Jewel Setter — Heat Set Helper Arm

[embed]

This post is largely focused on info that might be helpful if you want to try your hand at building yourself. If you want some more info on my thoughts during the design, you can find them here.

If you want to create your own variants, tool holders, improvements, etc., here is the link to the Fusion file. It’s a mess, but I’d love to see what folks can cook up to improve on the concept!

For all of the printed parts referenced below, I’ve uploaded the STLs to MakerWorld, Printables, and Thingiverse, pick your preference.

My First Impressions

Before diving into the build details, some quick first impressions from my testing/playing thus far.

The only aspect of the design that I am not in love with is the tip changing solution. I knew this was going to be a gamble (hence my requirement that it remain compatible with the original), and it is functional. But I’ve found it a bit clumsy to use.

I have made some updates to the design to help with the clearances of the tongue and groove features there. I think that should make it significantly easier, but I doubt I’ll buy replacement parts to find out (I was fortunate enough to get the first set for free from PCBWay). But, I made the changes, should you decide you want to give it a go…and if you do, could you PLEASE let me know how it goes and any suggestions to made it a smoother process?

The arm was really the main focus for me, and that portion has honestly exceeded my expectations. I was mainly looking for a good use case to test out the jewel bearing four bar concept on something functional.

The counterbalance mechanism is fairly straightforward, but I ended up being really happy with it. Just something so satisfying about watching it.

I was really on the fence going in to this build on whether I wanted a counterbalance on both joints. I ended up going with the second link being free hanging, and I think that was the right choice for this one (at least in my opinion).

All in all, I’m more than a little pleased with this little fella. It’s been a whole lot of fun, and I’m looking forward to getting a long life out of it…..and continuing to find other fun stuff to strap to the end of it.

Now, if that hasn’t scared ya off, on to the build info!

Build Details

Rotary

Printed Parts:

• rot_Base
• rot_TopDeck
• rot_Preload

COTS Parts:

• (~98) 3/8” Steel Balls
• (6) M5 x 9.5mm Heat Sets
• (6) M5 x 14mm BHCS
• (2) M5 x 30mm BHCS — For attaching arm to rotary…I designed it for BHCS, but these SHCS were what I had on hand
• (2) M5 hex nut

The assembly of this fella isn’t too tricky. The six M5 heat sets go in to the Preload (inserted into the top of the orientation shown in the above image).

Fill one side of the Base with balls, place the corresponding ‘cover’ on it, keep pressure on that assembly while you rotate it. Fill that side with balls, apply that ‘cover’, then tighten the fasteners….the actual ease with which you accomplish this quick list may vary :)

Arm

This one’s definitely the most complicated of the sub-systems (and don’t tell the others, but it’s also my favorite), but it’s not too bad to assemble. Once I got the hang of it assembling the Base bits, I found the rest went pretty quick.

Printed Parts:

• BaseCenter
• Base_Preload_R
• Base_Preload_L
• BeltLock
• DriveArm
• (5) PassiveArm (same arms are used for both upper and lower sections)
• MidLink_Hub
• MidLink_PreLoad_L
• MidLink_Preload_R
• ToolPreload_A
• ToolMountCenter
• ToolPreload_B

COTS Parts

• (24) 3/8” Steel Balls
• (1) M4 x 30 BHCS
• (13) M4 x 12 BHCS
• (1) M4 nut
• (13) M4 x 4.8mm Heat Set
• (1) Spring similar to this one
• (2) M3 x 5 Heat Set
• (2) M3 x 10 BHCS — would be better with a set screw, but I used what I had
• Section of GT2 6mm Timing Belt

I assembled this in three main sections. The below was the first section, minus the counterbalance elements, I assembled those with everything else complete already.

This subassembly is made up of these fixed parts:

And these moving parts (with 2x the PassiveArm):

I started by laying one of the “Preload” parts down on the table with the cones facing up. I think just stacked it up like a sandwich. I used some of the ‘MidLink’ parts just sitting on the table to prop the ends of the arms up a bit to keep things from getting unruly.

The BaseCenter is the only part in this sub that needs heat sets. It gets the four for attaching the Preloads and one more on top for the fastener that constrains the spring.

There is enough compliance in the arms to allow this sub to be fully tightened and still assemble the “MidLink”.

Speaking of, The next step was to add the ‘MidLink’.

This section is again made up of three parts:

Plus the three remaining PassiveArms.

