Adapting Extrusion Based Cantilever for Thermal Expansion.

Hello,

I've been thinking about moving up through the materials in terms of temperature capability from PLA to PETG and onto Nylon with PC being an ultimate goal. I've got another thread in the printing section of this forum trying to attain ball park figures for processing temperatures and this thread is to collect a few ideas on how I may adapt the Ormerod design to increase its ability to cope well with temperature.

All of my acyrlic parts have already been eliminated and my main electronics will be outside the chamber.

My thoughts so far:

• Targeting max chamber temp of 60C.
• Choosing components to the capable of 70C
• 'Protiens' where needed to be PETG. Carbon filled may be interesting.
• Check temperature ratings of standard bearings and the currently preferred dry polymer bearings.
• Constrain smooth rods at one end only. Bronze bushings for other end?
• Considering use of Open Builds open rail instead of smooth rods.
• Swapping to GT2 belts.
• Seperate, but enclosed filament spool enclosure. Possible 80C pre-heat.
• Moving to an extrusion based x-arm rather than the CNC cut plastic or metal versions.

I was thinking that with ensuring the rods are only constrained at one end the cantilever design could be quite geometrically stable with moderate temperature changes?

Anything else I may need to consider?

I will be considering safeties very carefully but as a rule don't run the machines unattended for a great length of time. The enclosure insulation will be PIR in order to minimise fire risk.

Belts: Out of curiosity, what belts are you using right now? I was under the impression that GT2 belts were pretty universal. Definitely choose good GT2 belts though like these nylon coated Gates belts.

Proteins: You should definitely look up the glass transition temperature of every polymer you'll have inside the chamber, PETG's is around 80C so you're cutting it a bit close, ABS is at 105C which is better but ABS is notoriously bad with warping which isn't ideal for proteins that need tight tolerances. I'm also not sure how well belts perform under temperatures like that, though it's not like there are any good alternatives to rubber belts. You should experiment with annealed PLA which has amazing strength and heat resistance, proper scientific testing hasn't really been done to get exact figures but from what I've seen it can outperform ABS and carbon fibre infused filaments in strength and heat resistance all with the cheap and easy-to-print qualities of PLA. The only drawback is the annealing process itself warps the print about though I don't think it warps as much as ABS

V-slot/Openbuilds: I love v-slot and hope to rebuild the motion systems on my printer with it but one thing you have to keep in mind for your purposes is that v-slot requires plastic wheels which usually have bearings. Definitely do some research into the heat resistance of whatever polymer the wheels are made of.

Hotend: What hotend are you using because that's pretty important for printing these materials.

I don't really know much about heated chambers so just for my own knowledge, how are you heating the chamber, just with the heated bed? What is the chamber itself built from? Finally, how will part cooling work at such high temperatures?

Edited 2 time(s). Last edit at 11/30/2018 02:02AM by NinthDimension.

Thank for your reply. Loads there!

I thought the same regarding belts and ende up with parts about 1.5% undersize. The RepRapPro kits ran on MXL belts. These are imperial and by eye unless you count 50 or 100 teeth look very similar to GT2. The pitch is about 2.03mm!

With regards the chamber temp it'll probably hit 40C with passive heating alone. Decent insulation and it won't take much heating to get it to 80 easily, but that'll mean quite a major shift in size from ambient and really the stepers should be outside as there won't be much difference between their temp and ambient for cooling. I probably should consider ABS again, but that would mean I'd need to mock up an enclosure and extractor early. I seem particually sensitive to those fumes.

Have a look at this for the v-slot: [ooznest.co.uk]
Metal wheels can be used with it and being removable it can be repaced without dismantling the frame. Ok it is giving one more thing to align, but it is little different to smooth rod + extrusions on many reprap with error in alignment to the base extrusiin being more tightly controlled. That and it is aluminium alloy and so thermal expansion shouldn't be far enough out to need to leave one end unconstrained in other to avoid slight bending of the frame due to differential expansions.

Hot end: E3D V6 (not lite). I woukd like to move to PT100 sensors in order to make it easier to compare settings with similar setups as I believe the PT100s to be more accurate. Equally I think they will need better shielding.

Mentioned how I will heat already but if it needs a little more help I've a spare heated build surface that I plan to put a few heat sinks on and a fan or two to blow air over it. Don't think I'll meed much capacity. With regards the hot end I believe you just meed to cool the cold end of the filament to the point where the filament has good mechanincal strength, and so I'm thinking to below the heat deflectiom temp. I'll be hoping 40C ish will do the teick but that may not be enough for the final goal of PC. Water cooling is the only rel option past that. I'd need to take a little care to avoid dew poimts, but also want every thing as dry as possible.

Quote

Constrain smooth rods at one end only. Bronze bushings for other end?

Teflon tube is prettty good as a smooth rod guide. Tie-rap it for tighter contact if needed. Usually I just cut it diagonally and press it into the bracket = sort off self adjusting.

Quote

Water cooling is the only rel option past that.

Go for water cooled steppers, too. Much easier to implement than rebuilding the whole frame to put them outside the chamber. Especially the extruder motor should be direct drive IMHO.
IDK if Bowden tube guided filament likes to linger in 80°C environment for long...Optionally go for remote cable drive extruders.

