Oh man, it's been a busy couple of weeks. Prepping for Maker Faire, exhibiting, and recovering from the flurry of interest and feedback were some of the most tiring but rewarding days we've ever had! In the meantime, we've been hard at work trying to sort out the motor issues we talked about in the last update.
To get the bad news out of the way first, we're still having troubles with our motors. Having confirmed and now understood why the old motors were failing (it turns out that they are rated for intermittent, not continuous, loads) we're now charging ahead on the stepper front as the cost of a geared motor (and associated drive electronics) would increase the costs significantly, and we want to keep them as low as possible going forward. Again, this will not affect campaign units at all - but we don't want to have to drastically raise our price post-fulfillment because of our motors! Having a larger DC motor also diminishes our current sensing ability - minute variations in load will not show up as clearly as they seem trivial to a larger motor. As an extreme example, imagine a small crack in the road. Roller blades or skateboards will feel it - a bike or car, much less so.
So a stepper motor is something we're looking at first and foremost. The costs will be kept similar, the reliability and longevity increases, and people who later want to hack together their own ProtoCycler will potentially be able to use a standard RAMPS or similar board, instead of needing their own custom board (we'll continue to use our own board regardless). The problem with the steppers is that there isn't any current feedback, and while a larger DC motor won't be as effective in this regard, a stepper just doesn't have the capability at all. We've done some research into using back EMF, etc, but it's seeming like within the scope of what we do load sensing on steppers just isn't going to be feasible.
The reason we value load sensing so much (which the DC motor provides through it's current sensing) is first and foremost that it allows us to account for any unexpected variations in the flow rate. In any feed auger system, there is a certain amount of slip, and for reasons beyond control this can vary slightly. Measuring the current allows us to see if the motor is pushing with more or less thrust, which directly relates the pressure the plastic is under, which in turn directly relates to the diameter of the filament. If we sense that the pressure has just risen, we can pull faster...and with a well tuned system, nearly eliminate any variance. Current sensing was also used for low pellet detection, a few other nice to haves, and was the basis for Intelligent Extrusion, which we were pretty darn excited about! But ultimately, our promise to you - our customers - is to provide the best filament extrusion device on the market, with Automated settings, a full manual mode, high quality and high tolerance filament, at an unbeatable value. Adding endless bells and whistles is great - but it's not our priority. And so really, the only reason we need the current feedback is for filament consistency!
So, we need to solve the consistency problems with the stepper. Those of you who were at Maker Faire saw that, even with the stepper, we're producing some pretty good filament. But pretty good isn't good enough when we've promised industry leading tolerances, and so we're trying a few things to solve the problem with steppers. The two approaches we're attempting are to solve the issue mechanically, and in code. To do so mechanically involves changing the feed system slightly so that it operates best at a constant flowrate - not a constant pressure. Current control provides a constant pressure, but steppers provide a constant feed rate - so we may be able to switch the mechanics so that a constant feed rate results in constant pressure. Challenges here are that mechanical development is inherently slower than software, and there's no guaranteed solution. To do so in code effectively involves placing a second diameter measurement sensor right at the nozzle, and watching the slight variances happen in real time to account for them. While this is much easier to test and begin implementation on our test setup, placing a sensor so close to the hot nozzle in a safe, easily manufacture-able way will be...difficult, to say the least. Alex is currently spearheading the mechanical side, while Dennon works on trying to solve it in code. This has meant a little slowdown in the updates, etc as we devote more and more of our time towards solving this issue - so please be patient and bear with us as we try to fix things!
Well, now that that's out of the way, a quick recap on Maker Faire - it was awesome! ProtoCycler worked great all weekend long despite the above challenges, and we even made a full spool of - wait for it - colour changing glow in the dark filament.
We added some blue pigment to the GITD green, so it's now blueish by day and glows green at night! How cool is that???
We also had tons of people come by and check out our grinder. The makers were happy to know how easy it was to power with a drill (which is something we, for safety reasons, definitely cannot officially recommend ;) ) while parents and teachers breathed a huge sigh of relief to see it was hand cranked. And boy oh boy, was the grinder a hit - *everyone* loved it. There's just something so satisfying about grinding up an old failed print!
We also got to see some awesome stuff from ProtoPasta - including metal FDM filament (apologies for the blurry pic):
AIO robotics was also there to share the startup booth with us, and I must say their all in one "Zeus" Printer does have a beautiful design...and quite familiar looking at that ;) ! We both realized we'd come up with our similar enclosure style independently, further proof that great minds think alike.
Backerkit also dropped by to give us some awesome schwag - they've been amazingly helpful in helping us fulfill, so it was nice to finally meet them in person and discuss the "post crowdfunding rush" a little more. P.S. PLEASE FILL OUT YOUR SURVEY IF YOU HAVEN'T YET!
Last but not least, there was some awesome stuff on display from all of our friends in Ontario in the 3D printing space - Mosaic and Structur3d were both there showing off their latest and greatest. It's pretty cool to think that between the three of us, you can print multi coloured recycled plastic of any colour and material, as well as any paste or gel like material as well. We all work with nearly any FDM printer on the market today, and the ability to print with almost anything is just too cool!
When we got home, we got a lovely package from Patio Interactive - they're the manufacturing guys out of / for PROJECT:SPACES. They gave us a couple of kilo's of plastic to recycle, all of varying quality and infill density, which is great for our grinder endurance testing as we're running out of stuff to grind... our printer just can't keep up! If you're in the Toronto area, shoot as an e-mail, and we might be able to recycle all of your waste for free if we can use it for our testing :)
Well, that's it for now - back to the motor troubles. We can't thank you all enough for your support and patience as we sort this out...please let us know if you'd like more information, have any concerns, or just want to swing by and check things out for yourself.
Take care for now,
-The ReDeTec team