There are four M4 heatsets, two on either side of the MidLink_Hub for attaching the Preload parts.

Also, I found that the cable for my soldering iron is just a tad too thick to make it between the Preload and the cable hook thingy. So I have to loose that Preload a bit to insert or remove my iron cable. But it also means my cable can’t come out, so I decided to leave it that way.

Just like with the Base section, I assembled it by stacking things layer by layer. Preload, Balls, Hub, Balls, Preload. On the top, the arms are sandwiched only between the preloads. For the lower sets, the Hub sits in the middle. The below cross-section shows the Base section, but all three sections come together in the same basic way. (Also, the balls are all the same size, I just did a poor job of cross-sectioning :) )

As with when I used the MidLink parts to prop up the arms from the Base, I used the ToolMount parts to prop up the arms while I assembled the MidLink.

The Tool Mount again was made up of three parts (catching a theme? :) ).

But apparently I did decide to mix it up a bit with these. For some reason I named the Preaload A and B, but with the others I went with L and R….no clue what Past Bubs was thinkin on that one, but I also can’t be bothered to change it.

I found this one the easiest of the three, since the arms were already pretty well constrained.

And that should leave ya with a working arm!

Hot End

This is the subsystem I most assume people will need to modify and/or replace. I designed it to work with the Weller WES51 iron that I’ve been using as my dedicated heat setting iron for some time. Unfortunately, I recognize that ironing station isn’t even on the market anymore :) so guessing it won’t be most people’s go to.

But, that is why I wanted to go with the isolated subsystems to begin with. This mount can easily enough be swapped out for one of your own (or, hop into the Fusion files and just modify this one to suit your needs). Same goes with the Arm. Both ends are just tongue and groove, so it can easily be swapped into other configs if so desired.

If you DO want to design yourself a tool mount, here is the info I think you’ll need.

The two holes for the fasteners are clearance holes for an M4 (I went with 4.65mm) spaced 20mm on centers.

The nominal gap for the groove is 8mm. The groove is a bit over 19mm deep and the holes are 7.5mm back from the opening edge of the groove.

There isn’t a specific mass limit that I’ve established, but, in general, lighter is better.

But anywho, all that ranting aside, on to Hot End I went with!

Printed Parts:*

• HotEndLockPin
• IronMount_lw
• IronCover
• HotEndRetainer

COTS

• (2) 4mm x 30mm dowel pin
• (4) M4 x 20mm BHCS
• (6) M4 nut
• (2) M4 x 25mm BHCS — For attaching to Arm

*I didn’t print any of these parts myself (at least not the final versions), but they are in fact printed.

The IronMount_lw (believe it or not, this is the version after lightweighting :) ) is what is setting the orientation of the iron.

I taper matched the V-groove in the mount the taper on the handle of the iron. Because my iron handle no longer has the insulating foam on it, this taper contact works on a pretty good length of surface, which is great…but it also means the mount will get a bit toasty.

That, plus the fact that this part also will make significant contact with the printed retainer, meant I wanted to give it some extra thermal protection. So I went with the UTR Therm material from PCBWay. I’d never used it before, but so far so good. The surface finish came out quite nice, and while I didn’t check it against a print, everything fit together really well. Full disclosure, I got the PCBWay parts for free because they were sponsoring the video, but if they hadn’t, it would have cost me $14.15.

I made one modification to this part in what I uploaded vs what I tested. I created the feature pointed out below, the little shelf there.

In my testing with tip swapping, I found that having the ‘cage’ (as I’ve come to call it, the part/file is the HotEndRetainer) able to overshoot into the groove makes it much more difficult to quickly get the locking pin in. So I extended this surface and added some additional clearance to the holes on the cage. That should allow this this to be in intimate contact with the flange on the cage and still get the pin in.

There are hex holes on the back side of the mount for the four M4 fasteners that hold the IronCover in place

Speaking of the IronCover

This one saw a couple of changes based on my testing. I added stiffening ridges between the fasteners. I added these because I found the existing part was already creeping quite a bit more than I’m happy with. So I added a couple of millimeters of extra material to stiffen this up. I think for my personal one, I’m just going to print some PLA stiffeners to put behind the flex material. This is my first time trying out PCBWay’s UTR Flex material. So I want to leave it in place if I can, to see how I like the material after some use. This one would have been $86.80.

Second, I thinned the walls of the ball of the handle. I had the nominal thickness at 3mm, and it just wasn’t as compliant as I had hoped. So this change is purely for feel.

Those are the only parts needed if using it without the tip swapper. The iron gets laid into the V-groove of the mount, and the cover gets tightened down over top to secure it in place. Bada bing, bada boom.

On to the thing I am most, “meh” about with this build. I think if I can work some of the kinks out (like the alignment issue I mentioned above) it could be quite a bit easier than swapping tips with the screw on. But I immediately feel like I’d like to improve on this.

The main issue is that nothing is guiding the cage and iron mount parts into alignment. So you have to try and hold them in relative alignment long enough to get the locking pin in. With stuff at elevated temps, this just becomes more challenging than it needs to be.

But, all my griping aside, I intend to leave the quick(ish) changer on the tool for the time being.

Also griping aside, I have to talk about how blown away I was/am by the 3d printed stainless part. Not only did the section that gets down below 1mm in thickness come out great, but the overall surface finish was quite a bit better than I am used to with SLM parts (or maybe I’m just too acclimated). Although the Hot End may have won for my least satisfied subsystem, this part easily wins out for favorite part of the build. It would have run me $24.89

And the final part (or really little subsystem) is the lock pin handle. I was excited to try out PCBWay’s Somos PerFORM material. It has heat distortion temps up to 276C, and I believe is ceramic loaded…in other words, I wanted to play with this stuff!….but…I forgot to model the holes for the dowels into the part I uploaded to them. What an insignificant detail, eh?

So for now I’ve just got a PETG part doing the job here.

I didn’t want to deal with trying to press fit into that handle geometry. So I went with a running fit and added a dab of adhesive. It shouldn’t really be seeing any significant loads, so it doesn’t need much.

Assembling the Subs

Once the subsystems are together, it’s time to assemble the beast.

I first attached my Rotary to my Thorlabs optical breadboard base. I didn’t discuss the Base above, but attach your Rotary to the solid base of your choice.

I then attached the Arm to the Rotary. Just slide the tongue into the groove, and insert the M5 fasteners. There’s line of site to these fasteners, so you can visually look to see when they are aligned, if it helps (same is true for the tool mount side). One side of the mount on the Rotary has hex-shaped recesses for hex nuts. Tighten the fasteners until the joint feels secure, but don’t overtighten.

At this point, I went ahead and attached up my counterbalance.

I first put my M4 fastener and nut into the DriveArm and just tightened them enough to remain in place.

I next fed the end of the timing belt into the gap until it was nearly to the end of the timing belt pulley portion. Holding the belt in place with my left hand, I tightened the fastener with my right.

I then fed the other end of the belt through the gap between the pulley and the rigid frame parts.

It was time to add the spring (or springs, in my case). I lightly secured one end of the spring to the M4 fastener in the BaseCenter part.

I felt the belt through the BeltLock, then through the open loop in the spring.

I decided to set my spring tension at the low load end of things. So I tilted the DriveArm set up to max angle (a little over 90 degrees).

With the arm in that position, I pulled the belt through the spring until there was only a light tension on the springs (slight visible extension). I pinched the belt around the spring to maintain that position while I fed the free end of the belt through the BeltLock again (teeth-to-teeth). Then I just got the BeltLock positioned where I wanted it and locked it down.

It wouldn’t hurt to make a smaller BeltLock, but as it is, this one just barely clears the arms for me…so it ain’t broke, I ain’t fixin it :)

This setup has worked great. I’m not sure if I just lucked out and the spring combo I grabbed are perfectly suited, or if it’s just pretty insensitive to selection.

The main features I was looking for when I was rifling through my spring drawer were:

• ~40mm of extension — This equates to 100 degrees of rotation at 45mm pitch diameter
• Max OD of 14mm — Gives 0.75mm clearance on either side for nominal arm positions
• “Feels about right” — My design spring had a load of 3.46 lbf at max extension. So somethin like that

And finally, attaching the tool to the arm goes pretty much the same as attaching the arm to the rotary.

A Little Extra Fun

To see how it would do assisting with holes of the less hot variety, I decided to model up a Dremel mount :)

It’s nothing fancy, it just clamps around the body of my corded Dremel. I have a cordless that I debated using instead, but I decided I’d rather have the lower weight than the cordless convenience.

I’ve included the files for this along with the rest, should you be so inclined (links up top).

** Some of the links above are “Affiliate Links”, and I receive a small commission (usually ~1–3%) of any sales from them.