Print in Place Mini Omni and Vector wheels (1 3/8" Diameter)

Allow me to introduce myself.

I have been doing FIRST robotics since 1995.

I am currently a mentor on Team 88, TJ-squared.

I am also an engineer at Formlabs, a 3D printing company.

I love all three. I wish that there was more overlap between these things I love.

I am trying to fix that.

I formed a group at my work called FormFRC to design parts that are (A) useful to FIRST FRC teams and (B) print super well on Formlabs Printers and/or in our materials.

This is the 1st post of what I plan to be many more (I have half dozen in the can and will post them as soon as I am able).

Print in Place Mini Omni and Vector wheels (1 3/8" Diameter)

Youtube videos (me yapping about the wheels)

Youtube video of the wheels interacting with a Note

GrabCAD with all the files (STEPs and STLs)

Space is always tight on a FIRST FRC robot. You need to move a Note or Cone or Cube or some such around you just can’t get it to center correctly.

Omni Wheels and Vector Wheels are often your friend in such cases.

But you can’t get them because they are out of stock (too often) or even if they are in stock you just can’t fit those giant things you can buy.

How about let’s print some. 1 3/8" OD rollers that you can batch process and get have as many as you’d want overnight?

It’s possible. I was able to start a batch of 192 mini wheels printing last night and the were ready to roll on a robot (well, if TJ2 had a robot ready – we are still prototyping) by mid morning.

These wheels are not very grippy because they are hard Nylon12 plastic but the surface of the notes are grippy enough to allow them to function tolerably well.

The Omni Wheel is a 3 Roller with a 1/2" hex drive, you can phase them to get effectively 6 rollers.


The Vector wheels come in 2 Basic Flavors. 6 Roller and 5 Roller.
It is hard to judge a difference between their performance. I think the both run about the same.

One difference between them is that the 5 roller version only needs a Right and a Left because you can vary the phasing between adjacent wheels in 12 deg increments (360/5 - 360/60) by just putting the wheel on the shaft at a different spot on the hex shaft.

BUT… that doesn’t work for a 6 roller version. If you rotate the wheel on the hex shaft you get the same phasing, so you can’t make the nice “chevron” shape that seems to work best. If you want that you have to have multiple versions of the Right and Left hand vector wheels. I have made 2 flavors of each hand so that there are 4 versions of the 6 roller vector wheel.

Note that we initially had the notes being driven from only one roller (pushing them along the table top) and that worked a lot less well at centering the note.






If you want to dig into the CAD the Onshape files are public:
PrintInPlaceMiniOmni
PrintInPlaceVectorRoller (5 wheel flavor is on a different branch – check the versions, you’ll find it)

I hope people find this interesting and useful.

Cheers,
Dr. Joe J.

20 Likes

Joe, you know I love the printed madness… have you tried these on FDM machines that are more commonly available to FRC teams?

4 Likes

There is a 1/2mm gap between the rollers and the housings. I really don’t know if FDM can manage that or not. I am complete spoiled by working at an SLA/SLS printer company. I would be happy if someone would try this and see what they can do.

2 Likes

Great looking wheels!

Any reason you didn’t print in PA11? I had some problems with printing similar projects/wheels in PA11 due to the thermal characteristics of melting/sintering the material but I was using SAF technology. Curious is the problem translates over to the more conventional SLS technology as well.

I believe that powder-based processes are best suited to the FRC environment (compared to material extrusion & vat photopolymerization) and I’d love to see a more economical way of introducing these parts to teams. Are you aware of any service bureaus currently operating the Fuse1?

Probably depends on tuning and orientation and possibly layer height.

When I design in custom supports on prints I typically use an offset of the layer height and tend to print in .20mm heights. It seems to handle that (filament typically droops a little so they are contacting at points but easy enough to remove)

My printer is currently on TPU duty for a few more days until another one shows up to take that over. But maybe I’ll get to giving it a shot over the weekend if someone else doesn’t by then.

I have indeed tried Nylon11 but it was Nylon11 CF (Carbon Fiber). They worked great but I think the better surface quality of on CF parts allow the rollers to spin with a little less friction (it is hard to tell by my hand but I THINK that the Nylon12 versions roll a little easier). I have not tried unfilled Nylon11 but I expect that it would be quite similar to Nylon12.

As to SLS for FIRST teams, it is a game changer. We basically never buy pulleys any more, preferring to print whatever we need, almost certainly printing it attached to whatever it is driving rather than having the torque transfer to a shaft and then back to whatever the next stage.

As I have said before Formlabs employees have “infinite prints” as a perk of employment. I almost never print in anything other than SLS since the Fuse1 has been available.

Cheers,
Dr. Joe J.

5 Likes

Couple of years back someone took the Thriftybot’s Vectored Wheels and made them into print-in-one piece models and we’ve printed those out on common FDM printers like the CR-6 SE with great success with them (using Duramic PLA+ filament). It was posted on here…just need to find that and post it up again.

1 Like

We just need more Formlabs employees in more FRC teams…

Step 1: Own a >$10,000 printer. :joy:

In all seriousness though, these are pretty neat, might have to try one out on one of our FDMs.

1 Like

I’ve used these a bunch.

But those are a smidge over 2"

2 Likes

Here’s the post that has the print-in-place wheels, and you’re right, they are 2" and not 1-3/8".

There is a version 2 that is easier to print with a support wedge and some extra clearance between the hub and roller

1 Like

Looks like I can order these for $3/wheel (when ordering 50) or $5/wheel (when ordering 30) at a 3D printing service bureau with a 3-day turnaround in PA12. While a team can print them themselves for a lower cost, $3/wheel is pretty cheap, especially if you’re looking to fill an intake.

1 Like

We have used the previous print-in-place wheel posted by 2910 at 0.75x scale. This at least gets you a 1.5” wheel that worked on our Markforged at any rate.

1 Like

yeah think that’s the one we ended up using. the wedges make it very easier to remove, just 1 cut and everything just works.

Update: tried it with FDM, with supports it came out but the rollers ain’t that smooth. Probably some tweaks could make it work though.

Hi Joe, thank you for this information our team has the wonderful opportunity of having access to 2 form lab resin printers. However we have yet to use them as we have don’t know exactly what to use them for, any chance this forum could be increased to including resin printers and some special item that can be made with these?

Thanks in advance,

We made pulleys, rollers, bevel gears and spacers on our resin printers for FTC and FRC this year. The surface finish and accuracy is significantly better compared to FDM and as long as it is used for a low-torque application we haven’t had any issues. Our alphabot is using 3D printed 24t pulleys for driving our flywheels 1:1 off of 2 Neo Vortex’s without issue. While we’re using ST45 from BASF, I’d imagine Tough 2000 from Formlabs would work just as well.

For FTC we use resin parts as competition parts, for FRC we tend to keep them to our prototypes and alphabots as we’re not as confident they’ll survive a hard hit, compared to the same components out of metal. We didn’t have much of a chance to do testing on our resin materials before the season started.

2 Likes

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.