banner perso

Some links about me. Many of my 3D designs are free. I also post on Google+ and in another blog, oz4.us

Tuesday, March 25, 2014

Quick comparison: pro versus low cost FDM 3D printers

Are so-called "professional" entry-level 3D FDM printers worth their price?


A LinkedIn user in my network asked for more data when I said that "low cost" FDM printers could compare favorably to "professional" (and expensive) printers. Sure, it is a bold statement, and it does not hold for every of the many "levels of quality", as there are just too numerous. But let me get into more details.
In my opinion, an Ultimaker rev. 2 (~2K€) compares very nicely to a Stratasys UPrint-Se (~23K€)
I certainly do not want to say the "pro" printers are crappy, nor that any brand is crappier than the other, but I had the chance to compare my Ultimaker prints with a "professional" Stratasys entry-level printer (i.e. 1200€ versus 23K€). The outcome is that I just would never swap the two even if I was offered the "pro" one for free! Huh.

Now I am a power tinkerer, I know how to print, I prototype the printer itself, and I cherish freedom, but I tried to keep this bias out of the following arguments.



By the way I use the expression "low-cost" versus "pro" , which is not fair to the Ultimaker as it just can print as "pro" as this "pro" one. Now should I use "low-cost" versus "high-cost"? I am NOT considering here other "pro and amazingly high-cost" industrial-grade printers, that are totally out of scope on this blog (as log as I am not given the chance to visit and disassemble one, an opportunity that the big "players" will probably never give me after this post).

Still I think that a client could have spared some money by NOT buying the "pro" FDM printer, and contracted people like me either directly for the printed parts (not my job actually), or to help mastering a low-cost printer he would have bought instead (more like my job). Or at least, he would have better bought a much more reliable selective laser sintering printer in my opinion (SLS). Nowadays, spending more than 5K in a filament-based printer is really throwing money out of the window.


For the following, I discarded the "very" low-cost printers (sub- 800€), because they are probably not reliable enough for business or industrial use. If you need to spend your time keeping the printer in a state of order, either the printer is crappy, or you are a masochist (or may be you are hacking the printer itself). So, I am considering here only the "hard wall" boxed printers such as the Ultimaker or the Makerbot (the latter being costlier for no added benefit and even worse perspectives in my opinion). Still, low-cost repraps are dirt cheap (about 500€) and good-enough results can be achieved by a skilled operator (just do not move them too often as setup and calibration are serious issues).

Also, "good" depends on how good and how often you want it to be! A rare but outstanding result may be better than just an "average" result each time, see this analysis! And the entry-level "pro" printers are just absurdly too expensive for the claimed reliability they offer in my opinion.

Some point of comparison

Support and maintenance: just like other FDM printers, "pro" printers also do fail (see below, reliability). And it takes much more time, especially when someone has to come and fix something on a "pro" printer: besides the cost, it impacts the very idea of "rapid prototyping". Also, because of the beautiful packaging and plastic housings of the "pro" printers, it takes both time and energy just to cleanup the mess around the print head.

It is much easier and faster to deal with a low cost FDM printer: e.g. 30 second to clean the nozzle, 5 minutes if the thing need to be disassembled or the nozzle changed (OK, 15 minutes when you really are out of luck and the entire world is against you?). You can also easily re-build some parts for easier maintenance without voiding the warranty. Nowadays though, the low-cost FDM printer manufacturers tend to make them sexier (eg. the Makerbot printers then the Ultimaker2). They produce much more beautiful printers that sell better, but it obviously does not help practical maintenance.

Reliability: to be honest, I will not say for sure because I did not get enough "pro" stuff yet to get an unbiased opinion. I really would like to think that that "pro" is more reliable, but I already saw a few issues with the print beds for example: the pro printer is printing ABS on an... ABS bed? Come on! Obviously it sticks well, but to the point it is sometimes not possible to remove the part without breaking it, or the bed (hence the "pro" printer came with a dozen print beds in the first place, wow!). This is just ridiculous for such a price, since I finally had to recommend reverting to makers' efficient techniques (eg. Kapton tape, liquid glue, Garolite boards, etc)

Interestingly also, you may afford a lower reliability for a much lower price if your business does not rely on the thing to be "certainly" ready in 5 minutes, or if it need not be exactly the same each time (artwork, for example). Note that I still did not say which is more reliable, as I do not have enough data to tell.