Edited 1 time(s). Last edit at 12/01/2018 01:48AM by o_lampe.

I have a Intamsys Funmat at work that I have run at 90c (max) chamber temp and it suprisingly does very well. Although I haven't torn into it yet the kinemetics appear to be the same as the Ultimaker. All of the belts, steppers, direct drive extrusion system and bits are inside the chamber and get very hot but still work as intended. Unfortunately I don't know what materials they are using but I will see if I can find some markings on the belts.

That is toasty! Which material needs that heat - or is it more with the heat the part is better rather than 'need'? Heard that PC wasn't really happy until the bed temperature exceeded 135C.

Quote
WesBrooks
Metal wheels can be used with it and being removable it can be repaced without dismantling the frame. Ok it is giving one more thing to align, but it is little different to smooth rod + extrusions on many reprap with error in alignment to the base extrusiin being more tightly controlled. That and it is aluminium alloy and so thermal expansion shouldn't be far enough out to need to leave one end unconstrained in other to avoid slight bending of the frame due to differential expansions.

Oh, I see that I misread your initial comment, you were referring to linear rails which is different from v-slot. You were talking about openrail linear rails, I was talking about v-slot linear rails, usually people just refer to openrails linear rails as "linear rails" and then v-slot linear rails as "v-slot." But yeah, openrails linear rails are probably your best bet as they are virtually all metal and are super rigid, miles better than smooth rods and, in your case, better than v-slot. You should definitely look into annealed PLA, I think that it could prove to be better than ABS and minus the fumes. Here is Tom's video on annealing, he only tests the mechanical strength, if you wanna see a heat resistance test I think CNC Kitchen has one here. I think o_lampe makes some good points as well, he probably knows more about heated chambers than me. You're right about the water cooling, as far as I know, if you want to go much higher than 50C you'll probably need some sort of cooling system that cools the hotend from an external coolant, the only two options I know of are forced air and water cooling.

On a side note, you really are investing a lot into a pretty humble printer, I imagine you've exceeded the price of the original kit many times over with the upgrades you've done and are planning to do but I suppose that that's what makes repraps so fun, just endlessly upgrading. Also, I had never heard of RepRapPro before since I've only been in the community for about a year but reading up about them it makes me really sad to see that they're gone, seems to me they were one of the only companies we ever had that made original reprap designs with the goal of minimizing vitamins and it was even run by the founder of the reprap project himself.

Edited 1 time(s). Last edit at 12/01/2018 03:25AM by NinthDimension.

Quote
WesBrooks
That is toasty! Which material needs that heat - or is it more with the heat the part is better rather than 'need'? Heard that PC wasn't really happy until the bed temperature exceeded 135C.

I haven't done PC on it yet but some 9085 Ultem and PEKK at 90 chamber and 150 bed. It does those materials ok but not as good as the Stratasys machines we have. The higher temp materials ideally need a chamber temp just below the Tg of the material and that's what the Stratasys's do. The bed goes to 160 on the Intamsys. My personal printer does that with a glass bed and 110v ,500w silicone heater. I have some PC I might get a chance to try next week and will find out what the temps are for that material on the Stratasys.

Edit- I just remembered I had recorded the temps for various Stratasys materials for their machines and their PC was one of them which prints at 345 with a chamber temp of 140. There is no heated bed. The steel bed stabilizes at the chamber temp

Edited 1 time(s). Last edit at 12/01/2018 06:57AM by Bill Clark.

@NinthDimension Sorry my bad. I was referring to the product family as v-slot rather than using the v-slot as a rail. The original company has gone, but RepRapLtd exist and they ship the Fisher Delta printer. They do keep en eye on the forums and still offer parts for the printer kits that were discontinued with RepRapPro. There's a piece on the closure of that company in the Ormerod section of this site. I'm not trying to follow the reprap ideals exactly but it did occur to me that the cantilever frame may work well at different temperatures due to a simple frame where expansion of one leg only (pre-heat soak) isn't likely to twist the frame.

I've made a minor itteration of the original version and I am now looking at a second to make use of some open builds extrusion for the x-arm and realised now is the ideal time to think about the implications of running warm so that There's less need to do a third major itteration of my design later.

Cantilevered printers are always going to suffer cantilever limitations. Unless you build them with huge crosssection frame members or run them slow, they wobble. It isn't hard to build a corexy machine to handle high temperature operation. You had the right idea about securing guide rails at one end and letting the other slide in a bushing. You can do that in the X axis of a coreXY using a linear guide with two bearing blocks. The X axis rail can be fixed to one of the Y axis bearing blocks and one of the X axis blocks can be fixed to the other Y axis block. That allows the Y axis rails to move apart without causing the mechanism to bind when the printer's frame expands as it warms up.



I print ABS with the enclosure heated to 50C all the time and it works very reliably. The mechanism never binds. I have no doubt it could handle higher temperatures, though it would be a good idea to check on the specs for the linear guides and their lubricants for higher temperature operation.

---

Ultra MegaMax Dominator 3D printer: [drmrehorst.blogspot.com]

I've built an open rail based core xy before thanks. I'm familiar with them but saving the core xy project for later.