One of my great annoyances with a lot of 3d printer reviews is how fleeting they are – the printer is unboxed, turned on, and made to go through a couple test prints (or fail to do so) before the end of the review. In my experience, that’s not enough time to really get a feel for the machine.
Fortunately, the good folks at Lulzbot were willing to loan me one of their new SideKick 289 printers for several months to properly put it through its paces. It’s a fun little machine with an admittedly premium price point, and now the question is: Does it live up to its potential, from the perspective of a wargaming hobby printer?
TLDR
The LulzBot SideKick 289 is a solid, reliable machine with a lot of quality of life features and a sturdy construction and an accompanying premium price point. The larger SideKick 747 is arguably a better value in terms of price, but for a wargaming hobbyist interested in terrain, utility prints, and some figure printing, you wouldn’t go wrong with the 289. Mac users should, however, be prepared for a bit of fiddling about to make the slicing software work.
Why the SideKick Interests Me
The Lulzbot brand is a fairly well known one on the broader 3D printing hobby space, with a well-deserved reputation for quality and good customer service. The Lulzbot Mini and TAZ 5/6 printers, in particular, were highly regarded as essentially being bomb-proof, extremely reliable machines for day-in, day-out printing with minimal fuss. But the company experienced a rough patch in 2019, with the somewhat lackluster launch of the Lulzbot TAZ Workhorse, followed by the original holding company, Aleph Objects, going out of business.
The SideKick line is, in essence, Lulzbot’s reintroduction into the hobby 3D printing market (under the new ownership of FAME 3D – Fargo Additive Manufacturing Equipment 3D) at a more affordable price point than their arguably prosumer TAZ line that’s marketed at universities, maker spaces, etc.
I was immediately interested in the potential for an easy, “plug and play” (insofar as FDM printers ever can be), Made in the USA printer at a price point that was a little easier to swallow than their previous offerings.
The other potentially interesting piece is that the SideKick uses Lulzbot’s Universal Tool Head mounting system – which means you can potentially swap toolheads relatively easily, allowing your printer to tackle different tasks with the right tool and without necessarily changing nozzles all the time, which is one of those things that is relatively straightforward to do, but if it goes wrong, it goes wrong fast.
For example, you could have a standard “daily driver” toolhead for most things, and then go down to a very detailed toolhead to small items, a toolhead with a large nozzle for stuff where speed is of the essence, or swap to a hardened nozzle if you’re printing abrasive materials like carbon fiber filaments that tear through normal brass nozzles. That’s both a neat idea, and would be something that’s fairly unique at the price point.
The Specs and Pricing
The basic TAZ SideKick 289 has a 6.375 x 6.375 inch square build plate, and can print things up to 7.125 inches tall. That gives you a build volume of just over 289 cubic inches – hence the 289. The actual body of the machine is about 17″ deep and 15.5″ wide, though you’ll want another 5″ of clearance on the right side of the machine for the fold out spool holder arm, and about 18″ tall.
One of the standard features I’m most excited about is the inclusion of auto-bed leveling via a BLTouch sensor. As far as I’m concerned, auto-bed leveling should be a standard feature on all but the most cut-down printers on the market today, and the BLTouch is a fairly reliable way to do it. Previous TAZ printers have used a slightly overly clever system involving touching a nickel-coated nozzle onto a round disk above the build plate, and using when that circuit closes to measure the four corners of the bed, but it’s always been a little fussy in my opinion. I’m glad to see the SideKick go with something mechanical and reliable that doesn’t reinvent the wheel and provides plenty of data points to the algorithm that actually compensates for all those measurements.
The specs past that are very variable as there are a ton of configuration options – something that I think may be daunting to new users, but for people looking for a 2nd hobby printer with a smaller form factor, might be great. You can look at those directly here.
The “Recommended” configuration is the SK175 toolhead, which is a 0.5mm brass nozzle toolhead that uses 1.75mm filament, a PEI-coated glass bed, and an LCD controller. This will run you $1085.
