My own printer design, and it´s huge! Need feedback

This is the place to show off your felix printer hacks and improvements.
Post Reply
Legion 2.3
Posts: 84
Joined: Sun Apr 20, 2014 6:23 pm

My own printer design, and it´s huge! Need feedback

Post by Legion 2.3 » Sun Aug 03, 2014 1:59 pm

Hello folks

I want to build my own, self designed 3d-printer, based on Felix-parts (codename: MOAP)

First of all, I´m (not yet) and engineer, I´m "just" a mechatronic, so I have some knowledge about mechanics and elektronics. I´m sure this design isn´t perfect, and it will never be because I can only use printed parts (nothing machined) and parts that can be bought at shops.
I found out quite a lot things about 3d-printing in the past by reading many articles and forum posts and by building and using my Felix 3.0, but I also know that there´s much I still don´t know or fully understand, so I still have some questions and I hope you can give me some answers.

My goals were the following:
- It should be buildable without milling and drilling ( I neither have access to cnc-mills nor have a workshop in the basement)
- Larger build space (approx. 400x400x400, maybe taller)
- ball screw drives to eliminate ringing and to increase the possible acceleration, jerk and printing speed
- Bowden extruder for a lightweight carrier

Here is a render of the current, early development stage:
Render 1.JPG
Overview
As you can see, I´ve chosen an X-Y-gantry-design. Due to the higher weight of the bed, a moving print bed like the Felix has wouldn´t work.

I wanted to use Item aluminum profiles. Unfortunately, Item only sells to companies, so I have to buy them from another supplier (SMT Montagetechnik)

I want to use a rod bearing system. These are 16mm rods with Hiwin linear ball bearings. Every bearing and every driving nut will be clamped, so it´s easy to assemble.
I want to use 2 bearings per rod, I hope to get an extremely stiff bearing system this way.

The Y-Axis should be driven by 2 stepper motors. That leads me to

Question 1: The felix board offers only 1 driver per axis with a maximum current of 2A, so this driver has to drive 2 steppers. Is it better to connect the to motors parallel or in series?
A parallel connection would split the current (1A / motor), so they won´t have their full torque.
Series connection would supply booth motors with the full current, but the driver has to deliver the double voltage. I´m not sure if it´s capable of that.
The Nema 17 from Felix is rated for a voltage ov 3,4V, but there are also steppers running at >10 V out there.
Could it happen that only one of the two motors looses a step or do both of them loose steps in this case?

The X - and Y-axes should be driven by ball screw drives. I found some with a diameter of 16 mm at 20 mm pitch. I hope I can achieve high acceleration rates and a high speed due to the stiffness of the ball screws (better than these flexible belts). This leads me to

Question 2: Whats´s the maximum speed (RPM) that these steppers can achieve? I can´t find any data, even not on the Wantai Motor´s homepage. I think the maximum speed depends on the mainboard electronics, especially how many signals / second can be transmitted to the stepper driver.
200 mm/s would be great, that would require at least 600 RPM.
Render 2.JPG
Coupling
I couldn´t find any couplings from 5 mm - 16 mm (stepper - ball screw), so I want to print the couplings myself. Since the torque of the motors isn´t really high, I think they should do the job.
render 6.JPG
E3d v6 Hotend
For the extruder, I want to choose a 3 mm bowden setup with the new E3d v6 hotend (single).
Later, I want to change to dual D3D autolift hotends as soon as they´re available.
The bowden setup makes the carriage quite lightweight, but I´m not sure if it will work properly.
Due to the size of the printer, the bowden tube has to be approx. 800 mm long or even longer.
I want to use a gearless extruder (the geared ones look crappy in my opinion).

