First Impressions: WanHao Duplicator i3 Plus 3D Printer

Late last week I picked up a WanHao Duplicator i3 Plus 3D printer. The cost is low enough that I could justify it as a backup and secondary printer, as well as just to have first hand experience with this popular model. These are some thoughts after a few days’ use and a number of prints, with a mind specifically toward miniatures terrain and similar hobby work. The upshot is that I think this is a great printer for the price, well suited for terrain production, and very accessible to newcomers to 3D printing provided they’re willing to look for and utilize other documentation.

I also have a walkthrough here for newcomers of some basic 3D printing concepts in the context of miniatures wargaming. A detailed tutorial on 3D modeling and printing using a miniatures wargaming example is here.

Wanhao Duplicator i3 Plus all set up and ready to go.

Cost & Availability

It’s stereotyping a bit to say, but in many respects the WanHao i3+ is typical of low cost but solid quality Chinese electronics. WanHao itself doesn’t produce retail products, at least for the US market, so there’s a confusing medley of rebranders under which the printer is actually sold. Monoprice is probably the most common; they sell their version as the Maker Select Plus on Amazon and elsewhere for $400. I bought a PowerSpec branded model from Micro Center for $350; it’s apparently sometimes offered on sale there for $250, a steal. I refuse to let it not be amazing that I can pop out to the store for under an hour, spend just a modest amount of cash, and come back with a robot that can make “anything.”

It’s worth taking some note of that availability. A big part of why I got this specific printer was because I could pick one up right away that evening when my main printer went down for non-trivial maintenance. Contrast that with Prusa Research, whose manufacturing is continually overwhelmed by their popularity and notoriously have backlogs of weeks or more. I produce a lot of prints for events and other deadlines, so it’s useful to know what I can likely grab with essentially no delay if/when necessary.

Tools

The PowerSpec edition, and I assume most rebrandings of the WanHao i3+, come with everything needed to get started. Included are the few tools necessary to assemble the printer and some extras for later maintenance; enough PLA filament for a a couple small prints; a spackle knife for removing pieces from the print plate; and an SD card with a few ready-to-print test designs preloaded. Including an SD card is a nice small touch, removing the need to hunt down or buy one to get started.

The spackle knife is a somewhat negative note among the included accessories. It’s useful, and more so than some other removal tools. But it has sharp corners that can easily gouge the print surface. I quickly switched back to the print removal tools I’ve been happily using for some time (especially the small tool).

A sidenote on the topic of tools to get started, I also use a simple pair of curved tweezers all the time. They’re handy for plucking at loose bits and threads of filament, or holding a cloth or paper towel to clean off a heated hotend.

Left to right: Spackle knife packaged with the i3+; knife that comes with Lulzbot printers; the Foreasy print removal tool I actually use; and curved tweezers for plucking filament threads.

Setup

Shipped in just a few major sub-assemblies, physically putting together the WanHao i3+ is intuitive and takes only a few minutes. Again typical of rebranded electronics though, PowerSpec’s documentation doesn’t quite track with the final product or packaging. The first indicator is that a description of the locations for the handful of bolts to be put in isn’t quite right. More problematic, the booklet doesn’t mention a cable you have to plug in (don’t miss cable E!). But it all makes simple sense: Bolt the gantry onto the platform; attach the filament holder on top; and plug in cables A through E. The cables & plugs in particular are very neatly and clearly labeled. The whole process is quick, the necessary hex wrench is included, and many good videos and writeups are available online to correct for the somewhat unclear documentation.

Plugging in well-labeled components.

Usage

Using the WanHao i3+ is also more or less straightforward, with the caveat again that the documentation is not great. Unfortunately this includes the touchscreen prompts. Sometimes these are just funny quirks and bugs, like the bed leveling procedure listing “x/4” steps when there are really 5.

Sometimes they’re less funny. Most notable are the instructions in both the documentation and the onscreen prompts for bed leveling to adjust the print plate to be “a millimeter away” from the hotend. That’s excessive, and even contradicted by more detailed notes in the booklet. Getting this distance right is where 3D printing starts to blend art and science, and there are tradeoffs: Too far and the first layer won’t adhere well and the print may fail; too close and the print’s first layer may blob up too much, or the hotend even potentially gouge the print plate. I’m still learning the best distance for the BuildTak print plate surfacing supplied with PowerSpec’s version of WanHao’s i3+ as I’ve found it to grip extremely well. That’s good in that prints adhere very well, achieving which is more commonly the challenge in 3D printing, but can make prints difficult to remove. I’m adapting by permitting a bit more gap in the bed leveling adjustments to compensate. A distance more like 1/3 of a millimeter seems appropriate, such that you can just barely push in a standard business card. For those just coming to 3D printing that might not seem like much of a difference from “a millimeter,” but it’s actually huge—under common settings each print layer is only 0.1–0.2 millimeters, and that first layer is the most critical in the whole print.