The SideKick as I would recommend it, and as reviewed, are:
- The same SK175 toolhead. It’s the most affordable, and is a solid workhorse toolhead. The SK285 is a viable alternative – it’s the same 0.5mm brass nozzle, but it uses 2.85mm filament. 2.85mm filament is sort of the Betamax or HD-DVD of the filament printing world – there are arguments for it, but it’s increasingly hard to find. All of Lulzbot’s TAZ series of printers used to use it though, so at the moment, the Universal Toolhead options are dominated by 2.85mm offerings. This may change in the future, but if you’re looking with an eye toward being able to use a bunch of different toolheads, the SK285 might be the way to go. Conveniently, they’re the same price.
- Magnetic Flex Plate. This is a PEI-coated spring steel sheet. Given the vast majority of wargaming hobby 3d printing is done on PLA, I think the quality of life of just being able to remove a sheet, flex it a bit, and have the models pop off is a godsend. I have one of these on my Prusa Mk3S, and I’m never going back. Also, it’s $30. Unless you can really articulate a need for a glass sheet, do yourself a favor here.
- gLCD Controller. This is just an old-style mono-color LCD screen with a knob for controlling your prints. Lulzbot gets marked down by some folks for not having a touch screen option, but honestly, I don’t mind this. And I highly recommend it. Without it, the only way to control your printer is via either an attached PC (which runs the risk of a ill-timed restart killing a print job) or something like Octoprint. Especially if you’re a beginner, this is essential. At a somewhat dear $95, I consider this to be a must have unless you really know you don’t need it.
- Filament Runout Sensor. This sensor will pause your print if it detects you’re running out of filament. It works pretty well, and again is a nice quality of life feature, especially for printing terrain, which is often rather filament intensive. It’s $20, but it’s also something you can add on later if you’re trying to shave costs.
That’ll bring you to $1135.
Let’s Talk About Price…
That’s not cheap. It’s just not.
I don’t necessarily think comparing it to the ultra-budget printers like the Ender 3 is a fair comparison, but there’s a big, Joseph Prusa-shaped elephant in the corner (alright, that’s a strained metaphor…)
The Prusa Mini, which requires some assembly but isn’t a full fledged build it from scratch kit is $419 with a filament runout sensor and spring-steel PEI bed with a slightly larger build volume (admittedly, it’s not a direct drive machine).
The Prusa Mk3S+, similarly equipped and as an assembled kit, and with a markedly larger build volume, is $999.
I’ll get into this more when I get to my final thoughts on the printer, but it’s definitely something to be mindful of. The SideKick, despite a considerably lower price point than the larger-format, arguably professional-level TAZ Workhorse and above, is definitely still competing at a premium price point.
Unboxing, Assembly and Build Quality
Unboxing was easy, with everything clearly marked out and nicely packaged, and while not technically assembled, the assembly of mine was extremely fast. All the various holders and fasteners you need are in a bright, unmistakable orange color, and things go together quite quickly. A time lapse of me putting mine together is below – admittedly somewhat heavily cropped, as “From the side and bent at 45 degrees” is possibly the most unflattering angle imaginable.
But really, it went pretty smooth, and there were very few puzzles to solve in the process.
An important note is that, because of the various toolhead options, the Y-axis coming as a separate part, etc. it’s not auto-calibrated from the factory or anything like that. The controls and instructions for dialing in the z-offset (basically, telling your printer how far it is from where the sensor says “I’m at the bed” to your nozzle actually being at the bed) are fairly straightforward, but they need to be done. Here there’s another comparison to the pre-assembled Prusa, which does come this way, but I found with mine that I needed to recalibrate things after shipping, setting it up, etc. so this isn’t much of a point against the SideKick.
Lulzbot has been talking about the SideKick as “the most 3D printed 3D printer”, which is sort of an odd approach to take, as a lot of 3D printer manufacturers have been moving towards fewer 3D printed parts, and more machined or injection molded components to give things a slightly more “professional” air. That being said, the SideKick doesn’t feel cheap or slapped together to me, even with the number of 3D printed parts it has – the combination of that and the garish (and in my mind, sort of endearing) neon “Lulzbot Green” coloration make a lot of callbacks to a sort of brash, hobbyist-centric image of Lulzbot in the past. I really didn’t mind it – the printer feels ruggedly solid, everything fits, etc. The print quality itself is okay – there are some visible print lines from bridging across gaps, some wisps of stringing, etc.