Question 3: Will a bowden setup work for this size? Is our Nema17 strong enough to push 3 mm filament through a 800 mm PTFE tube without using a gear?
Or do you think a direct drive extruder would be better? I´m wondering if my construction will be stiff enough to support direct drive extruders at high acceleration rates without ringing.
render 3.JPG
Z-axis
render 5.JPG
Z-axis explosion
For the Z-axis- I want to use a Felix-like setup (I really like this design). Though it´s a bit bulky, you only need a single motor, so you don´t get a problem with syncing multiple steppers.
I want to use a trapezium spindle tr16x4. I know the pitch is higher, but I still have a resolution of 0,00125 mm/step and the stepper should have enough power to move the bed (it´s not much more than 5 kg).
I´ve chosen a Nylatron trapezium nut, Nylatron seems to be a good choice for this kind of application.
render 4.JPG
heated bed
For the heated bed, I want to use a 3 mm float glass bed. I found a manufacturer who cuts them at custom dimensions and drills holes at custom positions. I think a 4-point-leveling should be sufficient, 3 mm glass should be stiff enough not to bend at these dimensions.
I want to use 4 independent heated beds. I want to make them switchable individually (simple relays should work, as far as I know , you can setup Repetier so that the bed is only switched on and off every few seconds.), this way, the power consumption is far smaller when I need only a small print area.

Question 4: How will glass behave if you heat it only at a small area, but keep the rest cool? Is it possible that the glass warps or even breaks? I usually print PLA, so the bed gets only 50 - 55 °C, the difference to the surrounding glass is only 20 - 30 °C.

On top of the glass bed, I want to use a removable 400 x 460 mm PP plate with kapton tape :https://shop.germanreprap.com/de/Druckp ... 60x400x2mm
I don´t want to break the glass bed when I try to remove a larger print (that can be quite difficult at larger prints)

I want to use a 850 W ATX power supply, this will provide enough power for the 4 heated beds.

Question 5: I want to use standard PLA for the printed parts (Innofil3d Filament). Are there other materials that are stiffer than pla? I don´t mean stronger and more flexible, I mean really stiff? What about PET+ or Taulmann T-glase? Or do you think PLA is stiff enough for this application?

This design is not finished yet, there are no endstops and no fans yet and there´s still something else missing.

Thank you for reading, I hope that I get some feedback from you, what do you like, what would you make different, and so on.

Saamec
Posts: 157
Joined: Fri Jun 21, 2013 9:28 am

Post by Saamec » Mon Aug 04, 2014 6:42 pm

I was surprise to see this design, I have like half a year to design something similar. Striking that the printer settings are similar. This table 400 400, and it guides 16 mm. very similar design. The exception is that my items are almost all steel and aluminium. The heater for a table, I will do according to their calculations, as the table 400 mm to warm difficult. And electronics will be completely different, as is the firmware. Going to compete with you))) 're Going to begin at the start of sales from October - November and only in Russia, this is a response to the sanctions))) good luck in the design, you have a very good beginning, almost perfect. I only wish that half a year of my work will be very much like your project, then you will have in the near future to give it up, and start another. But in any case, the customs shall eat the competitive advantages of your product. But I'm really surprised that so much in common... the photo is made as if from my editor -))) We are well done. Photo my project to lay it will not, because he is already on the finishing stage. All items defined, suppliers, and technologies for manufacturing of the details that not buy. Calculated electronics is ready firmware. Soon will be available software for your printer. You know... 6 months of hard work and sleepless nights.)) Once again wish you success)

seaton
Posts: 291
Joined: Tue May 14, 2013 9:04 am
Location: Bunbury, Western Australia
Contact:

Post by seaton » Tue Aug 05, 2014 6:40 am

Great design!

I as for motor couplings, I'm sure ball screws can come machined with 8mm shaft for coupling, probably even 10mm, and you can get 5-8/10mm couplings

My other comment is to do with the rod type linear bearings you're using, could turn out to be rather noisy if used on the X and Y axis, the ones used on the current Felix are great interms of low noise level.

would love to see finished design.

Stephen...
Stephen...

