Today our destination was a motor manufacturer called Lotus. This manufacturer is used often by AQS and so they were able to set us up with a very hands on tour. As opposed to the Huawei and CTS tours, pictures were allowed and we were able to get deep into the process and parts. One of the things I've taken a liking to is working with the small and medium sized manufacturers rather than the larger ones. At a smaller size, customized designs are much easier and the processes are flexible enough that changes can be made sometimes on the fly. The larger manufacturers are much more rigid in their processes and methodology. Although they're more polished, it usually comes at a price so it's difficult to make changes or do small batch test runs.
Lotus is a medium sized motor manufacturer and the facility we visited manufactures DC motors. They have a separate facility to manufacture stepper motors as well, but we were not able to go to that factory. Once we arrived at the factory, they took us to the sample room to talk a little bit about the types of motors they manufacture. After this, we went to a small room where they put together customized sample motors. These are custom DC motors that are requested by their customers and the initial sample quantities are handmade.
For each motor, its possible to modify the torque, speed, power, and shaft diameter. The motors are mostly modular so these are easy parameters to change. The only thing they won't do is modifying the form factor of the motor. There are about 100 different standard sizes for the DC motors so it doesn't seem like this is much of a limitation though.
They also do geared motors with custom gear reduction ratios which seems extremely useful. I asked about the minimum order quantity and the guide said that it'd probably be in the area of 1000 pieces. Considering that many of these motors are on the order of less than one to a few dollars, this seems extremely reasonable for a custom motor, especially a geared motor.
The gears in my head are also now spinning...
yuk yuk yuk.
After seeing the sample making room, we were taken to the main factory floor. The first stop was a test station where they test and balance finished motors. The test apparatus consists of a stroboscope to check the stability of the motor speed and a vibration sensor to check how much the motor vibrates as it spins. Based on these two parameters, a technician will add small amounts of weighted paste to the motor to balance the motor. Once it falls within spec, then he will put it into a finished pile and then start on the next motor. I don't think all motors get this kind of treatment, but it looks like the motors he was testing was for an immersion mixer and its likely this was a requirement by the manufacturer to minimize the amount of vibration in the end user's hand.
Next to the balancing test stations, they had a hand assembly line. These look like they're used to assemble the small parts in a motor like the end caps and the contacts. There are a lot of small pieces that are probably not very amenable to automation or perhaps have not been automated yet. The workers are in an assembly line to assemble all the small parts into a larger part that can be passed to the part of the factory floor where they have automated machines to assemble the motors.
The next area we were taken to is the factory floor where they do the main assembly of the motors. We were able to see the full process of how a motor is manufactured. The process is straightforward, but requires quite a bit of automation. For the rotor, they start with a bare rotor with no windings. The rotor and armature are modular and the rotor core is composed of stacked layers of a ferrite core. The modularity lets them adjust the length of the armature by adding or removing ferrite core layers to the shaft. The rotor is then placed inside a fixture that automatically winds wire around the armature. The thickness of the wire can be specified by the customer and this will affect the torque and speed of the motor.
They also have automated machines that assemble the motor enclosure. The enclosure consists of the motor casing, stator magnets, and various small components to lock everything in place. They had one particular machine that we all found fascinating. It was an automated stator assembly machine, but rather than permanent magnets, they used ferrite rubber, similar to the material you see on the back of refrigerator magnets. The ferrite rubber is originally non magnetic. The machine cuts the rubber, inserts it into a stator case, and then puts it between two huge coils with 500V passed between them. The electromagnet magnetizes the rubber and it becomes a permanent magnet. We were talking about how to get the material, laser cut custom shapes, and magnetize them for different applications. This is probably one of the great things of being able to go deep into the processes at these factories. When you see the processes, you can also see ways to hack the processes to achieve different things.
Another interesting thing is that the tour guide mentioned that the Little Bits motors are manufactured there. Jie used to intern at Little Bits so the factory tour guide brought some Little Bits sample motors and we took a picture of Jie with the Little Bits motors. Maker Shed also has some motors manufactured there for some designs.
This concluded our tour of the motor factory and it was an excellent tour that let us get right into the whole manufacturing process. We then had some talks about the process to have a custom motor made and he mentioned that we could contact the factory directly or talk to AQS and they would set everything up.
Around the factory there were also several small shops that were set up. We checked them out and they had some shops making connector terminals, silicone molding, speaker grills, and an injection molding tool shop. AQS was saying that its important to work with the right people in China otherwise your injection molding tooling might be handled at a small shop like the one we saw. That shop was a bit sketchy and seems to not have the proper equipment to get the tolerances that are normally needed for injection molding, at least at the factories that we've seen. On the other hand, in some cases, it might make sense to farm it out to a small shop like that to save cost. As always, there's never a correct answer.
The day didn't end there though. Since we were with AQS, they asked if it was possible for us all to give a small talk about what we're working on to the other AQS employees. Since AQS has been extremely helpful this whole trip and helped set up many of the factory tours, this was the least we could do. There's an additional benefit because a lot of the MIT Media Lab grad students were asking about what Bunnie, me, and David do. This would be a good chance to talk a little bit about ourselves.
As usual, the Media Lab designer's projects were awesome and generated a lot of interest from the engineers at AQS. David Craynor also talked about how he co-founded FormLabs and him and his partners came up with the idea for the a cheap, high resolution 3D printer. Bunnie was talking about his past life at Chumby as well as all of his current projects including some DARPA work and why he started working on his ARM laptop (which is awesome by the way). I talked about the UN project I'm working on, some of the stuff I'm doing with Wrecking Crew Orchestra, the Safecast work I did, and some various other projects that I'm playing with. I guess the Wrecking Crew Orchestra EL wire dance stuff was the most appealing part of my talk and the Media Lab students and the AQS people seemed to like it the best.
The night was capped off by going out to a huge dinner with the AQS staff. AQS paid for the meal and it was mainly to build tighter bonds between AQS and the participants on the trip. There was a lot of drinking and socializing involved, not only at this meal, but on the trip in general. In my opinion, I think this is one of the most important lessons for the designers on this trip. To do business in China, and Asia in general, there are alot of soft and social skills involved. These are all needed to build trust between the parties doing business together and its something that can't be learned in a lab or academic environment. On the flip side, I told many of the students that they didn't have to drink or do a bottom's up (ganbei) if they didn't want to. The liquor was standard chinese white liquor (bai-jo) at 53% alcohol, so its a bit dangerous if you're not used to it. Personally, I don't really like hard liquor. Anyways, in my opinion, it's good enough just to see what business social graces in China are like.
I also did have an interesting conversation with the owner of AQS and mentioned that everyone thinks that labor in China is cheap and that's why it can be so successful. The labor charges are not really that cheap due to hidden costs though. There's bus transport to and from the factory, dormitories, meals (lunch and dinner are always provided), and other costs. Also, it's getting harder to get employees to come work in the factories so many factories are raising wages to attract workers. There's actually a huge labor shortage in Shenzhen right now. This is goes counterintuitive to a lot of the stories that circulate outside of China and I was a little bit pleased to hear that Chinese laborers are successful enough to start getting a sense of entitlement. But from what I've seen on this trip so far, the real success factor in China is the huge eco system and network of markets, electronic components, suppliers, manufacturers, and logistics facilities located in such close proximity to each other. This proximity makes manufacturing in China go much faster than other places without this kind of environment.