They don’t detract from the experience, and again I find them a little bit charming in their own way, but I think it’s valid to ask if 3D printed components are what one should be expecting on a machine costing over a grand, and I’m not going to fault anyone for saying no.
On the other hand, it does leave things open for extensive conversion and modification work, which I’m looking forward to seeing. Personally, if I had a SideKick and was feeling particularly brave, reprinting all the parts in SnoLabs absolutely stunning Black Amethyst PETG would be extremely tempting.
Similarly, the SideKick uses a somewhat unusual design for its motion system, V-shaped rollers moving along the crest of an aluminum extrusion rotated at a 45-degree angle.
These are fairly easy to clean and seem reasonably robust, but I think a lot of people at this price point would be expecting either smooth steel rods ala the Prusa Mk3 or a more expensive TAZ printer or, in all our fever dream 3D printer fantasies, linear rails. That being said, while I would jot the current design of the system down as “curious”, the printer and print quality doesn’t seem to suffer from it, I found it fairly easy to maintain, and it’s likely easier to service than a more complex and “elegant” system. The motion is also very smooth – my wife commented that it seemed to be moving in a much more visually satisfying manner compared to my Prusa, and I think she’s on to something – the motion system seems to act with a…deliberateness.
I did appreciate a lot of the design and build touches on the 289 – the very easy and obvious belt tensioners, a very sturdy four pillar extruded aluminum frame, the ability to fold the LCD in and out to get at it from multiple angles (though having it on the right side of the machine with the SD card slot facing into the printer is a touch awkward), and excellent cable routing in nicely wrapped protective sheaths. It feels like a rugged and reliable machine in the tradition of the more expensive and frankly fairly bomb-proof TAZ line. Similarly, I think the ubiquity of the 45-degree rotated bars will open up a plethora of options for user-printed modifications, like overhead spool holders, etc. that I’m looking forward to seeing developed. Lulzbot itself already relies on this as essentially a free mounting system for the spool holder, filament sensor, etc.
Software
The software for the SideKick 289 proved to be the most consistent pain point for me. Lulzbot has decided to stick with a self-branded version of Cura, and I’ll be honest – I’m not a fan. Lulzbot is building off a somewhat older version of Cura, and to make matters worse, their Windows and Mac versions are just not in sync with each other. The Mac version doesn’t even come with the necessary configuration files to use a SideKick – I had to email support, and they sent me some configuration files I needed to add. That’s honestly not great, especially for a printer that’s otherwise rocking a very plug-and-play style to it.
This continued when I needed to update the firmware for the machine – despite getting new configuration files from support again, I simply could not get my Macbook Pro to talk to the printer – I checked the cable, the USB-C dongle I was using, etc. I ended up having to borrow my wife’s ThinkPad to install the software, when went like a breeze once I had access to the Windows version. I checked this with a recent download onto a clean machine, and this is still true. There’s also nothing on the Lulzbot website that tells me this – there is a note that the Mac version doesn’t support the M175 Tool Head, but nothing about the SideKicks not even existing in the Mac version. At the bare minimum there needs to be a link and guide to updating your files, but really – this needs to be fixed. Getting started in 3D printing is a fairly demoralizing process at times, and first impressions matter. Getting your printer out of the box, excitedly assembling it, and then just…nothing…and an email to support isn’t a great feeling.
The slicing itself is fine – it does the job, and does it decently. There’s some features I miss from Prusa Slicer, and I expect a PrusaSlicer 289 profile will show up eventually (there are files for the 747 available on the SideKick Owners Facebook group already), but really the difference between Cura and PrusaSlicer in terms out outright functionality and “Will I get a good print from this?” is minimal.
But overall? I’d like to see some platform parity here – Macs aren’t exactly rare and unheard of machines anymore, and we’re talking about baseline functionality, not novel and experimental features.
This does however give me a chance to talk about Lulzbot’s support – it is outstanding. They’re very responsive over email, and quite helpful – this has always been my experience with them. I’ve had them send me step-by-step instructions for performing some service, including there own annotations for where they have trouble in the process, etc. with my Workhorse, and that continues here. They’re great. This has not been my experience with the support at Prusa, which is allegedly known for their support – my filament sensor simply doesn’t work, and they’ve been very little help the few times I’ve tried.