Felix 2.0 -> Felix 3.0 dual
Simplify3D Slicer, Kisslicer
Have you added your Felix to the Map? http://goo.gl/maps/HajnZ

http://blog.strobotics.com.au

Legion 2.3
Posts: 84
Joined: Sun Apr 20, 2014 6:23 pm

Post by Legion 2.3 » Thu Aug 07, 2014 3:59 pm

It´s going on:

Here is my extruder design:
Extruder 1.JPG
Extruder 1
Extruder 2.JPG
Extruder 2
Extruder 3.JPG
Extruder 3
Extruder cut.JPG
Extruder cut
I´ve decided to use a direct drive extruder, it seemed to risky for me to use a bowden setup, I think the length of the bowden tube would be a problem.
I want to use the MK8 drive gear from Reprap.me, it should be the best dirve gear out there: http://reprap.me/extruder/mk8.html

The tensioner will be spring loaded, i hope the design works as expected. The distance between the drive gear and the plastic filament guide should be small enough to prevent the filament from kinking.

The layer fan duct (50 mm fans to reduce curling) is quite large, but that´s the only place on the X-carrier where I could mount it without reducing the build space.
Lüfter 2.JPG
Layer fan 2
Lüfter.JPG
Layer fan 1
seaton wrote:I as for motor couplings, I'm sure ball screws can come machined with 8mm shaft for coupling, probably even 10mm, and you can get 5-8/10mm couplings
Yes, they can be ordered machined, but they want 20€ for a 10 mm shaft, and that´s a bit too expensive. I think (hope) my solution will do it, too.

Makron
Posts: 110
Joined: Wed Mar 20, 2013 1:29 pm

Post by Makron » Wed Sep 03, 2014 10:43 am

How is this going.

Where did you find the ball screws ?


This looks amazing
https://www.youtube.com/watch?v=9pZyN5D0a4Y

dognotdog
Posts: 181
Joined: Tue Sep 04, 2012 9:22 pm

Post by dognotdog » Wed Sep 03, 2014 6:38 pm

Some of the details are really nice, but using non-supported guides like that for X/Y/Z is a step backwards from the solid design of the felix. The "ringing" you speak of due to the belts, you trade with vibrations perpendicular to the unsupported rods. Apart from that, if you were to use high quality linear bearings instead of the chinese ones, it won't be cheap.

Makron
Posts: 110
Joined: Wed Mar 20, 2013 1:29 pm

Post by Makron » Thu Sep 04, 2014 9:59 am

i think The "ringing" is only in the Xaxis and its because it runs over a bearing so you have a flat side as the belt tooth run over it.

Makron
Posts: 110
Joined: Wed Mar 20, 2013 1:29 pm

Post by Makron » Thu Sep 04, 2014 10:02 am

I am making this atm .

Al parts are made from alu and the belt is 15mm wide. bed is 500x500.
Attachments
groot.jpg

Legion 2.3
Posts: 84
Joined: Sun Apr 20, 2014 6:23 pm

Post by Legion 2.3 » Sat Sep 20, 2014 2:19 pm

I´ve redesigned large parts of the printer, now I use 2 Gates 9mm GT3 3M-Belts per axis. The ball screws need more parts to run than expected (+housing for lubricating and play-free adjusting, fixed bearing, floating bearing), that´s something I can´t do sufficient with printed parts.
Now, I´m printing the parts with ABS-plastic, I need the temperature stability.

The belts are stretch-free (that´s what Gates says), so i won´t see any ringing there. The Felix belts are quite soft (i see it already when I lock the motors and try to move the X-Carrier back and forth, and that´s defenitely not only coming from the bearing

The entire drive system is now more like the Ultimaker-style.

I´ve reduced the build plate to 340x320 mm, the power consumption of a 400x400 mm heat bed is too much for me.

For the elektronics, I´ll use a RADDS board with an Arduino DUE and 1/128 stepper drivers with 2,2A and 36V maximum voltage.

I can´t see why unsupported Linear rods should be worse than rails, I use higher quality linear hiwin ball bearings, 12mm for XY and 16 mm for Z (i know it´s overdimensioned). There is a very low radial load at the bearings and the 12mm rods won´t bend much at this small load.
Attachments
MOAP oben.JPG
MOAP Gesamt 2.JPG
MOAP gesamt 1.JPG

Makron
Posts: 110
Joined: Wed Mar 20, 2013 1:29 pm

Post by Makron » Mon Sep 22, 2014 3:25 pm

I have my parts delivered :D
Attachments
i (1).JPG

Post Reply