Prototype Torque-Limiting Set-Screw Wrench

[HO RacePro]

Awright, so I'm bragging. I think I've earned the right to a little brag.

Them tiny set-screws used to hold wheels on 1/32nd scale cars are easy to strip. Worse, the threads in those pricey aluminum wheels are easy to strip too. Which is why torque-limiting set-screw wrenches were invented for slotcar use.

Problem is, they aren't always available. And they do cost a fair bit. Also the tips on them can wear and need replacing. In theory the tips are replaceable, but I have yet to see the replacement parts for sale anywhere. Nuts.

So why did I design my own? Well heck, you know I love a challenge. And once one gets its hooks into me its either meet the challenge or load it into my permanent things-I-fuss-about-buffer, where it can annoy me pretty much forever.

So yeah, I designed my own. And I am obnoxiously proud to declare it a success.

My main idea was to start with a standard less-than-a-buck allen key and build a ratcheting housing around it. The ratchet limits how much installation torque you can apply, but not the removal torque. My design uses two parts for the housing and ratchet, plus a steel coil spring to load the ratchet. I 3D filament printed the housing and ratchet parts in PETG.

Using a standard allen key means the part is easily and cheaply replaceable. Or if you are really miserly (like me!) you can simply take a Dremel to a worn key and cut off the end for a new, sharp tip.

Since this was my first ever prototype I did not expect it to go together without a fight, and I was not disappointed. I needed to spend some time getting the holes sized correctly, I had to tap a 6-32 thread, and I had to figure out how to assemble the cuss without launching that danged spring across my workshop.

But, a bit surprisingly, I got my very first attempt to work. That doesn't happen very often.

So here are some pictures...







This particular tool is an 0.050 inch hex driver. I haven't calibrated the torque-limiting feature yet. The allen key I used was just one I happened to have laying around. I probably would not use a ball-end key for a production version. I'd also probably shorten the hex key a bit. I have a tool designed for cutting hard piano wire that does a fine job cutting small allen keys. I used that tool to trim the short end of the key so it's only just long enough to engage the ratchet.

At least in theory it should be possible to adjust the torque-limiting feature with the screw. That might need a little more refinement.

Anyhoot, there's the thingy, and I'm proud of it.

Ed Bianchi

[Kevan]

Nice one Ed 💪

[ourwayband]

Very Nice...

[4424ever]

Ed the type and frequency of your posts suggest to me that you have a lot more time for hobbies than I do
I like this one and where is heading 👍

[HO RacePro]

I'm retired. No money. Too much time. Happy as a swine in muck.

[Kevan]

Sounds wonderful Ed

[Kevan]

I like the idea of printing one myself as I don't have a 'real' one 🤣

[HO RacePro]

Kevan, you wouldn't be asking me to share my STL files would you? I'd ask you to get me a trackside seat for the TT in exchange, but I'd have to fly ballast class to afford the trip.

Let's wait until my design gels a bit more. I'm still trying to reduce the design to something that goes together sweetly right off the print bed. As I am sure you realize filament printing isn't always WISIWIG, especially when small features are concerned. As a result over the last few days I have made a whole bunch of minor tweaks to this dimension and that feature. As of this writing I have done -- what? -- 5 or 6 prints in search of the sweet one. But I am getting close.

Here is the latest complete prototype I've made. But I do have another body being printed as I type.



You might notice I've added a part -- that short stepped collar I'm calling the ratchet. It makes the operation a whole lot more consistent.

But I still need to buy a bunch of compression springs and hex keys to make the thing production ready. And then I'll almost definitely need to do some more tweaking. Still in all, good progress.

[OXO cube]

wonderful design Ed
almost everything in this hobby is way over priced imo
i understand its hard to make money from a quite small customer pool
hopefully when you do make your files accessible , its still affordable to us paupers

[Kevan]

Thanks for offering Ed...sadly I stopped watching the TT years ago, 5 riders have been killed this year and it leaves a bitter bitter taste I'm afraid.

[HO RacePro]

I understand. I follow Formula 1, but even a big crash where the driver walks away is still upsetting. You have to understand that the death of Jules Bianchi hit home hard.

As a former motorcyclist myself the idea of racing on a course lined by stone walls and trees is chilling.

[Kevan]

There's rider deaths every year, on average more than 2 every year. F1 used to be like that but change was forced and the really dangerous circuits were either changed or abandoned.
They say road racing is more addictive than crack and that's why they do it, they're addicts and can't help themselves.
There's a lot of ex-racers who suffered life changing injuries and would probably still race if they were capable.
I can't wave the IoM TT flag I'm afraid, even though I can hear the bikes racing past at the end of the road not 100 yards away.

