Crossing The Wall
Last month we told you about the two challenges holding us back: The bearing tolerances and the liquefier nozzle. So, let's cut to the chase and answer the big question… have we solved them?
Well, after a good deal of testing, we nailed the bearing tolerances and the whole sub-assembly has since been RTMed. Woohoo!
The nozzle, however, has proven more complicated and is not yet ready. Which sucks, because it’s holding up a half-dozen other components that are otherwise good-to-go. So, what are we going to do about it?
Before we tell you, let’s take a step back and discuss a problem us engineers commonly face. One of the quickest ways to run into trouble when designing a product is to have an “over the wall” design process. This happens when a team of design engineers create a component and/or assembly, then simply toss it “over the wall” to the manufacturing team, saying “go, make this.” Graphically, it looks like this:
We’re not a big company, and we’re certainly not big enough to have a manufacturing team of our own, but we are aware of this common problem. That’s why we've always worked closely with our suppliers, asking them for feedback every step of the way. Sounds good, right? Well, over the past six months we’ve learned a rather difficult lesson.
China is fiercely competitive, and when there’s a good opportunity, manufacturers will go to great lengths to secure your business. That’s great for prices and service, but there’s a serious side-effect from all this competition: manufacturers sometimes say they can make and do things that, in reality, they cannot. While we appreciate their ambition, it’s not helpful during the DFM (Design For Manufacture) process, as it leads us to design components that pose more challenges than they ought to.
Remember the chassis we started extruding in May? That was it, done, right? Ha, nope! It started life as a skin colored, striated, crooked and overall terribly inaccurate approximation of what we wanted, and it took almost six months of testing and adjustment to finally meet our design goals, and we only RTMed it quite recently.
The dimensional tolerances, material properties, color distribution, shade of white, surface finish, etc. are all interrelated, so it’s no surprise it took eight generations to perfect. If you take it to an extrusion company today, they will marvel at the immaculate surface finish and how precisely it’s made – but before the campaign, we were told it wouldn't be too big a challenge.
Numerous other components have followed this trajectory, but through hard work and perseverance, we’ve overcome these challenges and RTMed much of the BOM.
However, the nozzle still remains untamed. The tolerances we require are simply too challenging for mass production with our (previous) manufacturer – nobody warned us back when we designed and prototyped it.
Now, a redesign could be attempted, however it wouldn’t happen overnight. We’ve been working on it, but it takes a week to make a prototype, plus another week to (express) ship it to Canada. At that rate we can barely crank out two iterations per month – assuming we cram all our tests into a weekend. It just isn’t fast enough.
Worse yet, we can’t easily collaborate with the manufacturer when they’re halfway around the world, and not many companies put their engineers directly in contact with ours. Most communication travels through salespeople. Argh!
Throughout Tiko’s development we worked so hard to avoid “the wall” and yet there it was - a rift between us and our manufacturers. Back in May this was an (underestimated) inconvenience, but being November, this was downright unacceptable. We had to do something. Something drastic. Something bold. And we had to do it fast. So we crossed the biggest wall there is. The Great Wall of China.
The Other Side
We would go to the ends of the earth to make Tiko happen, so we did just that. Matt and Sharon, the other two founders of Tiko, booked a last minute flight to join Mike in China.
This selfie was taken in Beijing, in the ShiChaHai district (by HouHai Lake) the morning after we landed. We’re Canadians, and even this was too cold for our taste.
What followed was an incredibly busy, hectic, and downright amazing week of travelling across China and meeting our suppliers.
Down To Business
You can guess which manufacturer we visited first. Naturally, our (new) CNC machining manufacturer – the people producing our challenging yet revolutionary liquefier nozzle.
Above we see Matt and Mike explaining the design goals (with CAD drawings on the phone) to the company president, head technician, and vice president (left to right). During this meeting we collaborated on improving manufacturability while keeping the performance criteria we need.
It was pretty amazing to see their facility and meet their team. The conversations we had here (and at the lunch that followed) will likely prove instrumental in bringing Tiko to market quickly and without compromise.
These gentlemen regularly do contract work for medical equipment companies and the Chinese military, so they know a thing or two about small parts, Titanium, and generally pushing performance envelopes.
In fact, they even sourced a more exotic and better performing alloy of Titanium than before, one which many suppliers simply cannot get their hands on.
Now, bear in mind that all of this happened a little over a week ago, so the results of this collaboration are only starting to show. Thankfully however, we are now physically in China and next-day shipping is commonplace, so our R&D cycle time has essentially been cut in half. Booyah!
So Much More
But this was only the beginning. From there we toured the country, meeting many more of our suppliers. Beijing, Nanjing, Shenzhen, Xiamen, Shanghai, etc. You name it. We visited every place we could. Here are some favorites:
All Wired Up
Ever wondered how wires are made? Well, it all happens in places like these, full of mind-bogglingly complicated spooling equipment and hard working, dedicated people.
And you’re probably imagining that this is some kind of wide, flat factory like the kind cars come out of, right? Wrong!
Many of these factories have multiple floors and are located inside the city districts. Like, right next to the apartments where the workers live, in neighborhoods just like these. It’s fascinating!
Crown of Molding
That was exciting and all, but after eating some spicy frog legs and catching the red-eye flight, we visited the factory where our parts are injection molded. Here’s a photo of the mold used to make Tiko’s filament tray.
In fact, pictured below is the absolutely gargantuan machine used to injection mold our larger components. This thing is so colossal that in person it feels like standing next to a train. It’s just nuts, this thing.
