This was a hefty one! A customer asked me to make them a replica of a Bishop from the Lewis Chessmen, a 12th century Viking-era hoard of chess pieces which were originally hand carved (as you'd do) from walrus tusk ivory. It would be cast into solid sterling silver, weighing just a bit over 1 pound in total, and standing 3.4 inches tall. Of course, as they were all hand made, there were several types of Bishop in existence. The customer chose the Bishop from the D2 set.

If you'd like a Lewis Chessman of your own, take a look at the available pieces, or contact me if you'd like to discuss making it a different size, casting in a different metal (bronze would look cool), or any other customizations. Of course if you want one of the Lewis pieces that isn't already listed here, I'll be happy to make it for you. Just let me know which one(s) you're interested in!

For museum pieces like this, I generally get online first to see what models are already available. No sense in reinventing the wheel, and accuracy is key. Unfortunately a fairly in depth search didn't turn up much. The National Museum of Scotland has some 3D models but doesn't make them available for download (or even to purchase). I did manage to get some 3D scans from the British Museum, but they were sorely lacking in quality. Most of the detail was unfortunately in the form of texture files, which are not very helpful for 3D printing.

So, I modeled my own. I got as close as I could manage to the original piece, based on the photographs and 3D images from the various museums. I spent most of my time sculpting, and the face was the hardest part. The customer did request some minor tweaks to the mouth, which I obliged.

During this time frame I had to replace the UV display (these are consumable parts) and unfortunately the original display was no longer available, so I was "forced" to upgrade my printer to a newer model. Oh darn.

Once the model was complete it was a matter of 3D printing it in castable resin. I use BlueCast X10. This is an older resin, and as far as I've been able to determine has been superseded by BlueCast X-ONE. Until now I haven't been able to verify that X-ONE would work on my un-upgraded SL1, and so I kept using X10. Now that my printer has been upgraded I will make the switch, and all of my carefully-determined printer settings will change. Yay. Hopefully, the X-ONE settings are a little better documented out of the box, so I won't have to spend as much time experimenting.

So now I'm on the Prusa SL1S "Speed". The upgrade was absolutely worth it - the new displays are rated for more hours, lowering their cost-per-hour, and since their power output is much higher I was able to lower the cure time from 12 seconds per layer on the SL1 down to about 3 seconds on the SL1S. This brought the total print time from almost 20 hours down to only about 10 hours.

It seems like the time should be reduced by closer to 75%, but it's not quite that significant due to other actions besides the actual curing, like separating the layer from the FEP film, which doesn't scale with the cure time. Still, cutting the print time in half is more than good in my book, especially for such a tall model.

Print time depends on the number of layers, not the number of objects. And I always print at the smallest layer size, 0.025mm, to maximize fidelity. The height of the tallest object will dictate how long it takes to print. Furthermore, it's ideal to use the whole print bed as much as possible, to avoid hot-spotting on the UV display. Over time, the display will wear out. Areas that see more printing will wear out faster than other areas. So to maximize the printer's capability, I try to print more than I absolutely have to. In this case I opted to print two sets of standard-size (22mm) dice. I'll probably cast them in silver, but maybe I'll go with bronze for this set.

As an aside, the dice pictured above are an experiment. Printing dice is a tricky endeavor, you have to get the orientations just right  or the numbers won't come out clean. (Indeed several in this print run were already defective and will be cut away.) In this case I'm trying to print the dice themselves pre-sprued, so I can avoid the hour or so of wax spruing. I hope to just stick them in a flask and cast. Due to the orientations, though, the sprues are a little longer and more curved than I'd normally like. We'll have to wait and see how that works out.

It's hard to tell from the pictures but I actually hollowed the model, since it called for so much resin. There's a hole at the top and a hole at the bottom side. I filled in both holes with wax when I sprued it, so that investment wouldn't be able to get in. If I had planned to cast a hollow object, I'd have left those holes open.

Once the model was cleaned up, sprued together, and invested for casting, it finally dawned on me that I hadn't taken a single picture of that process, and that I should really have recorded the whole thing for YouTube. Oops.
 
Well, I started recording at the pour, so the second half of the project (including making a custom box to ship it in) is here:

And here are some close-ups of the piece at various stages of the finishing process:

Once again, if you'd be interested in a Chessman of your own, the models I've already completed are available to browse here, and if you'd like to discuss a different model, size or material then by all means please email me.

I hope you've enjoyed seeing how this project progressed!