We recently finished a project for Proctor and Gamble that set out to highlight the sustainability efforts being put into some of their most popular products. The concept was a three-walled interactive video display that was triggered by placing one of 6 different Proctor and Gamble products on a center pedestal. The product replicas had to act as both a trigger for a fully immersive video experience, as well as a clean white surface to be projection mapped on for the experience. Here are some of the things I learned that might be helpful for your 3D printing projects.



  1. How industrial design meshes with creative directors and strategists
  2. Scanning vs. modeling from scratch vs. manipulating pre-existing models
  3. Nothing works when you need it too – isolate, re-work, and repeat
  4. Manufacturing options, external vs. internal
  5. Finished products!

1. How industrial design meshes with creative directors and strategists

The process isn’t that far of from what it would be at a traditional industrial design shop. The hardest part generally is that the creative directors and designers aren’t as familiar with the materials and processes that are used in most industrial design projects. It’s important to show a number of different options to allow your manager the ability to make an informed decision, even though there’s probably always going to be a favorite.

2. Scanning vs. modeling from scratch vs. manipulating pre-existing models

We went back and forth on this one. At first, I was under the impression that 3d scanning would be easy – we’d be able to snap a few photos and load the models, no problem. I was quickly proven wrong. After snapping over 100 photos, and about 2 days of processing in Autodesk’s Recap, I got something that looked nothing like a bottle of Tide, but more like an abstract meatball.

Modeling from scratch would have been doable, but time-consuming. With the short timeline and the number of products it wasn’t the best option.
After doing some quick searches through Grabcad’s community library and TurboSquid, I was able to locate 5 models that were close enough to the products that we’d be reproducing. I purchased all 5 and manipulated them in Cinema4D. The outlier being a Charmin Ultra 6 pack, which I ended up modeling in Fusion 360. I updated the shapes to the current product, simplified the shapes for better (and easier) printing, and added registration keys for the bottom alignment system (each model had to be placed directly in the center of the podium, perpendicular to the projector for the graphics that were projected onto each product).


3. Manufacturing options, external vs. internal

This question can usually be pretty easily answered by considering 3 main constraints: time, quality and quantity. We had very little time to complete this project. Lead times for printing with a company like Shapeways can be upwards of 4 weeks, with prices nearing $1,000 (for the size of the products we needed). It wouldn’t have been possible to get six 10-inch models printed in time and under budget.
Upon weighing these options, we concluded that it was cheaper and more cost effective to buy another printer in addition to the one we already had and print all of them in-house. Also, for each of the 6 products we needed a back-up copy in case anything was damaged, bringing our total to 12 models. With an average print time of roughly 36 hours per item (with two printers), we were just able to get 12 models printed and shipped to Canada to join the rest of the booth.

4. Nothing works when you need it to – isolate, re-work, and repeat


If you’ve ever worked with 3d printers, you know that there’s a 99% chance it will not work the first time. In my case, I was able to get a few models printed before we ordered the second printer. This is where things really started to go downhill.

We ordered an identical Raise 3d N2 printer to the one that we had that worked pretty reliably and after plugging it in right next to our other machine, the quality and reliability of the prints went way down. At the time, it seemed like they were both getting enough power, as the circuit never tripped. But 3 hours into the print, the filament would start to bind in the feeder cog, sputtering out filament at an inconsistent rate, which would lead to the overnight spaghetti experiments seen above.

Once we got both machines back up and running, at about 3 hours in, we’d see the same problem, but on both machines this time. We tried switching out the nozzles, printing at a bunch of different temperatures, using different exporting software, and dipping the filament in oil with no real progress. Each day at around 6 pm I would start printing, say goodbye and good luck only to come back to more spaghetti experiments in the morning.

It was only once we were getting down to the wire – when we had a local company print one in white ABS (we were using PLA for sustainability reasons) to help relieve some of the stress of printing the last few – that I finally figured out the problem. I decided to unplug the new printer altogether and move it into another room. Although the circuits weren’t breaking, the fluctuation in power coming to the 2 heated bed printers was enough to temporarily clog the nozzles which led to the sputtering that was messing up our printing.

5. Finished products!


And here they are! To a client, creative, manager, or consumer, 3d printing may seem simple, but just like any new technology, mistakes are bound to happen (especially with tight deadlines). We hope that in describing our problems, you might mitigate your own. If you have questions, hit us up! Happy printing!