Setting Up and Initial Printing
First, a tip: Don’t try to calibrate a printer using transparent filament. I keep trying to do this, because I have a small stockpile of “Natural” PLA and PETG from Protopasta. And while I love it as a filament, it’s absolutely garbage for calibrating. Also don’t use black or grey – against the dark print bed, it’s just hard to see. Pick a nice, bright color. Like say…neon green.
From here on out, I’m mostly going to be talking about the SideKick as it pertains to printing PLA – which is what 99% of wargaming-related prints are made out of.
The first thing you’ll notice about the SideKick is that it is loud. Turning it on, there’s an immediate fan noise that, when putting my phone with a decibel meter for about a minute I got a reading of 55.3 dB. For reference, the Prusa Mk3S I have, while it was printing, was an average of 59 dB when not running in silent mode. The good news is the printer doesn’t get much louder than that while printing. The bad news is is that that fan appears to be the one responsible for keeping the electronics cool, so it’s running if the printer is running, whether it’s sitting at idle or not.
There is an ironic problem I have with calibrating LulzBot machines – some combination of the slightly larger extrusion from the 0.5mm nozzle and the excellent PEI-coated surfaces they use means that I almost never have bed adhesion issues unless the z-offset height, which is the only thing that definitely needs calibrating on the machine, is wildly out of where it should be. Even when the nozzle is high enough that you can visibly seem the filament dropping, like in the top panel of this figure from Raise3D, it will stubbornly stick to the build plate.
For the rest of a printer’s life, this is a good thing. When you’re trying to get it calibrated, it’s remarkably vexing, as the usual method of “brush your finger lightly over it, and if it comes off the bed, you’re way too high” starts not being true well before you’ve “arrived” at the proper z-offset height. This is great for long term reliability and making the printer less fussy overall, but it does mean a wider band of potential settings you have to work through to get the printer dialed in correctly. This is where I really have come to value the SK175’s fairly open design – there’s a lot more visibility than I’m used to with a lot of printers, which was nice for watching initial layers going down.
Once I got things mostly dialed in, and to a place where I felt like the z-offset was in the right place, I moved onto a more complex model. It’s been my experience that the various first layer tests that are out there can be “passed” and feel like they’re giving good results, while some more complicated models show signs of problems – this has been true for both resin and FDM printers.
First up was the mandatory Rocktopus. That came out alright. Not perfect – one of the tentacles decided to make a run for it, but alright.
But I wanted to do some more testing. My go-to “Okay, I think I’m calibrated…” test for FDM printers has been Aria the Dragon by the incomparable Loubie. My office is scattered with literally dozens of tiny dragons used as filament swatches. The reasons for this are many. It’s a relatively small object, which means it won’t take that long to print or use that much filament – but also importantly, it’s much taller than it is wide, which means if you’re relying on a lot of surface area of PLA holding to the bed to make up for poor adhesion…you’re going to have a bad time. There’s some detail elements, and the space between the wings and the neck of the dragon are prime places to uncover issues with stringing. Basically, it’s a decent test model to make sure things are going well before you plunge into 18 hour, hundreds of grams of filament style prints. This turned out to be the right call, as I ended up with a four headless dragons. Which is sub-ideal.
Much fussing with settings ensued as I experimented some, sat vigil as the models failed, etc. The big hint is that they all seemed to fail at about the same time – within a layer or two of each other. In essence, what was happening was that the printer was slightly over extruding, which was leaving a small “crest” of plastic on the neck of the dragon which would, slowly as that column was built up, catch the nozzle and let the print be knocked off the bed. Some tinkering with the amount being extruded as well as the speed the printer was moving at sorted this out nicely, and several intact and rather nice Arias joined their headless brethren.
Overall, this was a pretty straightforward process, and once I stopped using transparent filament, fairly straightforward. Z-offset calibration is, universally I think, probably the biggest starter pain-point for people, and this made it relatively simple. Narrowing in on the range of values where you genuinely need to start honing in takes a little bit longer, but I find it really hard to fault a print bed for being too good at keeping PLA where it’s supposed to be.