Misleading prompt for the bed leveling procedure.

Adjusting the bed leveling with a business card.

With the print plate pushing down just slightly to accommodate this business card under the hotend, for about 1/3 of a millimeter gap, prints seem to adhere very well and remove reasonably.

In general the touchscreen interface is workable but not great. I wish more or all screens had the hotend and print plate temperatures on them, e.g., to avoid having to go into subscreens to monitor cool down once a print completes. Some of the screen flow is not obvious or ideal, and there are some minor oddities or shortcomings like the limited number of visible characters in the file listing. But the control screen works and is ultimately straightforward once you learn the quirks. With the printer assembled and leveled, getting going just involves slapping in an SD card, hitting “Print,” and selecting a file. I’ll almost certainly set up an OctoPrint server on a Raspberry Pi to drive the printer, to have a better interface and remotely monitor progress, but it is nice to have a built-in SD card reader so the printer is ready to go on its own out of the box.

Fortunately, these issues with the documentation and limitations of the controls are mitigated by readily available resources. Since so many people have this printer, there are many guides online from which to learn the basic routines, and multiple active forums on which to ask questions. For those new to 3D printing, I strongly encourage watching a few videos or reading a few writeups about assembling and adjusting the printer before diving in. The aforementioned OctoPrint and similar open source projects can also supplant and improve the control interface if desired.

For slicing 3D models into G-code to drive the printer, I have been using the open source Cura and as expected had no problems. Several commercial and open source options exist. All of the settings needed to configure slicers for the WanHao i3+ are listed in the documentation, and many notes on the topic may be found online.

Although probably not the quietest available, I find WanHao’s i3+ to be very quiet. Certainly not a problem to run in a den or home office and not be noticeable elsewhere.

Print Quality

Straight out of the box, with no adjustment or tuning beyond basic bed leveling, I think WanHao’s i3+ produces very good prints given its price. Certainly they meet my expectations for tabletop ready miniatures terrain with which I’m happy to play games.

Some first prints on the WanHao i3+, from my Kolony designs.

Closeup on a building.

Fine details on small scatter terrain pieces.

With some tuning of settings I’m sure the quality can get even better, and no doubt some will be required for more challenging designs featuring tough retractions and bridging. Physical modifications of varying expense and difficulty to improve the printer can also be made and are extensively discussed online. An easy one for which I’ve already ordered parts is adding a brace to the gantry to further reduce unwanted motions. Thicker print plate chassis are also available and seem simple to install, reducing bed warping and making leveling easier and less frequently required. However, I think this printer is more than suitable for miniatures terrain out of the box.

Filament

WanHao’s i3+ uses 1.75mm filament, probably the most common size these days. Unlike some other popular entry level printers it isn’t restricted to proprietary spools, a huge boon for better availability and lower costs. Just as with the printer itself, it’s nice to know that in a pinch I can run to any of several nearby stores and grab more filament. A kilogram of PLA, almost certainly the most commonly used filament type for miniatures terrain, runs about $15 to $23 for typical quality without any shopping around for a better price and is enough to print quite a pile of models.

Cost and Longevity

From a filament price we can do some rough calculations on the cost efficiency and longevity of the printer. Obviously the value of a 3D printer can be hard to quantify. If you do custom design work it could be invaluable in expanding your capabilities. Just being able to acquire and build niche models is similarly hard to put numbers on. But we can do some basic calculations as a value floor. In particular, presumably one of the tradeoffs of a low cost printer is some reduction in expected lifetime. As a baseline, how long does this low cost printer have to last to make sense under the minimal use case of just printing existing designs? In considering miniatures terrain of the styles in which I am most interested (i.e., buildings, not interior tiles), we can put some numbers to that through comparison to buying MDF terrain.