The prints themselves are as good as what you get from a "pro" printer, and serious experiments have proved the fact.

Now, the "pro" printer comes with a soluble support that probably helps a lot for overhangs, as the single-nozzle entry level printers will usually require more post-processing (but without additional costly and nasty chemicals).

PLA support with a counter-nut and a nylon hub.
This thing tunes the free play for a rod,
quickly printed with "almost-rough" 0.18 mm layers.
Precision: so far I compare favorably what I can do to what my client achieved, mostly out of the box (experience may talk, but the "pro" printers are advertised as being ready-to-print and user friendly!). I never saw a part that was printed at a level of details that I did not achieve already. Reciprocally I am still expecting to see how fine they can print with the costly beast.

The numbers are not bad for the low cost FDM printers: the layer size can be set to whatever you want (I went down to 0.04 mm where mechanical noise gets noticeable). This is way thinner than most of the "pro" printers.

Also, low cost FDM printers let you change and use other nozzle diameters, either for finer details or for faster prints. It takes two minutes to change the nozzle on an "open" printer. I am not sure nozzles sizes are even available on the well-boxed "pro" printers, but it would be hell with all the clips and tortuous heated chamber and rest zones (note that a heated enclosure is mostly a good thing, to take into consideration for low-cost FDM printers). Now they most probably feature some proprietary couplings that would not let you try if you were OK to void the warranty anyway.

Robustness: the question is not easy to answer. Correctly printed, an ABS object is as robust as an ABS object printed on another printer. Now, low-cost filament may lower the quality of the print obviously, while the filament being sold by the "big ones" is of constant quality and technical properties. So if you need safe numbers, you probably better go "pro", because most of the low-cost printer filament companies do not publish data that are reliable enough given the variety of settings, nozzles and printing configuration that exist in the many "low-cost" printers. I still have to make some real field comparisons. But anyhow I do not like ABS as a sturdy material. It does not like UV, it is brittle imho. Now it is more impact resistant and more importantly heat-resistant than the harder PLA, for example.

But, hey, I can also print in Nylon when I really want a resilient part... Sure, ABS and Nylon are not the same material... but I am the one with choice! Professional printer owners just have to buy the very expensive sealed cartridges that they are being sold ($400+/Kg, i.e. tenfold more expensive!). And they probably will have no chance to hack it around without voiding the warranty. So make sure to check what you want before buying.

Printable volume, and portability: the Ultimaker (eg) has a bigger volume and larger surface than the "pro" printer I could check. May be even 4 of them would fit in the space that the latter uses in the office. This can be a false issue according to the client, but it also means I can store my printer elsewhere when not in use, or more importantly, that I can bring it to the field where my client need it for really rapid prototyping, which is simply impossible with a "big" one (it can even be operated on batteries as a rucksack!).

No way you can get this on a "pro" printer,
too bad if your goals are design and art!
Finally, the "low-cost" bring many more choices for the printable volume, e.g. the "delta" printers easily print very tall objects, still for a fraction of the cost of a "pro" printer.

Choice of filament: according to the "pro" brands, you have a few proprietary ABS variants, sometimes just nothing else. The "low cost" give a huge and always expanding choice: PLA, Nylon, PVA, PET, PC, carbon fiber-, and even wood- and chalk-loaded polymers... When Nylon or PET is more adequate for industrial or biological applications, Laywoo or Laybrick are just way better for aesthetics purposes (art, ergonomics, etc). And you get even more color than materials. It does not mean it is as easy as sticking with a very few materials, but you get what you pay for.

I would certainly not dare feeding mineral-loaded filament to a "pro" printer, as it would certainly "void" the warranty of the printer, if ever I manage to get around the proprietary and expensive cartridge system that would certainly refuse the test in the first place. New filaments are brought to the market at a pace that "pro" reseller just cannot afford.

