Fail-Safe Break-Away Torch Holder - Prototype

Fail-Safe Break-Away Torch Holder - Prototype


The fail-safe break-away torch holder prototype is working out great. I just need to change the mounting position to get the torch in closer near where it was originally, add a gusset where the magnet plate flexes at the bend, run the switch wires out the bottom, add screw stop to make the torch clamp angle repeatable, and put a knob on the clamp bolt for tool-less torch removal.

It has one bolt clamping the torch in, it doesn't have to be much more than finger tight, it holds the torch very straight and repeatable, so it takes about 10 seconds to remove the torch and put it back in, without having to re-square it every time. It has totally separate and fixed plumb adjustment in the X-Z plane and X-Y plane, so it is easy to fine tune torch alignment.

The magnet plate on the torch sits on top of the fixed magnet plate so it doesn't rely on magnet strength to carry the weight, the magnets just hold the torch down, and the torch doesn't fall when it does break away.

The switch trips open before the magnets completely disengage the pockets that they ride in, and it's a 3 point alignment, so it holds perfect alignment and trips if the plates come out of alignment in the slightest. It is wired as a limit switch so it detects all collisions during cutting and during rapids and plunges. With Mach3 soft limits, you can just soft-reset, back away, clear the obstruction, and go again.

It is surprisingly easy to make. It just takes a 5/16" drill, a tapered plumbers reamer, and a 90 degree countersink to get the magnets just right. And the torch clamp just folds right into shape so easily and works like a charm.

I will explain it all better later, I still have some testing and refining to do on this and the touch sensor circuit. For now, I just needed to get this much done to proceed on further into basic height control. It's a big relief to have the prototypes working. Stay tuned...

By heck that looks a great bit of design work there Gary!
I like the 3 point ball idea, might have to re-model mine after seeing that, I slipped up by only having two.
I din't fit any trip switch on mine neither.
Well done.
thanks, it shouldn't take too long to finish the design and post instructions, press fitting the magnets to the right depth in the right size hole was the trickiest part.

I used 3/8" ball magnets from and the switch is a normally open super-micro detector switch from, on 1/32" pcb material it came out just the right height to fit between the plates, but it's so small I totally lost one just taking it out of the package, and the contacts are only rated for 1 ma at 5 volts, so I might have to try a more durable compact through-hole push-button type switch.

The big mystery now is why the Y axis makes a noticeably wavy edge finish while it cuts very smooth along the X axis. I suspect it must be torch alignment and/or I must have the wrong tension springs on the Z axis rollers because they're not holding very tight. I'm hoping that goes away with better alignment and stronger springs. I shored up the table and grating but that didn't help.
Always something! I thought there looked to be a bit of a zaggy cut looking at the bracket on video, does your table have a motor on either side of the gantry? is it all running smooth down the rails, hopefully it's an easy fix.
You have a lot more thought gone into that than mine.

Plenty thought went into the table designing though I can say for sure ;)
Cnc Plasma Cutting Table
I'll have to check everything. The wobble is so consistent I'm thinking it has to be the torch or the rail. It has an axle to drive both sides.

Your machine looks nice. Is it new? Is the carriage riding on one rail? It looks (and sounds) rigid and solid.
Thanks, I started out on the project about 18 months ago, it's taken a while to iron everything out and start getting to grips with all the software side of thinks, but getting there now I think.
They are 30mm rail/carriages sturdy enough there, 2 blocks on X and 2 on either side of Y.
It's a strange one that especially being driven on both sides.
Sean, I have to say that is a very comprehensive pictorial of your table build, thanks. As Gary has said a lot of thought and planning has really paid off for you.
Can I ask, do you know your total gantry weight?
Great Job'
Thanks Murray, I did do a final tally of 120lb and that's using CandCNC's 620 oz motors, geared at 3.75:1.
It works really well, I possibly get a little inside corner cutting on less that 1.5mm thick, but that can be dialled out either with the new rules or using the settings below in Mach.


