Repairing a thirty-year old workcentre with 3D printed stainless steel

For those of you who are not Antipodean, the Triton workcentre is an Australian icon. How many other power tools have an owners’ club? Each one mounts a circular saw onto a carriage, mounts that carriage onto a table, and lets you cut wood across and along. In 1975, that was all any man needed to feel fulfilled.

The classic workcentre is the Mark 3, from 1984. The Mark 3 brought along a major step forward for the Triton world, the bright orange paint that they still use today. My Mark 3, like all good tools, came from the in-laws. It’s covered in shavings, a bit rusty, and does the bloody job.

For bonus Eighties nostalgia:

Or rather, it did, until I lost the locking pins while moving house. Bugger, as they say around these parts. The workcentre has a removable table and to switch from cutting across to cutting along, you have to move the table from the cutting across position to the cutting along position. To hold the table into place, there are two pins that slide into holes, lining everything up, and then a quarter-turn locks everything. Without the pins, you’re screwed.

Sweet, fine, they still make the parts, I’ll order two more. Oh… bollocks. They still make nearly all the parts, but they don’t make locking pins. Fine, I’ll make them myself… except they’re a really awkward shape. A bent peice of metal, I can do. In fact, most pieces of metal that I make end up bent, whether deliberately or not. But those two little forged lugs on the side? Those are a pain in the arse.

So, after some head scratching and the requisite degree of umming and ahing, I decided the way to go was 3D printed stainless steel. Given this is the Twenty-First Century, bodging up a repair starts with sitting in front of a CAD package, drawing away, with occasional trips down to the workshop with the calipers to measure the various holes in the workcentre table and frame. I’m not too flash with any of the various CAD packages I have to hand, but it’s a sufficiently simple shape that drawing it up didn’t take too long.

Sent the design off to Shapeways and, three weeks later, these babies turn up:

And then I realised two things. Firstly, 3D printing in stainless steel is really expensive. US$8 per cubic centimetre. Ouch. Still, it’s cheaper than a new workstation. Secondly, three weeks is a long time to wait when you want to know if your design is going to work. And of course, the first version didn’t. Or rather, what I drew fitted just fine, but I’d put the lugs at the wrong angle by ninety degrees. What can I say? I am a special and unique snowflake.

Anyway, the second version fixed that. To reduce the price by a third I cut the length down and cored out the centre of the bar which is the kind of thing you can only do when you’re printing. Here’s the two versions:

So now I’ve fixed the workcentre and the last twenty metres of skirting for the house are underway. A quick look at the Triton forums suggests there’s a few other people in a similar situation, so if anyone else wants these pins, then you can buy them from Shapeways.

POSTSCRIPT:
Of course, once I’d made the perfect replacement locking pin, what did I find?

Yup, the old ones were buried downstairs, in a box, ready and waiting to pop up just as soon as I’d sorted out the new ones. Anyway, at least I now know that other people can replace their locking pins, should they do a better job of losing their parts.

35 thoughts on “Repairing a thirty-year old workcentre with 3D printed stainless steel

    1. Well, it’s stainless powder, the big laser melts just enough of the surface so that the powder sticks together, then it’s consolidated with a brass infusion. 680 MPa tensile strength, so stronger than mild steel, not as strong as proper stainless steel.

      And I always bugger up the first one of anything that I make. I’m pleased to get the second one right, normally it’s the fourth…

        1. Titanium can be done. It’s just very pricey.

          I’ve not come across anyone doing (or even researching) 3D printed aluminium. I’m guessing that the oxide on the surface of any aluminium particle gets in the way – certainly back when people were sintered aluminium powder for high-temperature creep resistance, the oxides really don’t want to break down and let the aluminium get out from inside each powder particle.

          And aluminium is easy to cast, so you can just 3D print a plastic form, make a mould, then cast the stuff. That’s trickier with steel or titanium.