Software: here again the "low-cost" printers give you a larger choice, with more frequent updates as it happens with open source tools: contributions and bug fixes every time it is useful or needed. Some, like Cura, are especially suited to the Ultimaker, which in turns make the experience very smooth to set up and use this printer. Moreover, it accepts plugins very transparently, for some specific goals like the variable temperature print above or setting changes at some place in the print (eg. pause to change filament, change in temperature at some given height, etc). You most probably are not allowed to tinker with the "pro" software, that even sometimes will let you choose only between two layer heights!

This is also the same for the printer firmware itself, which can be upgraded for new features or bug fixes, and which can be tweaked at will. Note that I did not have to upgrade mine for the last 6 months at least because it is just stable. Well, I compiled my own version to have a few specific command shortcuts and macros in the LCD menu, once again something you can forget with a closed-source printer of course.

When you buy a "pro" printer, you are certainly stuck with the specific software, which ought to be adapted to the printer but which will not let you do anything else easily. It may also come with embedded planed obsolescence in the long term, a risk that is hard to evaluate in the first place!

Conclusion

In my hash opinion, the clients that buy the "low cost" variants of the "pro" FDM printers are just being fooled. These printers ought to be wiped out of the market because of their incredible price tag compared to the features they brag about. They only exist because of decades of "serious (aka BS) business" marketing inertia that should have simply vanished a few years ago. As a client, running a real business certainly does not mean you should think that the price of a 3D FDM printer is a warranty of its efficiency. The consumables are outrageously more expensive for no added benefits as show by independent studies.

It is no surprise to anyone that Stratasys bought Makerbot Industries, because they made "low-cost" printers that started to address a market that the former could not because of their incredible margins. These "low-cost" printers were reliable enough for a "big one" to invest massively in them without shooting themselves in the foot (aka "serious and costly business" image). So the "big one" smartly bought a smaller promising company that probably manufactured printers as efficient as their own entry-level line of printers, but for people that would certainly not buy the too expensive ones. And since some "business" clients keep on buying the more expensive former line of printers, they make even more money.

Now, with no surprise, Makerbot immediately closed their open-source-built printer (nasty move), then raised their own price tag, and finally used the usual corporate bullshit language. Beware buyer, Makerbot is no good brand to look for, unless you still want to pay a lot for troubles.

In any case, I feel sorry and even sad for the latest victims of the low-cost "professional" printers, because the deal is outrageous. Let me be clear though, I guess the mid- and high-end line of their printers may still be where "really serious" business happens, with real reliability and much better controlled prints, but this was quite out of scope in this post.

Final note: Stratasys versus 3D Systems?

Do not believe I am complaining here about this printer from Stratasys so as to promote 3D Systems indirectly (probably the other "major" company).

As far as their "low cost" printers are considered the latter produces an amazing set of false claims and corporate BS. Check this post from Hrvoje Čop for example, dedicated to the "lowcost" 3DS Cube Pro. Would-be buyer, do please check it first! The cost may be significantly lower than the herein depicted UPrint, but it comes with a lot of amazing, straight lies!

This is really NOT a proper behavior for such big companies. As best, it looks like they relied too much on their patents, and were overwhelmed by the useful, real and independently validated improvements makers brought to the market while they were just making abusive money on old pricey technologies. When their patents are over, they will start to be in real trouble bragging about obsolete features for overpriced printers, accessories and consumables. Oh, they already do it for the low-end of their printers.

Now, who on earth can reliably tell where the "real" threshold is between their useless vs. useful printers?!

update (jan 2015): I was told about another side-by-side comparison, with the added view from the "pro" owner here. Make your mind :)

About Me

My photo

If you know me and you cannot tell exactly what my real job is, then you probably found the right Jeremie. Check zax.fr for some pointers.

I am self-employed and I help start-ups, research centers, small companies with their needs related to computers, sensors, data processing and mechatronics. If you have a project and know what "R&D" is, then you already sparked my interest ;)