I’ve been holding off on this update until I’ve gotten solid confirmation from our manufacturer. Here’s the update everyone has been waiting for. Fitbits will begin shipping from our manufacturer during the week of July 27th to our warehouse in northern California (about an hour drive from our office). We are shipping everything via Fedex Priority Air to get it to our warehouse as quickly as possible. The moment the shipment touches our warehouse, we plan on turning as many as we can around the the same day and shipping them out to you.
Have a lot of you been waiting a long time? Yes and I’m really sorry. However, we’ve worked closely with our manufacturer to come up with this final schedule and they (and we) are confident we can can hit it.
So what is happening between now and then:
Currently, there are 22+ plastic and metal pieces and 100+ electronic components that make up the Fitbit. Here’s renderings/outlines of some of the plastic and metal pieces:
I really hate the metal pieces in the 3rd row / 2nd column 3rd row and the 1st row / 4th column. Those are our charging contacts. If you remember, we had charging reliability problems in our early prototypes and it took forever to get those contacts right.
Each of the plastic parts needs a mold created for it. A mold is essentially a large piece of steel that has a cavity carved into it. The cavity is the shape of the part that you need to make. This piece of steel can weigh 100+ lbs. Designing the mold and cutting the steel is a slow process. It can take 5-8 weeks to cut and polish the mold. Once the mold is created, it’s placed into an injection molding machine. During the injection molding process, melted plastic is injected into the mold at high pressure and the end of the process, the plastic part is formed inside the mold and ejected out.
Here’s a tool /mold. Note the complexity of the mold and all the pins sticking out. Those pins are used to “eject” the plastic part out of the mold.
Actual injection molding machine and video of injection molding process:
For us, mold creation began a while back and the last molds will be finished by mid-June. There are a lot of parts and therefore a lot of molds. