Latest Articles

Most Popular

Assorted (hopefully) "Useful Stuff" Projects

Like the bucket of assorted fasteners on that bottom shelf, this category is for stuff that I didn't know how to group...oh, and speaking of those fasteners, check out the little sortin fella!

2020 Aluminum Extrusion Hardware

Quick Bolt Sorter

General Purpose Turntable - Gen1

Battery Racks

Details
Parent Category: BubsBuilds Projects
Category: Assorted (hopefully) "Useful Stuff" Projects
 
I don't remember when I first made one of these little racks for holding my camera batteries, but with each new camera battery type I've added to my shelf/bag, I've made a new  holder for em. I have found that the simple thing of 'contacts down = charged' / 'contacts up = needs charging' has been wildly helpful for me. So as a result I've also now started making some for other rechargeables.
 
For most of the below, the CAD files can be found on OnShape here  (the exceptions are the ones I made that predate OnShape)

Common Batteries

AA

 
I've been gradually shifting over to rechargable AAs, so I now have quite a few in rotation.  So I   went with  a 22 position holder for these. I printed mine in Polymaker's translucent blue PETG. I really like the look of the filament, but the surface finish leaves a bit to be desired after my recent switch  to a 0.8 nozzle....clearly gots some tunin ta do. But it holds the batteries just fine, and looks aren't exactly top of my list of needs on this one :) 
 
 
 Based on my super thorough research (aka quick Google image search for AA battery drawings), AA batteries have a diameter between 13.50mm and 14.50mm. My calipering said mine land on the top end of that. So I went with a nominal hole diameter of 15mm for the holder. I have found that a 0.25mm radial clearance results in a snug fit with my printers.
 
I didn't actually print one of the below test cells, but if you want to dial in the fit to your liking I'd recommend printing one of these first. Then you can adjust the scaling in your slicer to get  it dialed in. The dimensions of the single match those of their full-sized buddies.
 
 BattRack_aa_sketch2_fde50.jpegbattrack_aa_model2_cdaf6.jpeg
 

AAA 

 
For the AAAs, I basically just scaled the design above for the AAs. I printed mine   in Sunlu's black PETG.
 
  
I didn't actually print one of the below test cells, but if you want to dial in the fit to your liking I'd recommend printing one of these first. Then you can adjust the scaling in your slicer to get  it dialed in. The dimensions of the single match those of their full-sized buddies...and yup, copy/paste fo sho
 
battrack_aaa_model2_2634d.jpegbattrack_aaa_sketch2_2d9d1.jpeg
 

Camera and Related Batteries

Sony  NP-FZ100

My two main cameras use these NP-FZ100 batteries (I use a Sony A7R IV for photos and recently upgraded to a Sony ZV-E1 for video...and hey, if you want to help me get out of that debt, please feel free to buy something from any of these affiliate links...come on, you know you need some filament....eh? :) ) So I recently decided to update this guy from a 4 slot to a 6 slot. I like the larger size, but I won't lie, I miss the cool look of the gyroid fill through the clear filament on my 4 slot...granted, this seems like my fault.
 

Six Slot version

Printables    |     Thingiverse
 
This is a recent upgrade to the below that's been my workhorse for a good bit. I printed the new one from Polymaker's translucent green PETG. 
 
 

Four Slot version

Printables    |     Thingiverse
 
This is one that I've been using for many years now, I love this little guy. It's been on many a backpacking trip (yeah, I'm an idiot, but I just love me some backcountry astro photography  ¯\_(ツ)_/¯) and still looks/works as great as it did the day I printed it. Not gonna lie, I have no idea which translucent PETG I printed this  in...but it's one of em.
 
DSC03095_1ca82.JPG
 
 
 
 

Single

Printables    |     Thingiverse
 
For test fitting, or for just throwing an extra battery in the bag. I don't think I'd made any singles prior to this most recent batch of holders. But I'm actually looking forward to this little dude. My A7RIV seems to be quite a bit more efficient than my A7ii  was. So I can frequently get by with just one in the camera and a backup for a day of out and about. I  printed this one with a few base layers of Sunlu's black PETG and finished out with Polymaker's translucent green PETG. 
 
 
 
BattRack_fz100_model1_5e81e.jpegBattRack_fz100_sketch4_3e07f.jpeg
 

Sony  NP-F970

Printables    |     Thingiverse
 
These big boys are for some LED light panels I have, plus I recently got myself a little video monitor that can run off of these as well (although it zaps em pretty quick, so I prefer to have it on DC when possible.)
 
I went with a Four Slot version, since that's how many of these I seem to have at present. I printed mine in Polymaker's dark green PETG, and I really dig the color on this stuff.  May be my favorite shade of green PETG that I've found to date. 
 
  but with how large these things are, I'm tempted to also make a Two Slot version that would be easier to put in a camera bag. If you'd be interested in a two slot variant, let me know in the comments (below or whereever ya like). Maybe a couple of echoed thoughts will be enough to overcome my inherent laziness :)
 
 But I did go ahead and toss together a single/tester, as with the others up yonder. I printed this one in Polymaker's translucent green PETG...also like some others up yonder.
 
battrack_f960_sketch10_62c30.jpegbattrack_f960_model1_85a0f.jpeg
 

GoPro 

 
 
For the GoPro batteries, I went with a six position holder. The GoPro batteries have a bit of a trick geometry for getting the same sort of fit I like with my other batteries. They have flanges around the ends that are wider than the bulk of the battery. So if you make them snug, these fairly thin surfaces catch on the layer lines of the print. I'm sure it could be dialed in, but to be honest I don't use my GoPros a ton, so they don't rank too high on my list to get 'just right' at present. This will definitely work well enough for now!
 
I do really like how it looks though. I put down a few layers of Sunlu's black PETG before switching to Polymaker's translucent blue. I really like the way the contrast came out.  Although the photo of it less so, so you'll just have to use your imagination :)  
 
 
 

Caster Wheel

Details
Parent Category: BubsBuilds Projects
Category: Assorted (hopefully) "Useful Stuff" Projects
 
 
Playing around with an idea for some  caster wheels.  

Update - Load Testing to failure

 Failed at a little over 600 Newtons! (~140lbf)

 

Build

I think the little video above pretty much covers the build process. The only part not shown is inserting the heat sets into the FrameMount and one of the wheels.
 

BOM

  • Printed Parts
    • FrameMount  
    •  Hub 
    • Retainer 
    • Wheel (x2) 
  • COTS
Just pop in the heat sets, toss in some balls, and tighten...sounds so easy in theory, doesn't it? :)
 
If either of the bearings feel too tight, you can add a washer in between the mating parts to free things up a bit. You could also probably just extrude the face in your slicer also. For the one shown above, I had no spacers on the main bearing and one washer between the wheels. It puts the wheels on the sloppy side, but that's kinda what I want (I'm hoping this will make it hold up to pet hair and the like a smidge longer  ¯\_(ツ)_/¯)
 

 Design

CAD files available on OnShape, here.

A caster wheel is essentially just two rotational axes,  othogonal to each other and with a radial offset (and if you read on  all the way through the Design Notes below, you'll see that I managed to overlook this last one initially...oops.)  
To create these rotational axes, I decided to use my old favorite, bearing races integrated into the printed parts and fillin em with balls.
 caster_v3_xsec10_3b0fa.jpeg  casterwheel_v3_model_8a72a.jpeg  casterwheel_v3_xsec11_26a58.jpeg
 
The light blue part in the image above is the FrameMount. It provides one of the main bearing races and is also the attachment point to the structure to be wheeled about. 
The light grey part is the Hub. It houses the inner races for both the main bearing  and wheel bearing. It's also the part that I suspect to be the one that fails under heavy load :)
The dark blue part, shown best in the image on the left, is the Retainer. It is the other outer race for the main bearing and also provides the preload for the bearing.
And the yellow and dark grey parts are the wheels. They are both the same Wheel part  and in addition to being the wheels they, can you guess?...that's right, they're also bearing races, kinda repetitive, eh? 
 
Although only one ball is shown in each race in the above, they should be fully filled to ensure proper load distribution.
 
The nominal design includes small (0.5mm) gaps between the Retainer and FrameMount, as well as between the wheels. This is to allow for developing a preload in the bearing as these surfaces are brought into contact. The thin-walled geometry of the Retainer is intended to act as a  source of compliance, instead of using something like a spring washer or the like. I was too lazy to do the same for the wheels, and it shows in operation. However, in iterating on the design, I realized I actually kind of prefer these bearings to have a very low preload. This is mainly because I don't really care at all about the runout in these bearings (or at least so I believe today....we shall see), and I do want them to spin as freely as possible.
 
So I decided to just leave the gaps and such as is, and dial in the preload I want by inserting a washer or washers in this gap.
 

Design Notes

 V1 - Epic fail

 Well, isn't this embarrassing.  In my excitement at the idea of just nesting integrated bearings to obtain  a caster wheel (but also just kind of intrigued by the idea of using this concept for gimbals, trunions, etc.) I seem to have overlooked a very basic principle of caster wheels. An image of this model is shown below. Can you see the error of my ways? :)
casterwheel_v1_design1_1ef54.jpeg
 
No? It's ok, clearly I couldn't either. Yes? Well, where were you a week ago?
 
For a caster wheel to do its castering thing, it relies on the moment generated by the offset between the wheel/floor contact point and the axis of the main bearing. That way, when being pushed any direction where this moment is produced (any component of the push direction being normal to the rolling direction of the wheel) the main bearing rotates, aligning the wheel to the direction of the push. 
casterwheel_stock1_84349.jpeg
 
In my love of symmetry, I just completely overlooked this until I assembled the thing and was dumbfounded by it's lack of  desire to rotate. 
 
Luckily, I did realize my error prior to just iterating on my oopsie. So, on to V2
 

V2

2024/01/12 - Design notes

The below image shows the main bearing concept xsec (I haven't patterned the balls yet). The light blue part on top is the rigid attachment point that will go to the frame being wheeled about. The dark blue part will be what attaches to  the wheel, and the grey piece is the retainer. The  thin walls and contour of the retainer is intended to provide some compliance. There is a nominal gap of 1mm between the retainer and the frame mount. The frame mount will have an M5  heat set, and tightening this M5  will deform the compliant retainer, preloading the bearing. The plan is to fully close this 1mm gap, but as usual, 3d printed flexural elements can be difficult to predict/calculate, so I'm just going with an 'estimate and test it' approach.
 casterwheel_design1_c1487.jpeg
 I think I have the main bearing where I want it. I'm curious how this bearing config will feel. Given that it's getting late, I think I may just print this assembly overnight  and see how I like it in the morning.
 
 

2024/01/13 - Design notes

I lied above. I was apparently more tired than I realized, and those parts didn't even make it to the slicer :) 
So  instead I went ahead and finished out the V2 design this morning.  As you can see in the image below, I may have overcompensated a smidge with my offset.  The extent of the offset was really more the result of liking the small(er) overall package size this offered. My assumption is that this connection structure between the main bearing and wheel bearing will be the failure point when overloaded, but I also wouldn't be surprised if the loads on the retainer are higher than I am expecting, letting it earn the award for weakest link...I suppose we shall see.
 
casterwheel_v2_design2_f67da.jpeg
 
 
All of the parts fit comfortably on a single build mk3 build plate, and with my print settings that yields a print time of right about 6 hours.  Total material consumed should be ~190 grams.
 
casterwheel_v2_print1_a0ead.jpeg
 
2024/01/13 - Build comments/takeaways 
  • Good  lord was this thing a pain in the ass to assemble!  The configuration of the main bearing allowed the balls to easily fall too far out of position to be self-aligned during preload...aka balls went f*&%ing everywhere :)
  • The wheel bearing, however, was substantially easier and self-aligned quite  well in repeated tests.
  • The wheel bearing could use a better-defined  preload. I inserted stacked washers in between the wheels to set the  spacing, but ended up going with basically zero preload in the bearing. Although  this makes the bearing pretty sloppy, I'm not sure I actually mind for this application. As long as it stays assembled and allows free rotation, I don't really care about runout.  May go with an intentional zero preload design for V3.
  • As usual, support structures suck :) Although not a significant problem to remove on this one, I really don't like having the decreased surface finish so close to the bearing race surface. Changing the geometry to get the outer diameter of the main bearing race is  tangent to the flat on top of the wheel bearing (to remove the ring of supports shown in the above screen shot.
  • IT CASTERS (I know this isn't actually a verb...but hey, it's the internet, I can make up whatever I want, right? :) ). Other than the assembly woes, it does indeed function as desired! No clue what a realistic load operating range it will be suited to, but at least it does it's base function. 

V3 

2024/01/14 - Design notes

 There were two main issues from V2 that I want to address in V3, which I'm hoping will be the final rev for the time being. 
  • The main issue with V2 was with the reality of assembly for the main bearing. Although it is full constrained when assembled, it was an absolute nightmare keeping the balls in their races while putting it together.  Since the wheel bearing assembly was substantially easier to work with, I'm going to replicate that bearing configuration for the main bearing as well.  The below cross section shows the revised bearing along side the V2 config.  Essentially it moves the ball plane definition to the Hub, instead of in the FrameMount and Retainer.  
  • Also visible in the below cross sections, the wheel axis was moved down to make the outer diameter of the wheel race tangent to the top surface of the main bearing race
  •  
 casterwheel_v2_xsec1_0d8a3.jpeg  casterwheel_v3_xsec1_51980.jpeg
 
 
casterwheel_v3_sliced_9d015.jpeg
 I think V3 seems good enough to call it for now on this bad boy! (at least until I actually install it on something and realize it's deficient is some yet-unknown way :) ). 
 

LilWinch

Details
Parent Category: BubsBuilds Projects
Category: Assorted (hopefully) "Useful Stuff" Projects
I decided recently that having to manually open and close the blinds on my windows is basically barbaric. So I set out to come up with an automated solution, but in the process, this little in-line winch that I made to actuate the blinds turned out to be super fun all by itself….yes, I’m still manually opening blinds while I get distracted…what can I say, I’m a barbarian ¯\(ツ)/¯

V1

Printables  |  Thingiverse 

Build

BOM

If you need/want hooks for the ends of your winch, I reused these little hooks that I previously made for cable organizing. They are the “CableHook.stl” file included in the model repo.

Build


  1. Print the printed parts….obviously

  2. Insert the M5 heatset in the Bracket

  3. Place the 9.5mm ball in the cone on the end of the worm and then put this sub assembly inside the bracket, nesting the other side of the ball in the cone of the bracket to complete the jewel bearing.

  4. Hold the sub assembly from 3 so that the hole in the worm is aligned with the motor axis opening. Look and see the orientation of the D flat on the hole in the worm, and bring in the motor with its D flat aligned to the hole.

    1. NOTE: you may want to check the fit of the motor shaft on the worm prior to assembly. This is intended to be a pretty tight fit, and anything obstructing the opening will make it difficult or impossible to assemble without damaging something.
    2. You can put the assembly with the back side of the jewel bearing against a flat surface to press the motor into position. Don’t push TOO hard on this, the motor’s thrust bearings likely aren’t particularly high quality.

  5. Attach the motor with at least 3, equally-spaced M3s. I would recommend tightening them in stages, working around the bolt pattern. If there is additional travel in the shaft-to-worm fit, this will draw it in evenly.