          1. If you were going to do aluminum, you’d probably have to do the whole thing, including making the powder, under argon. But you couldn’t infuse with bronze, maybe with zinc. I also wonder about titanium: I’d think the infusion would have to be pure silver.
            In any case that’s pure awesome, that they can do it.

          2. Aluminium’s affinity for oxygen is so excessive that you wouldn’t just have to make the powder under argon, you’d have to use argon with only parts per zillion of oxygen. Probably the best way of getting argon that pure is to react it with aluminium as you finely divided that aluminium into powder. Oh, and I’ve a nasty feeling that aluminium powder without surface oxides will stick to itself very effectively…

            For the titanium, my reading of the advertising bumpf is that they don’t infuse, their laser can melt the titanium enough that you get a fully densified result straight from the printer. Which is nice.

        1. Don’t know. I’m guessing that the material cost is small compared to the capital and laser cost. If you could have mild steel for nine dollars or stainless for ten, then I know I’d be going for stainless every time.

    2. I have a Triton Mark 3 work centre and had exactly the same problem … missing lock nuts and unable to buy replacements. The solution? Easy and inexpensive.

      Go to your local hardware and buy2 threaded bolts size M8 and approximately 100-120mm long. Make sure that you purchase 2 matching nuts for each bolt. Cut off the heads of the bolts with a hack saw and create a 90 degree bend in the remaining shank of the bolt by fixing it into a vice and hammering it. Be careful not to damage the thread while doing so.
      Now feed the end of the bolt through the first keyhole and screw on the first nut.
      Now continue to feed the bolt through the second keyhole and screw on the second nut.
      Now place he bench top in place and align it correctly. Screw the bolt a few turns until the tabletop is locked in place.
      The second nut is just the right width for the gap and the bolt locks it all in place.
      Problem solved.
      Regards
      John Thorley
      Sydney
      Australia

    1. Solidworks.

      I’ve a student license, which is a tad handy as it costs silly money. Then again, it’s worth it. Oh god, it’s so sweet.

      But for a shape this simple, you could do it in all sorts, 123D, SketchUp, OpenSCAD, depending upon which particular flavour of pain you want.

      1. Ah very nice. I’m self-constrained to open-source for this project, but it seems like there are a few options.

        I dig the tablesaw, looks nice and simple. thanks for posting it

        Mike

        1. It also turns into a sliding cross-cut saw, if you don’t mind moving bits and pieces around. Thirty years ago, this was the bee’s knees. Now table saws and mitre saws are much more affordable so there’s less need for a flexible workcentre like this, but I like it.

        2. Oh, and yes, this geometry wouldn’t be too tricky in OpenSCAD. Well, it’d be about as easy as anything is in OpenSCAD, which is to say a complete pain in the arse once if you’ve become used to a professional tool like Solidworks.

  1. As they say, the easiest way to find a missing something is to buy a replacement!

    I’d be stymied by the CAD package for starters, although I could always draft the youngest son in for that job (he’s just finishing up 2nd year Mech Eng at Canterbury, and seems to play with CAD packages for light relief)…

  2. when I saw your first picture showing what they should like I thought… “hang on… you’ve lost the pins but still have a photo take one day when you still had them…in your new living room….?”

  3. Small correction

    It was actually your mother-in-law’s Triton. (Not used for anything requiring great skill, I hasten to add.) Fred points out that when he has to cut wood he prefers to use an old bent handsaw…
    But we are both very glad that the Triton is proving useful.
    Mary

  4. Now with extra 3D

    I’m intrigued that printing steel was the option you went for here over, eg, using a lathe or mill and brute force to put the bend into it. Or welding on lugs. It seems like there are only a handful of dimensions that matter (corner to lugs, profile (and orientation!) of lugs, diameter of section that goes through holes) and the rest could be almost anything. Which ought to make it relatively easy to cut. (Give or take the issues of cutting and bending steel — especially if you’re determined to make it out of stainless steel rather than, eg, mild steel and galvanise coat it to reduce rust. For that application it doesn’t look like you really need the extra strength.)