[HO RacePro]

I originally posted on June 6th. Here it is June 18th and I'm finally ready to stop calling my torque-limiting wrench a prototype. I now have a production model. Call it the Mark 1.



What makes it a production model is the spring and the hex key are both catalog items I can buy from a reliable source -- McMaster-Carr. That, and I have pretty much run out of dimensions and features that I feel need tweaking.

One big change is the nose of the tool is now hexagonal. That allows me to fit a standard 1/2" wrench on it so I can calibrate it. Another change is the cup now sports my new 'Bee-On-Key' (Bianchi) logo. Very small and on the limit of my printing resolution, but hey, it's important to trademark my creation, yes?





And now you know how to pronounce my name. Not that I get huffy about folks who mangle it. I've had seventy years to get past that issue. No worries.

Now the question is how to market the gingy. Yes, I can offer STL files, but then the customer would have to supply his own spring and hex key. And perhaps more of an issue, tap the body for the retaining screw, which he'd also have to supply. Then there is a short bit of brass tubing that lives inside the assembly. That would need to be sourced, cut and fitted.

Finally, there is a trick to assembling the tool. That spring is pretty stiff and puts up a real fight. I've figured out how to do it. But if you don't know the trick you'll be struggling, and launch that spring, guaranteed.

Of course I can sell the complete ready-to-rock tool as a mail-order item. But believe it or not the cost of shipping could be almost what I'd charge for the tool itself. Shipping costs have gotten nutty.

I have thought of offering these tools as a set of three -- 0.028", 0.035" and 0.050". That covers the common set-screw sizes used for 1/32nd scale slotcars. That would make the shipping cost less of an issue.

I suppose the best idea is to offer the tools individually with the option of buying a set.

So what would they cost? Oh geeze. Don't hold me to this, but I was thinking maybe US$15 each, with shipping extra, or $45 for the set, shipping included.

How does that sound?

[MrFlippant]

I think you're right that your design is probably not suitable for customers to download and make. If you have to buy tools (taps) to make it, that's a bit too far for most.

Have you considered working with a retailer? Make a bunch at whatever price you want for it, and then they can sell it with whatever markup they need, and people can toss it in with their other slot car stuff when they order from them.

Also, I'm sorry if you mentioned it before, but what is the torque value set to? And since I don't know it off the top of my head, is it more/less/same as the slot.it torque driver?

Edit: oh, and is it possible for the end user to replace the key, in case it ever gets messed up?

[HO RacePro]

Mr. Flippant,

I've worked with distributors in the past. Pluses and minuses to that. If I get enough sales I'll consider it. But their markup could make the tools just too expensive.

I'm working on the torque. I made one measurement today and got up to 11 inch-ounces before the ratchet snapped. That is in the same range as the Slot.It tools. But the torque is adjustable. I need to come up with a good way to measure torque so I can calibrate the tools. And I need to discover how much torque can be applied to the three sizes of set-screws before things get damaged. There are instruments known as torque watches that appear to be suitable for those measurements. Quite a few on eBay at manageable prices. I'm researching them.

And yes, it is possible to replace the hex keys. That was a feature I designed in from the get-go. But I did tell you that assembling these tools is tricky. It will be much easier to trim and dress the end of the tool to get a new, sharp end. You should be able to do that quite a few times before it gets too short.

I guess I'll need to create a YouTube video to show how to replace the hex keys. Not too hard once you know the trick.

If I get enough sales I'll be able to justify a new printer. Creality makes one with a conveyor belt instead of a print bed. It is designed to run continuously, printing the same parts over and over, unattended, until the filament runs out. The parts can fall off the end of the conveyor into a bin. Last time I checked that machine retailed for just under a kilobuck, US. I could swing that.

https://www.creality.com/products/cr...dex.header_1.1

You didn't mention the bee. I'm disappointed. It is cute, yes?

[MrFlippant]

Heh, yes, I like the logo. Very cute and clever.
I now understand that the key can be replaced, but it requires some disassembly and reassembly. Grinding off the rounded tip makes a lot of sense, as you say, it could be done many times before the whole key needs to be replaced.

I look forward to your find decision on pricing and stuff. I'd like to get some other sizes than just the slot.it one.

As for the belt printer... I'd hold off for a bit. They are even more fickle than a regular printer, and if you're already getting good, reliable prints for your machine, I'd sooner buy another one of those than the belt printer. Also, depending on your printing surface and how easily parts can be knocked off, there's some gcode that can be added, so long as the printer supports it. I did this at the beginning of the pandemic...

[HO RacePro]

Mr. Flippant,

Verrrrry interesting...