Thoughts on Bed Size
The 289’s bed is small – there’s no getting around that. And at this point, I’m seeing fewer and fewer kits that have “small bed printer” options where STLs come in logically split pieces. There are definitely some pieces where I’ve run into problems – the largest pieces of the kit I’m using to build my “wrecked void ship” board, based on these files don’t fit unless scaled down. But a lot of them do.
I’ve been able to print fairly large terrain files, like the wall kits from my Samurai castle/Defense of the Imperial Palace board, or the pieces I’ve been using to make a small shuttle for my Five Parsecs from Home crew using the Dragon’s Rest modular vehicle kit. Some kits are definitely pushing it – for example, those I have custom made by The Terrain Machine only fit if I scale them down from “32mm” to “28mm”, which is more accurately Star Wars Legion size down to slightly smaller game sizes. Which, I would admittedly do anyway, but it’s definitely a factor. I’d say probably 80-90% of the kits I go looking to print will fit on the bed, but it’s often a little stressful.
What it does preclude is having multiple pieces going at one time. On larger bed machines, I will often load up the bed with several different pieces, which while not making the print go any faster (practically) does mean I don’t have to clear off the first print and start up the second. If I’m about to go to bed, one twelve hour print is more efficient than two six hour prints. But it’s also a recipe for failure if one part detaches from the print bed or a number of other potential misadventures. It’s definitely something I have mixed feelings about – I’d prefer the larger bed of the 747 just in terms of never having to worry about any of this again, but I’ve very rarely felt limited by the smaller bed. Ironically, the thing that most blatantly didn’t fit on the bed is a utility print I designed to corral some of my model weathering supplies.
Long Term Use
I’m going to be honest here – the SideKick is an absolute machine.
I’ve been hitting is fairly hard for several months now, having it primarily make terrain for various upcoming projects. It’s been churning out ruined scatter terrain and generic Sci-fi/Cyberpunk buildings for my Five Parsecs from Home boards, along with some of the pieces for my Defense of the Inner Palace board.
I’ve also used it for a number of utility prints, where the ability for a 0.5mm nozzle to move more plastic in thicker layers has been key. That includes a number of trays for my resin 3D printing area, where I won’t be overly fussed if resin gets all over them, to some hooks for rods in my “Closet of Opportunity” and to allow me to hang some things off a shelving system I have in my hobby room. It also churned out more of my airbrushing sticks which have been invaluable while I’ve been painting my Black Templar forces and trying to keep the white shoulder pads and black armor separate. And it’s done it largely without complaint – I think I had one noticeable print failure after I got things calibrated, and another where I looked at it, went “That probably needs a brim, but I’ll let it ride…” and sure enough, it needed a brim. I’ve had the thing for about four months now, and it’s definitely been a pleasure to work with.
Seriously, it’s achieved the same level of “fire and forget” as my Prusa Mk3S, which is genuinely saying something – the combination of the solid build quality, the print bed itself, and the BLTouch auto-bed leveling is just perfect for a solid, reliable printer. If the name wasn’t already taken, “Workhorse” would be an excellent descriptor.
A few specific observations:
Nozzle Size
I was worried the 0.5mm nozzle would result in obvious differences between things printed at 0.4mm. And this did come up once, when I wasn’t being particularly mindful of the larger width of each perimeter for the larger nozzle, and a piece that was meant to be a tight push-fit that held in place (a homebrew airbrush painting handle I’m working on) turned out a little too small. But generally speaking, I haven’t noticed the difference save that the shells of the prints on the Lulzbot have a reassuring thickness to them, and print speeds can be a little faster as the nozzle moves more plastic. A good example is this wrecked wall set from my Imperial Palace project:
Made in three pieces, two of them were printed on a Prusa Mk3S at 0.2mm layer heights, and one on the SideKick at 0.18mm layer heights. Visually, they’re extremely hard to tell apart, and I usually have to check the bottom to see if I can make it out. And once it’s got a layer of primer and some paint on it? Any difference is going to be completely lost.
