Have not posted in quite a while as I was quite busy starting a new FRC community team and we are getting ready for an off season in november
Well I got myself a SOVOL SV04 printer (IDEX) and as a test we decided to print the take in rollers for the off season competition. It will be driven by 1/2 in square aluminum shafts connected to belt/gears and motor and the hubs are HIPS or PETG and the “tire” is TPU
This is during the print to give a view on the inside. The “squish”/ give of the "tire is controlled with the infill and the hub is 100% infill and square so no chance for the tire to slip and there are extra “fingers” as a profile
Just for completeness sake that this part of the mechanism that will be strung together with 1/2in square al. tubing
If someone asks why? now? the reason is as I said new team and offseason competition with robot from schratch to hopefully train the kids - none of whom have ever seen an FRC robot well enough to do all right in the competition which is starting to look more and more likely as it looks like we are getting the $$ together by november to sign up. For now we got a temp number I think something like 20240423 or something like that. So the ball drive has been put on the back burner amongst some other things as I am also very much involved with FTC atm. Being retired is very busy lol.
I was just thinking it had been awhile since you’d posted one of your projects. Good to see you’ve been keeping busy.
I like the idea of printed intake rollers. This tracks with some of the stuff we’ve been doing with printed TPU tire treads. I think there’s a lot of potential to optimize rollers for the game piece.
My multi-filament machine can’t switch between TPU and other materials. Can you tell us more about your multi-filament setup on the SV04?
SV04 is an IDEX (Individual dual extruder) machine and I bought it new for 429 incl shipping and tax. I use prusaslicer with it. So the left extruder has one filament in my case PETG or HIPS and the right one has the other - in this case TPU. You do the cad. I use either inventor or F360 and you create 2 bodies 1 for the hub and one for the tire and make sure they match - you can generate the geometries from the other body and then expand on it. Then you load both stls (inventor) into prusa slicer or in case of F360 I prefer to export as a 3mf. And then you assign the extruders and set the slicing parameters
So for example in the above case with HIPS it was for the left extruder 250C, 6 perimeters 100% infill and for The right one (TPU) 240 C 3 perimeters 5% 3d hex infill (10% initially with 4 perimeters which made it imo a bit to hard)
NOW the SV04 has a couple of short comings which I will fix
1.) I am not a big fan of the PEI spring steel build plate and will replace it with a glas/ceramic based or a G10 which I prefer.
2.) The extruder is cheap and not as well constrained as I would like so I most likely will replace that probably with an orbiter as I had good results with that.
3.) I just invested $12 for some bimetal heat breaks which I will replace at the next lul or when I have to do good gears. The reason is that that will let me print higher temps. As for gears I have the idea I want to try to make what I would call “cheap nylonish gears” If you print a gear out of nylon you in most cases do not need lubricants as Nylon to Nylon is quite slippery and tough. Drawback is Nylin is expensive about $80 + and up per KG. So the idea is to print the outside mm or so as a “gear tire” in Nylon and the rest of the gear - which is more than 90% depending on gear size out of PETG or HIPS etc. This can cut the total cost of the gear by 80-90% depending on gear size. It can also fix some “nylon problems” namely that even a hard nylon has often too much give which can cause a gear skip at high torque. So one could now reinforce the Nylon tire which is in place for the non lube surface and wear resistance with a hard plastic - probably HIPS or ABS or PLA etc to stiffen the teeth. Now most nylons will need to print at 260-290 which is above the stock capabilities and hence the upgrade to the heat break and a little edit to the firmware.
Over all quality wise on par with most good bed slingers of the Anycubic and Creality variety and out of the box quite useful. And easy to assemble (less than an hour) if you can read and use an Allen wrench.
For the Nylon thing I probably will also fashion some kind of enclosure to up the ambient temp to reduce/eliminate potential warping
If you are starting from scratch you probably should also get a filament dryer (about $50 at Sovol) and some desiccant packs like they are in the packages that come with the filament. As any filament except PLA needs to be dried usually.
