Best 3d printer to make from scratch

Quote
Polygonhell
I don't often comment on these threads, but my 2c, on Cartesian designs, I've built hbot, core xy, ultimaker and more conventional gantries, and I'd pick the ultimaker design for a new build.
Both Hbot and Core XY designs require a high degree of rigidity in the frame and high quality bearings to perform well IME, conversely it's hard to build a working ultimaker gantry that doesn't have close to perfect XY alignment, pretty much everything in the design forces it. I also prefer the shorter belt runs, IME in designs with long belts, you get an effect where there is a measurable difference in the way the belt responds close to the driving force of the motor and further away from it.
The ultimaker and zortrax have reputations for high quality prints for a reason.
The raise N2 version of the design mounts the extruder on the gantry and they seem to get away with it, so it doesn't require a Bowden style extruder.
I'd seriously consider a eustathios or similar design, if the plastic parts are off putting, most of them could be machined or easily modified to be.

The reputation of the Ultimaker has indeed it's reasons.
That's why I choose to design a Ultimaker(isch) printer. From what we learned the past 3 years, building many different printers, we added our own ideas to improve printquality. One of the major reasons why the quality of my Ulti-Printer is outstanding is because I abandoned the linaer LMUU bearings completely. I designed my own bearings. These bearings are adjustable, so there's absolutely no play at all on all axis. Needles to say that makes it very difficult to tune the axis because they have to be perfectly parallel. A few posts back in this thread I posted that bearing system. To be clear : the quality of the parts in these pictures is not good at all because I had to print my new parts on the printer I used for so many years, a Prusa i2.
A second and also important reason is not to design slim-line parts...! True... slim-line parts looks very nice indeed, but it's absolute nonsense. Constructing CNC-machined aluminum parts it will work making parts somewhat thinner, but not possessing a CNC workshop I'm forced to design and print my parts somewhat thicker to have these good results.

Would you like to share info about your diy bearings?
Or can you suggest the proper arrangement for the same ,which is based mostly on off the shelf parts.
I was about to order linear guides and metal plate for milling to make a G&C all metal type printer.
But if you can provide some sample prints and comparision (not random prints but sustained printing comparision of um gantry and corexy or delta) it would be not only helpful for me but other newcomers to the hobby as well to Jude the results on their merits or requirements.

(P.S: can also you elaborate over the issue of using smooth rods and profile guides as the um gantry I guess ,cannot be modded to use the guides and bushings are said to yield better results.)

------------------------------------------------------------

Pains of not having a 3d printing text book

Edited 1 time(s). Last edit at 06/17/2017 07:11AM by sudarshan.

Of course I will share, what exactly do you want to know..?
I can tell you this already : like I told you no LMUU's but 12 ball bearings taking over the task of the linear bearings. In the XY-carriage eight ball bearings are fixed , the rest is adjustable. The adjustable bearings can only move very little, not even a millimeter. There's absolutely no need for a large adjusting range.
Two more ball bearings to guide the teflontube in the XY-carriage and on more in the idler part of the carriage, so that makes 15 ball bearings in the XY-carriage.

@ chriske:
Your Printer does look very interesting! - It avoids many of the Ultimaker-Problems, but keeps the good parts of it!
Could you share your files (stl) with us, so we could build such one too?

The thingiverse-Links I found in your Posts do not work... :-(

Quote
chriske

Quote
newbob

Quote
chriske

Quote
newbob


I would not dismiss delta design too quickly. As far as I am aware only delta and light weight corexy printers can print at over 100mm/s with decent quality.

Not a Delta but mine(own design) can print also with that speed and even faster and with a very high perimeter quality.
The printer in the image is a prototype and works perfect. Busy making a larger version this very moment (heatbed 200x300mm), almost done.

For a brief moment, after I wrote that delta and corexy 'rule' , I considered Ultimaker design but overall I think it's performance sits in between speed and precision, and probably 100mm/s is too fast for good precision (especially for larger printers). Please post your results when you get your printer up - I'm curious how it works out.

That printer(I named it Ulti-Printer(The name Ultimaker was already taken... ) is already in service for a few months now. It outperforms all (about 20)printers by far we've built in the past. The perimeters are superb, no ripple at all, no ghosting around holes.
I expect the larger version I'm building right now will do the same.

Very nice design, too bad you've removed it from the internet. In your design, you decouple rotary motion from linear guides mitigating vibration caused by spinning linear guides- solution that allows larger printers and faster speeds. Bearing design is similar to MPCNC. Overall, substantially different and hardly an argument for ultimaker design.

Edited 2 time(s). Last edit at 06/19/2017 11:47AM by newbob.

I'll post the new files as soon V2(200x300) is up and running. That shouldn't take to long. In a few day's I expect 'first print'.
One major difference compared to V1(200x200), I installed a siliconen heatbed instead of the classic PCB. That siliconen bed heats up in about 10/15sec to about 70°C...!...

As soon the printers is finished and running I'll start a new thread about this Ulti-Printer.

See you...

To be clear on the matter, it takes lots more time to print/assemble this Ulti-printer compared to say a i3Mk2.

I am going to see a creality CR 10 within a week ...but....can anyone explain me how different it is from the Son of mega max....I have a little confusion.....with all that sturdy frame , industry actuators etc... It still has to print slowly for good output....but so is the case with CR 10....then where is the difference??
If my prints are great with the slot bearing guides....why should I go for more costly guides...I can simply replace them after a year or two...etc etc.

I am going to see a creality CR 10 within a week ...but....can anyone explain me how different it is from the Son of mega max....I have a little confusion.....with all that sturdy frame , industry actuators etc... It still has to print slowly for good output....but so is the case with CR 10....then where is the difference??


If my prints are great with the slot bearing guides....why should I go for more costly guides...I can simply replace them after a year or two...etc etc.


Also..if um style gantry wil not give more precision then we should go for corexy...anyway speed is never a priority....it's better to start printing in the evening...let it print overnight and the next morning , you get up to see a wonderful print....so it won't matter if it takes 8hrs or 12 hrs..it's night time
Plus....in case if the expiry of the polyjet printing patent leads to developement of polyjet extruder attachment to fdm mechanisms...then my machine should be accurate enough for the mount(but that's still a far cry)

So how will I be able to maximize accuracy....that's the final question.