For those guessing, the Lulzbot did the one on the right. And actually bailed me out – PrusaSlicer absolutely freaked out about something in the file and thought it was some sort of noneuclidian horror, while Cura just plowed right on through and printed it. Admittedly, the tolerances for this set are pretty loose, so I’m not worried about clip systems being slightly misfit or anything like that. By and large, I’ve been trying to keep one printer doing one project for consistency’s sake, but I haven’t run into any notable compatibility problems.
What is did result in is much better vase-mode printing. For those who are unfamiliar with it, vase mode printing is where, rather than printing a solid object, the printer prints only a single outside layer on one continuous stream of plastic. This is decent for hollow objects (though ironically, not vases if you’re putting water in them). This has recently been popular for 3d printing wargaming terrain, because it uses way less plastic, is much faster, and because it’s a single extrusion, you don’t get the little marks and blips of plastic where the printer starts and stops a layer. If a model’s designed for it properly, this can be really great.
The issue is, that single layer is all the structural support you’ll get. On a 0.4mm nozzle equipped Prusa Mk3S printing at 0.2mm layer heights, the default width is about 0.42mm. On the Lulzbot, because of the larger nozzle, it’s about 0.58mm (I measured with a set of digital calipers). That may not seem like much of a difference, but that’s 38% thicker – which is a significant amount. I don’t really advise vase mode printing for anything that’s meant to survive very long with a 0.4mm nozzle, whereas the little “Definitely Not Zone Mortalis” pillar I printed on the SideKick is reasonably durable. If you’re really into vase mode printing, an even larger nozzle equipped print head might be advisable (more on that below). By way of comparison, the Lulzbot printed piece currently serves as the place where I temporarily stash makeup sponges while removing oil washes, and serves this purpose well, while the Prusa printed piece didn’t survive to make it to the review.
Layer Height
Lulzbot Cura’s recommended layer heights for “Standard” and “High Detail” are 0.25mm and 0.18mm respectively. I’ve settled happily into using the High Detail setting for the most part – it’s close to the 0.2mm I print on my Prusa, and is a good balance of speed and detail for terrain projects. This is a side-by-side comparison with a 0.18mm print and a 0.25mm print of the same object – a 40K-style heavy gun emplacement.
The 0.18mm height one is noticeably better in terms of detail, albeit when photographed in particularly unflattering light and very close up for terrain. That being said, 0.25mm is a decent standard for functional prints, and if you have terrain with less complex detail – like Necron terrain for example – 0.25mm could serve very well, and would absolutely fly in terms of speed. Which might matter if you’re an event organizer or the like and you need to scramble to fill some tables. But generally? For terrain, I’ve settled at 0.18mm.
Changing Toolheads
Changing toolheads, which is one of the major selling points of the SideKick line (and indeed, the whole idea behind the universal mounting system) is indeed dead easy. You remove three thumb screws and unplug the single cable that connects your toolhead to the rest of the printer, get the new one, reverse this process, make sure to tell the printer you’ve changed things and…there you go.
I timed it for the sake of curiosity. It took under five minutes (4:47 specifically) to do the swap. Total time until I was ready to print, which involved changing filament, heating the new nozzle back up, etc.? 7:29.
If you’re looking for the SideKick’s killer feature – this is it. Hassle-free toolhead changing in an entry level printer like this is fairly rare, and adds a high degree of versatility to the machine. You can easily swap to a larger nozzle for really fast functional prints (I’ve made everything from a tray for our dog’s booties to dry off on after we wash them in muddy winters to pillars to support an airbrush booth to put it at a workable height – neither of these needed detail). Or, if you’re interested in printing with abrasive filaments, like those with carbon fiber in them (which may or may not be stronger, but look fantastic), you could switch to a toolhead that uses a hardened nozzle, while switching back to an easy printing brass nozzle for “daily driver” PLA printing.
Or, in the case of this review, switching out to a SL toolhead, which features a 0.25mm nozzle, for fast, highly detailed prints.