Kolony 4×5 Habitats

A Kolony 4×5 Habitat, rendered above, is arguably a bit more detailed but fairly directly comparable to a simple MDF building commonly used for Infinity and other ~28mm games. It consumes ~120–150 grams of filament depending on whether or not it’s printed with a floor (the roof is also designed to be optional and easily made with foamcore or styrene instead to save print time, and burns about 1/3 of the filament, but is included here to fairly compare to similar MDF buildings). So a standard 1kg spool for $15 will produce 6–10 of these buildings for $1.50 to $2.25 each in direct filament costs. Electricity consumption is negligible for home use (i.e., not mass production).

Compare that to about $8–$9 for a comparably sized simple MDF small building (e.g., from Shark Mounted Lasers or Black Sheep Industries, both of which I play on often and like a lot). With the printer factored in at retail pricing, the 3D printed buildings using $15 spools become cheaper at about 50–60 small buildings [e.g.: 54*$1.5+$350=431 while 54*$8=$432]. Using $23 spools the crossover is higher, but not considerably.

So, as a very raw measure of pure economic sense, will this low cost printer produce at least 50 small buildings before additional costs are incurred, such as its semi-consumables needing replacement (like the print plate surface), or more serious repairs becoming necessary? I don’t myself know yet for sure. But I’m fairly confident it will.

Total costs for collections of small buildings for MDF versus 3D printing.

From that simple evaluation the capabilities scale very differently. 3D printing simply takes a long time and as such isn’t well suited to producing large terrain. It’s much better applied to producing small pieces to augment and detail larger constructions. On the other hand, 3D printing scales well with increasing model complexity. Even simple structures like the Kolony Outpost and Storage Shelters, the prints pictured earlier in this post, would be somewhat more complex and costly MDF models, but still only use about 100–150g/$1.50–$2.50 of filament with typical settings and take similar or less time to print. The Kolony BioDome, pictured below, also only consumes about 108g/$1.62 in filament and similar print time. But an MDF version would be very complex with numerous parts and almost certainly sold at a good bit more than $9.

Kolony BioDome (printed in HIPS on a Lulzbot Mini).

What these very rough calculations indicate is that if you’re producing a non-trivial but modest amount of terrain (a couple tables’ worth, figuring a typical generic layout is equivalent to ~15 small buildings in the analysis here), then there’s some reason to believe that this low cost printer will last long enough to make basic economic sense: Provided it does last that long, it’ll be similar or better in terms of pure dollar outlay relative to prices for reasonably comparable market offerings.

This isn’t to say 3D printing terrain like this makes sense for everybody. Maybe the finish quality isn’t acceptable, you feel it simply takes too long to print, or printers are just too much hassle to work with and maintain—these are all reasonable viewpoints to hold! Further, maybe you simply don’t envision printing enough to make it worthwhile. Alternatively, maybe printing is extremely valuable to you because you love a niche game for which models are really only available through 3D printing, or you have grand plans for crafting many boards full of bespoke, personalized terrain. I’m just arguing that under some plausible assumptions, parameters, and requirements (such as acceptable quality level), the per-piece cost using this printer is reasonable.

Newcomers

For people coming to 3D printing new, the WanHao i3+ seems a very reasonable option for those willing to put just a little effort into researching, thinking about, and using the printer. It’s affordable, prints well, and is ultimately pretty straightforward. My only hesitations for true beginners stem from issues like the touchscreen prompts and documentation. You have to know just enough to realize those are a bit off and then either figure out or track down better information. Light searching will also yield a number of improvements that are easily made but come built into some other, more costly, products, such as better G-code boilerplate to plug into your slicer to make printing more convenient by moving the hotend entirely out of the way when done.

A more subtle example is that the printer’s design essentially has three points of contact along the Y axis: Front, back, and the gantry. If they or the underlying surface (i.e., the table or shelf) are not level, it can rock a bit. So, for example, I assembled mine on one table and then moved it to another as its current home. A very slight difference in these surfaces led to the printer rocking just a bit front-to-back on the latter. So I loosened the gantry bolts, re-settled it in place, tightened the bolts back up, and the problem was solved. Anybody could do this very easily. But you’d have to be paying just a bit of attention to notice the rocking is possible, realize it could be a problem, and be just the tiniest bit mechanically minded to resolve it. This particular issue doesn’t come up with some other common printer configurations, so it’s an example of a small potential issue that could trip up an unwary user.

All of this is to say that, despite Micro Center’s questionable shelving decisions, 3D printing and especially this printer are not as thought-free as a typical 2D inkjet printer or similar appliance. If you’re happy going into it with just a hint of a hobbying and tinkering mindset, then WanHao’s Duplicator i3 Plus is probably a great option. Otherwise it might be worth looking elsewhere.