I've never mucked around with home-made gcode. You've shown it can be done, and that it can be useful. But for what I'm doing I don't think it would help.

And I take your advice about the belt printer seriously. The parts I'm printing aren't the best for that angled print head. What I could do instead is fill up my bed with parts. Just printing the three parts for one wrench takes an hour. How many parts could I fit on my bed? Dunno, but probably enough to keep it busy for 8 or 12 hours. That might work out okay.

Also, for what a belt printer would cost I could buy three, maybe four conventional printers. Which means I could be printing 4 or 5 times as fast.

So look. What if we say the wrenches cost US$15 each plus shipping. What sizes would you be interested in? I'm already planning on offering 0.028, 0.035 and 0.050 inch sizes. It should not be hard to supply other sizes if you need them.

Here is a link to my ordering instructions page...

http://box5921.temp.domains/~bianchi7/?page_id=49

[MrFlippant]

It did take me some research and trial and error to get that code working, but once I did, it's easy to replicate. This is for a Prusa printer, but I believe Creality uses the same fundamental firmware. Whether they've chosen to support the necessary commands, I can't tell you. The other trick is that the bed needs to hold the material firmly enough for a successful print, but also completely release it when cool, so that the pushing code doesn't just break the print.

If you're confident in your printer, then filling the bed is a great way to get a lot of parts done, especially over night. It might be worth only putting on the needed number to run overnight, and then you can start another bed in the morning. Even with my printers, I'm often wary of filling the bed, especially if I'm not 1000% sure of success. One detached print before they're all done will ruin ALL of them.

Another option worth looking at is sequential printing. I know PrusaSlicer supports this (and so the latest versions of Slic3r and SuperSlicer) and that it supports a variety of printers, including most of the Creality offerings. The trick is to order the parts on the plate, and to provide enough room between them, that it can complete one print, then move over and start another one from the plate, without ever hitting the already completed print. This way, even if one of them fails at some point, the rest are not ruined as a result.

Anyway... I'd probably get all three of your wrenches. I don't know the measurements off the top of my head, but my cars consist of slot.it, NSR, ScaleAuto, RevoSlot, BRM, and some others. So long as I don't have a useless wrench, I'll be happy. If there's a size you know I'll need that you don't offer, I can wait for that, too.

[HO RacePro]

Mr. Flippant,

As always, interesting and informative...

I once used the sequential print feature on my Creality printer, so I know that is an option. And an intelligent one. Of course I have forgotten just what command ju-jitsu was required to get that working, but I'll dive back in and puzzle it out.

I've never worked out how to get parts to break free when the bed cools. 'Nother thing I should research. I'm still using a tap with a hammer and a drift to release parts. That still works far better than you'd imagine.

You should never have a useless wrench. The printed parts will accept a wide selection of hex keys and they are replaceable.

The tools I use to replace a hex key is a hemostat clamp, a half-inch box wrench and a small slip-joint pliers. I insert the new key in the body, sliding it in almost all the way, then clamp the far end to hold it in place. The collar goes on next -- a press-fit on the end of the body. The spring is installed next. It takes a bit of fussing to get the end of the spring past the 'L' of the hex key.

Now here is the trick. I rotate the spring so its coils wind past the 'L' of the hex key. I hold the body against rotation with that 1/2" box wrench, and use the slip-joint pliers to grip the spring and turn it. It does take a fair bit of torque to get the last few coils past the 'L', which is why the tools are needed. Doing it by hand is just too hard.

Once the spring is in place I slip on the ratchet and the nut. With them in place I install the screw -- not all the way tight. The ratchet needs some space to move. Once the nut is screwed down I can release the hemostat clamp.

Voila! Assembled wrench with a new hex key.

I hope that all makes sense.

[MrFlippant]

I think I follow, but a video would certainly help. ;-)
I'll see if I can dig up the actual sizes for the wrenches I would use most often.

[MrFlippant]

As for the print removal... I've been spoiled by the magnetic flex plate system on my Prusa. I can't imagine having to use tools to remove prints from a solid bed. I know you can get such mods for other printers, WhamBam Systems probably sells what you'd need. The textured flex plates work really well for ABS and PETG materials, though PLA doesn't like to stick to them. I usually use a smooth PEI sheet for PLA prints. Because of the texture, and expansion/contraction differentials of the cooling plastic and metal, the prints are usually completely free once the bed has cooled down. There are other materials that have similar properties. Angus (MakersMuse) showed off using sheets of G10 not long ago.

[HO RacePro]

I still need to calibrate my wrenches. And determine just how much torque the different sizes of set-screw can tolerate.

Dunno how much difference there is between 'not enough' and 'too much' torque. Rather important to find out.