The SL Toolhead and Miniature Wargaming
The appeal of the SL toolhead is its small size – putting filament out of a 0.25mm nozzle means smaller layer heights, and thinner perimeters, which will make for more detailed prints. And the detail you can get from these printers is remarkable. It rules out using some more exotic filaments – I wouldn’t do wood or carbon-fiber filled filaments and expect anything other than nightmare clogs, but the point of this toolhead from my perspective is printing pieces for wargaming, which is best done with a boring, standard PLA anyway.
Setup, as mentioned above, is dead simple. I had switched from the SK175 toolhead to the SL toolhead in essentially no time at all. I did have to go find a roll of 2.85mm filament, which meant I couldn’t use the nice guide tube that came with the machine, but this isn’t really an issue – if you’re planning on heavily using the 2.85mm toolheads, get all 2.85mm toolheads. For the moment, there are a number of ones from Lulzbot that are only offered in that size, but I have hopes for the future.
I did have to do some firmware tweaking to deal with some z-offsetting related issues, but the poor Mac support notwithstanding, this was also just not a big deal. Switching in Cura was similarly swift, and we were off to the races!
The SL toolhead needs a very precise amount of z-offset calibration. Essentially, the thinner the layer, etc. the less you can rely on “Maybe it’ll work itself out” and you really need to be dialed in. To be perfectly honest…I never got there. I got close, but never quite to where I was comfortable declaring myself done. Admittedly, if I wanted to spend more time with the machine I probably could have done this, but at the end of the day the machine has to go back to Lulzbot.
I did get decent results using a raft, which is one way to compensate for some of the issues I was having.
Now here’s the tricky part – switching toolheads worked wonderfully. This enabled the attempt to use a 0.25mm nozzle in a way I’d never do with my Prusa MK3 for fear of breaking a perfectly good machine in a way where it would never be quite the same again. And in doing so, I ended up thinking that the average 3D printing wargaming hobbyist is best off…with an SK175. Here’s why:
This is a small house from a recent Kickstarter for 8mm scale Gothic Definitely Not Adeptus Titanicus terrain. I picked it because it’s a simple file, but one with some details. The black figure on the right is printed with the SL toolhead on a high detail setting (0.10mm), and the grey one with the SK175 (0.18mm). Ignoring the somewhat mangled bottom due to the raft, the detail on the SL-printed figure is better. There’s less ringing, because a 0.25mm nozzle puts out less plastic per second, so the printer moves more slowly. The details are solid, etc. It’s good.
But it also took eleven hours to print. The one on the standard SK175? It took a little over two. And it’s fine. It’s markedly better than the print I got off my Prusa Mk3, the windows look like windows, the grates on the vent and the garage look like grates, etc. And importantly, both of them still have visible layer lines, etc. They’re both still going to look a touch 3d printed from up close, but we’re talking about terrain that realistically is mostly going to be engaged with from afar.
When it comes down to it, I think the age of the high detailed FDM nozzle has come and gone, because resin printing has been advancing in leaps and bounds recently. Most people who are designing things for FDM printers are designing things assuming ~0.4mm nozzles, and designing their details accordingly. And the folks going for really detailed models, the kind where printing at a 0.10mm layer height is a good idea? They’re now designing for resin printers. An entry-level resin printer from AnyCubic or Elegoo is price competitive with just the SL toolhead. They’ll print the same model in higher detail, and they’ll do it faster (~4 hours is the estimated time on my Epax X1-4K).
Admittedly, you can pull off larger objects on an FDM printer than you can all but some limited large format resin printers, but the print time for those larger objects will be agonizing.
If you only have room for one printer, my thought is to decide if you want to tackle larger objects like terrain, or smaller objects like character figures, and choose a FDM or resin machine accordingly.
If you embark on high-detail FDM printing, know that you are headed into niche hobby territory. That being said – people do get some remarkable prints out, and I’m fully confident that if you’re going to spend the time to tune an SL-equipped SideKick to the level it needs to be turned, you can make awesome stuff. But for everyday printing for wargaming, especially 28mm-32mm scale terrain? I’d start with an SK175 as your workhorse toolhead.