Thanks for that Sean, Nice looking table and lots of neat ideas. I thought it looked like it was up around that figure, the SHS beam looks to be about 75 x 75 x 6mm going by the radius so figured there was about 22Kg there . I'd be happy with 8 metres per minute, it's more than I can currently run at and still maintain cut quality.
I like the way you have used RHS to build your reduction drives too (along with pretty much everything else you've done )
I'm working on a new table build and have been looking at the Hiwin type flat linear rails and also the round shaft type linear rails. Is your Z axis rigid enough with running it on the single rail on the gantry? I was thinking parallel but if I can save weight and maintain rigidity I'm more than happy.
Thanks again
Your not far away with the sizes there Murray, it's 70mm x 70mm but only 3mm thick, my original idea was to use nutserts for the linear rail mounts but the fixing tool wan't up to the job so I just tapped M8 with all the rail bolts it seems solid enough to me, tapping the 3mm box is more than enough for the rack fixings as well.
There is zero movement just using the single rail especially using the 30mm, I reckon 25mm would be enough really, if using as a router it might be a different matter I suppose.
Any slight flex comes from the backplate on the Zaxis (bought from Velox), but again for plasma it's fine.
Yea the reduction drives were just using up off cuts from the top rails, those bearings worked really well as they are adjustable so could get a perfect mesh on the drives.
I got a few ideas from dnelso's table on cnczone he used a fixed pinion gear but that would take a lot more work and accuracy, those spring loaded tensioners which again wasn't entirely my own idea really work well, I don't see any backlash at all.
You can email me at seanpalmer 'at' if you need any other info no probs.
wow, I totally missed the link to your photo gallery, duh,,, that is nice,

my torch holder is a bust, must rotate the magnets 180, the torch handle wants to pivot downward, that's where all the weight is, having one magnet opposite of that weight is backwards as it breaks away too easily in that direction

on the waviness I went to try my one spare set of consumables and they just won't cut through right, while the original set works fine, so aaarrgh....
on the waviness I went to try my one spare set of consumables and they just won't cut through right, while the original set works fine, so aaarrgh....

Gary, it may sound a bit odd but I was getting serrated cuts like that when my swirl ring got all oiled up and dirty. I thought it was something mechanical and chased it for weeks, replaced bearings, vee rollers, cleaned the rails, etc. etc. Replaced the swirl ring and gone.
Interesting, the ring was new to, but since I didn't pick up an extra one that's probably what it is. There's also a different fine cut shield and swirl ring (with one row of air holes) for hand cutting, and I'm not quite sure which swirl ring should be used for machine cutting with the hand torch, I think I tried both, but I'll have to check....
I run a PM 45 with standard consumables and the swirl ring is the same for both hand or machine torches. The electrode needs to be able to freely slide inside the swirl ring to maintain an even arc distance to the nozzle with a blowback system as used by Hypertherm. Overtightening can also damage the swirl ring I'm told as it can deform it. I'm also told to only clean it with warm soapy water and blow it dry if becomes dirty.
Ok, version 2 is good enough to paint, still needs wire routing and set screw to hold repeatable torch handle angle, but should be able to make a final version with a good instruction video soon. It trips me out how easy it is to make, and how easy it is to square the torch with the table.

But, now I've got a new idea on a way to use a router that is going to be awesome, so I might try to work that in.

To rotate the magnets I had to make the removable torch plate straddle the fixed magnet plate. And I added some torch lead holders that not only takes that strain off the little magnets but it also supplements the magnets by holding straight down in the very center of the torch plate.

I give up on fine-cutting 10 ga for now. One worn set of consumables cuts fine, all others require 47 amps and reduced speed. 45 amp consumables work a lot better for this sort of cutting anyway.

The wavy cut is almost definitely a table vibration issue, it's been there since day one with 3 different torch mounts, fine and 45 amp consumables, new or used consumables, anywhere on the table, in 18 to 10 ga material. I think I just need to build a hand cutting roller guide and make a test cut by hand to confirm for sure that it has nothing to do with the torch.

For now, I must quit playing with it and get some work done. That's so hard, I must have triple OCD.

Top Bottom