Matter of fact, everything in this machine is so big that it can take several hours to evenly heat both the machine and mold to operating temperatures (else the parts come out warped). And if you need to test a mold, remove and adjust it, then re-test it… then oh boy, you’re looking at an entire day. So refining the molds can easily take a week or more. Don’t even get us started on texturing.
Ultimately, Mike spent so much time there, he got his very own cubicle:
Seeing the factory was very cool, but by meeting the people who run it, we discovered the many subtle difficulties involved in making parts as big yet precise as ours. No wonder it took so long to perfect them, but hey, we wouldn't accept any less.
We saw so much else that week, but at this point, you get the idea. It’s been a pretty mind blowing trip.
Bringing It Home
So what can we learn from this? Well, it’s that, you guessed it… manufacturing is hard. We’re embarrassed to admit it, but it was all too easy to shout “Make it better! Improve the color! Tighten the tolerances! Do it faster!” from the comfort of our office chairs and snazzy café’s back in Canada.
However, coming here and seeing how it’s done, well, let’s just say it’s a humbling experience. Many of our suppliers are self-sufficient, meaning they mix/pelletize the plastics themselves, design the tools, CNC the mold cavities, operate the machines, package the parts, etc. and yet we demand things be done in days not weeks.
You saw the filament tray mold, look at that crazy-complicated thing, it was made completely in-house. It’s hard to blame them for needing over a month to make it, and many weeks to refine it.
In fact, in the case of our chassis, the extrusion company had to design and build their own saw (below) just to cut our chassis to the right tolerance, because there weren’t any suitable off-the-shelf solutions. Now that’s dedication.
In all honestly, after having seen all of this, we’re rather embarrassed about all the pressure we put on our manufacturers to make these parts. Instead, we should have come here sooner and collaborated with them. We could have done it efficiently, rather than through brute force.
But hey, water under the Xihoumen bridge, right. Most of these tools and parts have been finished - and to an exceptionally high standard at that. So, with all of these done, we can focus on the few remaining challenges.
So, What's Next?
Mike has since returned to Canada because of personal commitments, but that’s to be expected after being abroad for half a year. Meanwhile, Matt and Sharon have taken the torch here in China, and are now overseeing manufacturing until every Tiko has shipped.
Aside from being the company’s CEO, Matt is the original inventor of Tiko and our lead engineer, so he will be here to:
A) Receive and test (mostly liquefier) parts almost a week ahead of headquarters, and physically go to manufacturers to tie up any loose ends before mass production. Full testing will still happen at HQ, but having validation several days ahead of time will no doubt help us get back on track.
B) Oversee assembly and QC. Nobody knows Tiko better than Matt, so he will be at the assembly facility standing over shoulders and making sure every Tiko leaves the factory up to his standard.
Meanwhile, Sharon is our CFO, and boy has it been a challenging job. International banking can be tricky, and numerous parts have been delayed due to banking problems. Sending wires is tedious enough. Sending them to China is a pain in the you-know-what. However, now that she’s here, the accessibility to and communication with our suppliers has greatly improved, and we are now able to process these transactions in hours instead of days.
What's Happening Back at HQ?
Mike is now in Canada and overseeing operations at HQ. Much of his work still revolves around organizing the supply chain, however he’s also guiding the ongoing R&D work that we do day in and day out. And hey, sometimes we actually print some stuff too :)
(Unfortunately, not in PLA, though)
Aside from the sheer fun of it, we use running Tiko's to confirm tolerances, optimize/debug software and firmware, improve the slicing engines parameters, etc.
Ultimately, things are busy on both sides of the Pacific, and every new day we're that much closer to the big one!
What Does All of This Mean for Ship Dates?
Technically speaking, from the day we RTM the last liquefier-related component, it should take 40-60 days until we start shipping out Tiko’s. So we need to get moving quick, or we’ll run into Chinese New Year come mid-February. If you don’t already know, pretty much everything in China shuts down for the majority of February.
It’s hard to say for sure how quickly we can resolve the liquefier problem. It could be a week or a month, or more. We can’t predict the future on this one, however, with the way things have been going, we’re cautiously optimistic that we’ll be shipping before the Chinese New Year.
Some of you have asked us if we’re increasing our production rate to compensate for the delay, and the short answer is yes. We’ve been doing some pretty clever things to work around the liquefier. However, we’ll cover these in detail in the next update, because you’re going to want to see pictures of this stuff and we just don’t have them yet.
In the meantime, Matt and Sharon will be working closely with suppliers to move everything along as quickly as possible, while doing it at the quality level that you’ve come to expect of us.
Back to Work
That’s all for this update. Although the majority of Tiko’s components have been RTMed, there’s still plenty of work left to do on the liquefier, assembly process, and QC program – all necessary for these Tiko’s to end up in your homes. Sure, we don’t have a 100% locked in date for when they’ll ship, but we just want you to know that we’re here doing everything in our power, and we’re on top of our game.
That said, we used to throw designs and demands over the wall to our suppliers, but we’re a little wiser now. Likewise, we hope that you can join us (in spirit) on this side of the wall, and that we can move forward as one big team. The challenges are immense, but together, we can accomplish anything. We love you guys!
Yi qie shun li! (All the best!)
P.P.S. While in China, we’ve seen some pretty amazing things. Not tourist attractions, but the real China from the local perspective. Much of it doesn’t pertain to production, but it’s awesome in its own right. We think everyone should come see China/Asia at some point - it’s just so different here! If you like, we can do a post about it somewhere to give you a taste. What do you think?