  6. Thread string through the small hole in the worm gear from the outside. You only need enough string on the side to be able to hold onto as you insert the bearing.

  7. Hold the string segment on the ID such that it sits in the small V-groove along the ID. With the string held in place, press in the roller skate bearing until it seats against the gear. For my builds thus far, this has been sufficient to hold the string in place, but if you want some extra peace of mind, you may want to tack it to the side of the gear with some superglue.

  8. Wrap the desired amount of string around the pulley inside the worm gear. Try to keep even tension to get a consistent winding and do make sure the wrapped string does not sit so high in the groove that it might obstruct the gear mesh. It will work with the string wrapped in either direction.

  9. Feed the free end of the string through the guide hole in the bracket. Mesh the worm gear with the worm and align the bearing hole with the heat set. Try to keep some tension on the string as you work with the assembly to avoid future tangles. Insert and tighten the M5 to secure the worm gear in place.

    1. NOTE: The worm gear should be mounted with the smaller opening side facing the heat set.

  10. Attach string to the fixed attachment point.

  11. Have fun and let me know how it came out!

Design

Design files are available here on OnShape

Essentially this design is just a worm gear drive mechanism with a groove cut into the worm gear to act as a pulley.

The Worm is supported on one side by the motor’s internal bearing(s) and on the other end by a jewel bearing formed by cones in the Worm and Bracket, with a 9.5mm bearing ball between them.

The worm gear (and pulley) has a roller bearing with a snug fit at it’s centerline, providing radial support. The expectation is that the string will provide a self-centering force axially, so no axial/thrust support was included.

Printables  |  Thingiverse 

V2

Printables  |  Thingiverse 

Gear Mesh

Tightened gear mesh 1mm by translating axis of Worm.

I went with a slight interference at nominal, since I expect an offset in the Worm Gear axis due to the clearance fit between the bearing and fastener (assuming the force from the worm will ultimately seat the bearing on the far side of this clearance).

Backplane

In testing rev1 build2, I noticed the C frame was opening up somewhat due to the side load from the string tension against the guide.

To mitigate this, I’m going to make a couple of additional small tweaks. The first is to add a back plane to fill in the C frame. So I basically just filled in that empty space with a 5mm thick plate and added some healthy fillets at both the motor and jewel bearing end.

Adding this material cost me one of the mounting holes for the motor…guess I’ll somehow have to make do with only 5.

Guide hole location

Initially I just set the guide hole at the centerline of the spool, without really giving it much thought. But I realized quickly that while this allows the spool to be loaded in either direction, it also puts a significant amount of load into the guide (and also will probably wear both the guide and the string pretty quickly).

So I decided to move the guide hole closer to tangent to the pulley….at first… Then as I started modifying the model, I realized it might be nice to just leave the existing hole and add options!…plus I am a terrible designer, and changing the previous hole placement by THAT much breaks everything :)

So I basically just tied the guide to the jewel bearing support with a loft feature and added two more guide holes at 10mm and 20mm offsets from the pulley axis.1.5 mm

Using the guide holes closer to the tangent will not only reduce the side load into the guide, it will also move the line between the two attachment points closer to the expected CG. So there should be less rotation of the assembly as the cable tensions.

Fixed Attach Point

Final modification for V2, a beefening of the attach point for the fixed cable. I didn’t have any issues with this in V1, but with the string pulling directly in shear along the layer lines, I just felt like a little bit of reassurance would be nice.

The existing wall was 5mm thick, so I just added another 5mm to that surface. The load path down into the frame isn’t ideal, but I think it’s good enough.

Printables  |  Thingiverse 

  1. Productive Susan
  2. Folding Table Leg Hinge
  3. Pulley - V1
  4. Bolt-Sorting Sieve
  5. General Purpose Turntable - Gen1
  6. 2020 Aluminum Extrusion stuff
  7. Desktop stands, holders, and the like
  8. Hooks and Hangers
  9. Assorted Frame-making Hardware

Page 1 of 4