I’m more curious how do you get the HIPS and TPU to bound together? Looking at the photo (only the 1st one shows the HIPS center and TPU black outside) it doesn’t tell if there are any bounded interface layers (I’m interested to see how strong this bound would be if it exists). Or if it’s internally locked by some sort of mechanism or special shaped parts.
The bond between the material is rather weak. Hence the hub is square and that makes a good mechanical interface on each end it steps up 5mm so even if there is a lot of lateral force it cant slip off. 3 perimeters of tpu (.44 + .63 + .63) perimeter width gives 2mm of TPU which at this TPU (Zyltech.com) wont stretch enough to skip. Even though tpu sticks ok to quite a few materials I would not trust that alone. In this case I know it will work as I did prior 2 part hubs like on this
A prior year shooter that was printed in pieces and then the hub halves held together with fasteners on the square axle. Now the IDEX option offers a way to make smaller hub/tire assy as you do not leave room for fasteners. On the 2 piece hubs I had to allow .4mm clearance between the hub and the TPU to assemble it without the use of a hammer. In the IDEX version I have 0mm clearance so even tighter. At the 2 piece hub with the .4 mm clearance the “grip” was quite sufficient so with the 0mm clearance it should be even better no matter how much it adheres or not.
Nylon wont stick to anything except Nylon so the mechanical interface will do all the lifting. Petg for example sticks very poorly to HIPS or ABS so it makes a great material as support or vice versa as it comes off nicely.
Is it perhaps possible to sandwich layers of TPU with whatever other material is being used?
Kinda fuse the two materials together mechanically?
Not sure if that made sense but if it’s possible to modify the printer g-code or Model such that the join between the materials is spread over such a large area relative to the part that it’s almost a uniform transition?
The two parts would need to be printed at the same time with a multi material printer.
They are printed “kinda” at the same time as you can see in the picture that is taken while printing. In normal printing you got 1 print head and that one gets the G-code commands - generated by the slicer - 1 layer at a time. That is how FDM printing works. With an IDEX printer in dual mode you have extruder 1 lay down layer 1 for filament 1 then it goes and “parks” and extruder 2 lays down layer 1 for filament 2 - then it parks and extruder 1 lays down its next layer and so on.
When you design stuff and slice you can design things like what each extruder prints. You can have extruder 2 for support material and extruder 1 for the part in which case you want the 2 materials close but not stick too much. You can also have 2 parts that are designed to intermesh somehow and assign extruder 1 to part 1 and extruder 2 to part 2 etc. in the hub/tire example that is the case
Here is a short piece of the Hub ( I started the print and then stopped it and tore it apart) you can still see a little TPU stuck on it
This is about 10mm up so past the 5mm step you see at the cleaned up piece. The step is on both the bottom and top to keep the tire from sliding off. So its sandwiched between the endes of the petg or hips hub. And the square shape of the hub makes it not slip when rotating. You of course can make it round and put other indentions in to make it not slip like a profile on the inside
This is an example of a 2 piece hub with grooves to grip the tpu. that was printed on an Ender 3. So I first printed the 2 halves and then the tpu part and then slid and bolted them all together. Compared to the new IDEX method the drawback is that I need room for fasteners in the standard mode (bolts, nuts etc) and in the IDEX I do not. So that means I can make skinnier hubs and tires. The rollers in this example with the 1/2in square shaft at the OD clock in at 2 in diameter. The smallest printing it in pieces before was about 3 in. If I use lets say a Neo or CIM with an 8mm keyed shaft and put a nylon outside on it I can probably go as small as 15mm diameter maybe smaller.
And there are probably many other options yet I have not thought off. But like with any thing it is not just buying the “new toy” it is now learning to properly use it and do things with it that are of benefit (in this case for FRC robotics)
Cura does this (kinda)
interlaces TPU with pla or whatever, really great for IDEX printers. @mpirringer you might want to try slicing with cura to be able to do this.
Yes. I just prefer Prusaslicer/Superslicer for a host of other reasons mainly Volumetric flow control etc. IOW things that become useful if you print with bigger or high flow nozzles. The SV04 ships with Cura. I just opted for prusaslicer. Cura is easier to use for a Novice. Prusaslicer has more features etc. So that was part of my decision too as I like to dive under the “hood”
