14 gauge and like I said bevel bevel everywhere. Some say a machine torch works better. I’m beginning to think its not the actual machine torch but the quality of the machine torch or lack of quality of the Razorweld torch.