The Value Proposition
This is where the SideKick 747 stands out, in my opinion, unless you can only accept the 289’s smaller footprint. A similarly equipped 747 comes out to $1345, and I think that’s a much more competitive offering when considering a world where the Prusa MK3S+ exists. And from this picture taken from the Lulzbot website that’s a mostly side-by-side shot, for a vastly larger build volume the 747 isn’t much larger. When it comes down to it, things like the various stepper motors, power supply, fans, etc. don’t scale down linearly with bed size, so more of the 289’s total space is taken up with “infrastructure” instead of “build volume”.
As much as I like the small form factor of the 289, I find myself eyeing its big brother, where the flexibility of a larger format printer that brings a 41% increase in space on the X and Y axis and a 37% increase in Z for 19% increase in price is arguably a much better value, especially when going toe-to-toe with the likes of a Prusa Mk3S+, where some of the Lulzbot’s features like the swappable toolhead system, the somewhat more rugged feel and superior service might be worth the extra cash.
Closing Thoughts and Future Improvements
So where are we at, when the dust settles and with a small pile of empty filament spools?
The perennial question is “Is the SideKick a Prusa Killer?”
The answer to that is no – but I also don’t think that’s a failing. I think the FDM space could use more competition, and slightly less ossification at both the bottom and top ends of the hobby/consumer market.
What it is is hot on Prusa’s heels as an absolutely viable alternative. Are you alright with a slight premium both for something Made in the USA and/or with top notch product support? Are you someone who would like to delve into playing with a lot of different materials and printing styles via different toolheads? Do you just want a printer community that’s a little smaller, more concentrated, but also fairly passionate about the products?
I think the SideKick would make a good first printer – solid support, the sort of quality of life features that make FDM printing easier (filament sensors, flexible PEI-coated beds, auto-bed leveling using a reliable mechanism, etc.), etc. But – and perhaps this is hinted at in the name – I think it would also make a great second printer, for folks looking for something a little more specialized, and straddling that line between “Plug and Play” and tinkering friendly, but in a way that’s very much designed to be easy and accessible. That’s where I think the 3D printed components, Lulzbot’s open source ethos, etc. all sort of meet each other.
There are a few improvements I’d like to see, most of which are quality of life iterations on an already solid product:
First, the support for Macs, software wise, obviously.
Second, Lulzbot uses a characteristic side-mounting spool holder, followed by a long span of PTFE tube to bring filament up and over the body of the printer and deliver it to the extruder. This is fine, except for two issues:
- It adds considerably to the width of the printer – ~4.5″ to be moderately precise – and
- The filament sensor being at the beginning of the PTFE tube means when the sensor says it’s out, there’s still a fair amount of filament left.
I’m not particularly worried about this for a full 1kg spool (I confess I didn’t have the courage of my convictions enough to test if the holder can bear the weight of a 3.5kg spool), but it makes it hard to use filament samples, which ironically is one of the places I end up needing filament runout sensors the most, as a lot of “approximately 50g” coils come with a lot of variance there. I would love to see an official overhead option for mounting the spool holder and filament sensor – I’m pretty sure the frame can take it, and for a small form-factor machine, there’s often more height available than there is width.
That’s really about it.
I think the SideKick 289 is a remarkably pleasant little machine to print with. It’s been soldiering right along, printing out concrete barriers and sidewalk Ramen shops for my Five Parsecs from Home crew, walls for the Imperial Palace, and because I wanted to see what could do if I really went for detail on the standard nozzle with a model designed for FDM miniatures printing, this Giant Worm from DragonLock which is absolutely going to terrorize players in an eventual Adeptus Titanicus narrative game.
Years ago, when I was car shopping, I read a review that stuck with me. The magazine was reviewing a huge swathe of performance cars, from the very, very expensive to the fairly practical. Of course, a lot of those ultra-luxury cars had extremely good ratings. But at the end of the article was one question for all the reviewers:
“What would you buy with your own money?”
If you asked me if I’d buy a 289, I think the answer is “Maybe”. The SideKick 289 checks a lot of boxes – but I would likely end up with a 747 if I was in the market for another printer. Because I have the space, and the larger form factor does offer a little bit more freedom and flexibility. But my bright green 289 has grown on me a lot – it’s a plucky little machine that’s done everything I’ve asked of it with both quality and reliability, and I think is an excellent option if one is looking for a smaller form-factor printer.
I’m going to miss it.
Permalink
Permalink