Summary

Long story short, my early impressions of WanHao’s Duplicator i3 Plus, and specifically the PowerSpec version from Micro Center, are very good. It and its filament are affordable and easy to find. Installation and usage is fairly straightforward. The output is good enough for ~28mm miniatures terrain, my main interest, even before any significant tuning. A tremendous number of people have one of these or a closely related model, so there are many tutorials and notes online. Numerous modifications are also possible and available to improve it even further. As a low cost printer I think this is a good option, and newcomers willing to put just a bit of thought toward the process will get a lot out of it for the money. Good luck!

WanHao Duplicator i3 Plus, hard at work!

Space Marine Detachments

Earlier I wrote a quick walkthrough about army list selection in Warhammer 40,000 8th edition. Unfortunately, aspects of those rules demonstrate again that 8th edition is very good, but neither as slimmed down nor formal and clean as it has been hyped to be.

The issues crop up from the new incarnation of Objective Secured, which enables models selected as part of an army detachment with uniform faction to trump other models for control of objective markers.

Preview of Objective Secured rule from the upcoming Chapter Approved.

Copy & Paste

To begin, there isn’t really a rule for Objective Secured. There’s a rule for Objective Secured, Defenders of Humanity, Knights of Titan, and so on. All the different codexes are coming out with the same rule under different names, e.g., the Space Marines’ Defenders of Humanity. Meanwhile, to maintain some parity for the factions that don’t have a codex yet, GW previewed an Objective Secured rule to be included in the upcoming Chapter Approved supplement giving that ability to all of those armies.

This copying & pasting is strange and unfortunate. Defenders of Humanity made sense at first when the ability was initially presented to be a unique Space Marines advantage. The presumption was other codexes would then arrive with different but similar abilities. But it turned out they all just have the same ability. Note that the abilty per se isn’t the problem. Diversity of abilities would be interesting, but also harder to balance. It’s the presentation of this universal ability that is problematic. Even minor carve outs such as for Fabius Bile and Fallen being compatible with all the legions and thus not breaking the faction uniformity of Chaos Space Marine detachments don’t merit copy & pasting the entirety of the basic idea into each book.

To look at just one issue, because there’s no universal term, the rule concept can’t be expressed in as simple and formal a fashion as before. E.g., to paraphrase previously:

Models with Objective Secured trump models without Objective Secured for control of objective markers.

Instead the new rules have to be worded in terms of trumping models with “a similar ability” because they all have different names. What does “similar” mean? In practice it’s understandable, but it’s not as formal as it could be.

All this duplication under different names is unfortunate but a hallmark of this edition, a consequence of “streamlining” the game by eliminating common text from the core rules and main rulebook and instead pushing it out to numerous copies in the codexes and datasheets. If all the factions are going to have this rule, there should just be one straightforward universal rule. That centralization would reduce the overall volume of rules and text, foster understanding, be more formal, and prevent errors from creeping in over time with successive copying.

This is also another example of how GW went to all the trouble of introducing a keyword system, but isn’t actually using it well. Objective Secured could easily be handled more elegantly and formally by granting units keywords if their detachments meet given conditions and wording the ability around those keywords.

Ambiguity

More critically, these rules contain ambiguities in their prerequisites. For example, Defenders of Humanity relies on the definition of a “Space Marine Detachment.” The other books are structured similarly. As explained below, that definition isn’t 100% unambiguous on whether or not a Space Marine Detachment can include multiple chapters within the codex, e.g., Salamanders and Ultramarines. It is explicit that Space Wolves and the other variant codexes are not included.

Beyond Defenders of Humanity, this ambiguity also raises rules questions about stratagems. Those listed in the codex are unlocked if you field a Space Marine Detachment. So must you field a detachment drawn from a single chapter to get access? Or would a detachment comprised of any mix of Space Marines suffice?

Mixed Chapters

In defining a Space Marine Detachment, the codex provides a list of chapter keywords which a “Space Marine unit” might have and then says:

A Space Marine Detachment is therefore one which only includes units with one of those keywords.

By far the most natural reading of this treats “with one of” as simply requiring each unit to have a faction keyword from that list, essentially grouping the last clause as:

… [units with one of those keywords].