    But the Star Trek Universe (go, go, gadget replicator!) bit is very cool.

    Ewen

    1. Re: Now with extra 3D

      The bend isn’t the problem, it’s the lugs and they are an annoyingly awkward shape.

      In traditional metal-working, there’s only three things you can do: subtracting, adding, or bending.

      Subtracting (cutting) to get that shape would be a mission – you can’t rotate the shape to cut it on a lathe. Cutting it out with a mill would either need a four-axis mill or you could cut something close with a small ball-ended cutter and then use a second operation to cut the sharp internal corners along the flanks of the lug. I don’t have any of that kit.

      Adding would be easier, either weld on beads and file to shape or cut a slot through the main piece and braze a flat piece in there. I don’t have any of that kit either.

      Bending, well, forging, would be ideal and that’s how the originals are made. You could churn them out for pennies each, but you’d need a great bit stamping press, specific tooling made from fancy steels, and a lot of set-up time. That’s how they were originally made, but they made hundreds of thousands of these things. I really don’t have that kit, nor do I have a factory in which to put it.

      So in all honesty, unless I wanted a few hundred, then 3D printing is probably the cheapest way to go, both in terms of cost per piece and design/set-up time. For a few hundred, I’d probably look at cutting slots and brazing pieces in there. Few thousand? Forging.

      1. Re: Now with extra 3D

        Yes, I appreciate the lugs are the tricky part. As you say I suspect adding the lugs (eg, beads) would be easier than subtracting all but the lugs — but lathing out a smaller outer diameter for the bit of the shaft without lugs, and then milling/filing to get the lugs shouldn’t be that tricky given suitable tools. At least to sufficient precision to be suitable for purpose (IME machining tolerances on such locking bolts are of the “nearest mm or two” variety).

        Possibly I’m biased by knowing someone with CNC machine tools. But I guess I’m kind of surprised that sort of service isn’t available (and cheaper) as a contract outsourcer. At least compared with US$8/cm^3. (And, yes, I agree if you’re making any decent quantity you’d just forge them. It’s rather impractical for small numbers though, as you note.)

        Still. 3D Printing. Of Steel. Welcome to the future.

        Ewen

        1. Re: Now with extra 3D

          For the milling, have a think about the tool path and tool shape needed. It’s doable, but not ideal and there’s quite a volume of material to be removed when you’re going from a 12 mm round bar to a 6 mm bar with 3 mm lugs.

  5. I’m guessing that you don’t have access to a welder? For something of that sort (not great precision needed, really) I would have taken a piece of mild steel rod, bent it, added a couple of weld beads where those protrusions needed to be, then filed down
    with grinding wheel and hand files until the shape was good enough. Sort of a PITA, but
    still not much more than an hour to do.

    1. Hmmph. stupid browser.

      For some reason my browser wasn’t showing me you last few comments until after I just posted that. Never mind, you clearly had already considered that option and rejected it based on lack of tools needed.

      The 3d stainless printing is pretty damned neat. I’m wondering what the mechanical properties of such an item are; If I asked them to print me a plain old piece of
      rod stock, and I pull/bend/work-hardening tested it, how would it compare to a real piece of stainless steel rod?

      1. Re: Hmmph. stupid browser.

        The internet answers: 3D printed stainless versus 18-8 stainless

        But in short, crap ductility and much lower melting point, everything else pretty similar.

        Oh, and it turns out that people are doing direct metal laser sintering of aluminium, and 6-4 titanium, maranging steel, nickel superalloy, and cobalt-chrome-moly medical alloy. Thus this machine is now officially on my Xmas present list.

  6. Thanks – I’ll remember that if I manage to lose the pins from MY Triton Mk3. But first I need to replace the broken knob. Luckily, I’m building a RepRap 3d printer …

Leave a Reply

Your email address will not be published. Required fields are marked *