Trying to decide

[robert72]

I have been looking at both the delta printers and the cartesian printers. I prefer the delta for several reasons, but am concerned about the potential flaws.

The printers I have been interested in are the V2, the Blue Eagle and the Openbeam; on the cartesian side I have looked at the Makergear M2, the lulzbot Taz 4 and a couple of chinese printers.

I like the V2. It has the largest print area and is the most mature of the delta printers I have looked at. However, I have seen the thread about the problems with leveling the bed. It appears that it is not a problem that can be resolved mechanically, but is due to errors in calculations (rounding errors?) that are inherent with delta printers. The large build area won't do me much good if I cannot use it. I noticed that the blue eagle and open beam seem to solve the problem by automatically checking at least 32 points on the bed and basing their corrections on that data. I wonder if that is essential for this type of printer. I know that this thread has gone on for quite a while without a final resolution, although I am think they may be close to that.

So, does anyone have experience with large builds that take up most of the area of the print bed? and if so, what is their experience. My impression is that the problem only appears with prints done outside of the triangle formed by the three posts. And, does anyone know what and when the final resolution will be.

Sorry for the long-winded question. I like to do a lot of research before I buy.

[BenTheRighteous]

I printed a spool for one of my heavy-duty extension cords which took up the entire bed. It printed fine. It's been awhile since I looked at the thread you're talking about, but I think those guys are talking about errors that are fractions of a millimeter. For my purposes, IF I have the same symptoms, I cannot tell.

In fairness, it did take me a lot of patient effort to calibrate the positioning to where it is now, but I feel that I finally have it perfect and I am quite happy with the performance both inside and outside the build triangle.

Whether my definition of perfect is the same as someone else's, I can't say.

[robert72]

Ben,
Thanks for the reply. i agree that they are talking about fractions of a millimeter. However, a millimeter is 40 thousandths of an inch, which is in no way a fine increment of measure, particularly when the V2 is suppose to print to a layer thickness of .1 mm or 4 thousandths and several printers claim down to .05mm (.002") or better. The errors they seem to be discussing run to about .020" or .5 mm or more and that is with a lot of work. All of this means that if you try for the thin layers on a large piece, you could end up with your print head hitting the bed. Again, I am new to this and so take this with a grain of salt.

I am encouraged to hear that you have printed a very large item with good results. You are probably correct that, for most things, fraction of a millimeter tolerances are more than adequate. I am not sure of the final use of my printer, but if there were some fairly precise and large part that needed printing, I would like to feel that i could do so.

Thanks again for the balanced and informative rely.

Robert

[0110-m-p]

It does take a long time to calibrate it "completely" but freshly assembled my printer was good to 0.009" between the towers when the center and towers were at zero. Now that I have learned delta printers better and have spent a lot of time tweaking my tower rotation, arm lengths, etc I have my bed flat within 0.002" (0.05mm) at 7 points. The point at the center of bed and at each tower at R=110mm is zero while the 3 points at R=110mm 60-degrees between each tower are between 0.000" and 0.002" "low." I think this is pretty damn good and my prints reflect that.

Does it take longer and more effort to calibrate than a Cartesian machine? Yes
Can you get as good of or better results than a Cartesian machine? Yes
Is it worth the extra effort to get to this point? For me, yes...for you???

As for the people with 0.5mm of error...no idea what there issue is but I have a hard time believing it is an issue inherent to the Rostock Max or deltas in general. Everyone builds and calibrates there own machine, so there is a lot of room for error.

With that said, I'm still in the design process for a custom 300x300mm Cartesian printer so that I can get even more bed area without a significantly larger footprint. Personally I don't care about build volume as much as I care about build area, so the print height advantage of delta printers doesn't do a whole lot for me....I just love watching them move.

[Jimustanguitar]

As for the people with 0.5mm of error...no idea what there issue is but I have a hard time believing it is an issue inherent to the Rostock Max or deltas in general. Everyone builds and calibrates there own machine, so there is a lot of room for error.

And there are thousands of Rostocks out there. The thread you're worried about is me and the other 4 or 5 people on earth having that problem

[Earthbound]

Since none of the machines we are discussing use a truly flat build surface, take any "precise" measurement in perspective. The math to position the extruder on a delta machine is established. The arithmetic ability of the control unit is imperfect, but at an error level that is extremely low and likely smaller than other variables. Think I'm wrong? Tell me the calibrated thickness of the gluestick layer or blue painters tape.

Getting good prints is a matter of setting the machine up properly, calibrating it thoroughly, tuning the settings on the slicing program (after finding the best one for your application), tweaking the model to remove or improve problem areas, and managing expectations.

I have a part that I am printing. I've got everything dialed in to where it consistently produces an excellent result. I have been printing them one at a time during adjustment. When I plated several of them to produce eight in a single run, half of them were unusable. Could be spacing, could be a draft in the room, could be many things. I'll need to do further testing if I want to run a plate like that. And I need to accept that there will be variation on cold or hot days. I shouldn't expect identical results on the next spool of filament, particularly if I change suppliers or even use a different color. There is a lot of trial and error in this process. As much art as science.

[Proto]

Hi Robert, and welcome to the forum

I just went through this decision process about a month ago. Let me share with you what I found out. First, there is no "perfect" 3d printer. They all have strengths and weaknesses. If I remember correctly, Makergear and Lulzbot are fairly expensive (outside my budget).

I purchased mine used (A Rostock Maxx V1) from one of the forum members, Cambo3d. He had issues with setting up this machine which boiled down to some of the laser cut pieces were mis-cut which effected the alignment. He documented his whole build, problems, solutions, even updates and modifications.

The link to his build is http://forum.seemecnc.com/viewtopic.php?f=42&t=1228 Warning, it is 14 pages long.

I have been printing on it for about a week now. Other than a couple USB communication faults, a mis-set z level retract, and a few adherence issues of the part to the bed. It has been working great. By the way none of the issues I listed could be considered Delta only.

[robert72]

Thanks for all of the information everybody. The reason I am looking at the Makergear is 100% top ratings from customers, pretty decent build area and they offer a kit for significantly less money. Taz 4 has a good reputation and very large build area and they also sell it as a kit. I am trying to buy used, but am unable to find anything decent and Money is an issue.

With the problem we are talking about, only the customer that uses the entire build volume will see it. I am definitely encouraged by the note from 0110-m-p. It appears that he has managed to get the bed as flat as can be reasonably expected. I have to admit that the visual of a delta working is very cool.