The wording doesn’t actually make a binding to a single chapter keyword across the whole detachment, it permits a selection per unit. A detachment made up of Salamanders and Ultramarines would indeed quite obviously “only include units with one of those keywords.” Note that units can each only have one of the keywords in the given list, so “one” doesn’t instate any additional information as compared to, e.g., “any,” and is a more natural wording anyway for a selection of that type.

So in that straightforward reading, a detachment of mixed chapters would receive the benefit of Defenders of Humanity.

Single Chapters

Alternatively, a somewhat strained but plausible reading interprets “one of those keywords” as containing an additional stipulation that the units all share the same keyword from that list. It puts more meaning into the “one,” depending on how you look at it either adding an unstated “which must be the same for each unit” or grouping the last clause to the first clause:

A Space Marine Detachment is therefore one which only includes … one of those keywords.

In that case a Space Marine Detachment may only be comprised of “one of those keywords” and mixed chapters won’t receive the benefit of Defenders of Humanity.

A more complete definition under this reading would be, for example:

A Space Marine Detachment is therefore one which only includes units with one of those keywords, which must be the same for each unit.

Or, better:

A Space Marine Detachment is a detachment comprised solely of Adeptus Astartes units, all with the same <Chapter> faction keyword.

Intent & Practice

In trying to resolve that ambiguity, the intent is impossible to decipher from the text alone. Certainly it would be most traditional for a “Space Marine Detachment” to be comprised of a single chapter. However, it would also be quite reasonable and not at all out of line for Defenders of Humanity to apply to any mix of Space Marines. Chapter Tactics, which is explicit about being available only to uniform detachments, would then be an extra benefit for selecting units entirely from one chapter and strongly encourage that traditional makeup. The latter would be a more interesting structure and create somewhat more list building decisions, so it could easily be the intent.

In practice this ambiguity doesn’t matter much. The benefits of Chapter Tactics are so considerable and the ability to mix factions between detachments so flexible that there’s little reason to not field single chapter detachments. In theory though you might, e.g., if a future codex had weaker Chapter Tactics equivalents but strong characters from different chapter equivalents that you wanted to combine in one detachment but still be Objective Secured.

Next Level

As with the previous examples of textual (as opposed to mechanical) shortcomings in 8th edition’s rules, right now these generally aren’t actual problems due to the freshness of the duplicative copies and the competing interpretations of the ambiguity being obscured by the in-game strength of one. But they create openings for problems, particularly as the game evolves. With each new book there’s a chance to introduce a real error while pasting in its copy of Objective Secured. At some point there might be a real reason to field a detachment of mixed chapters/legions/dynastics/etc. and face real questions about what buffs it would receive.

The elimination of Universal Special Rules and other centralized concepts no doubt minimized the number of pages in the main rulebook. But it necessarily expanded the cloud of text replicated throughout all the books, with identical rules under formally and often even textually different symbols. Objective Secured and all its variants such as Defenders of Humanity are a prime example. That unnecessary inelegance and volume of text is unfortunate, a product of over-minimizing and false simplification.

The ambiguity in defining Space Marine Detachments isn’t hard to see and the definition trivial to improve. So it highlights that although much improved, Games Workshop’s ongoing reviews of 40k material are still very limited and non-technical. The playtest group is still too small, coming in with too many shared assumptions, not analyzing the language or mechanics formally enough, and sharing too much extra-textual context and intent to catch all possible interpretations and other potential issues. It is a much longer discussion for another day, but for a such a complex technical system (the game) and with such a large, diverse, and disconnected audience implementing that system (players worldwide), large scale public review is strongly warranted. That could be done in an efficient fashion to minimize burden on GW’s end, and, as other games as well as GW’s own recent public rules releases (for both 40k and Age of Sigmar) have demonstrated, in such a fashion that it wouldn’t hurt sales.

To take 40k to the next level of elegance and excellence, the design process should:

  • Focus streamlining on minimizing overall rules volume and duplication, rather than superficial main rulebook page count reduction;
  • Continue to expand the scope, diversity, and extent of peer and public review.

40k 8th edition is very good. But it could be even better with minimal effort and cost.

Battle Hex

Going through my folders and deleting a bunch of old stuff. Wanted to post this up for posterity. Draft for a series of tiles to get printed on heavy cardstock, used in various miniatures games (with and without hex grids). Later this would have been a great Kickstarter project, but at the time there wasn’t any great way to build up a market and confidently order enough to make the price reasonable. These days I don’t do enough drawing and graphics work to crank these out.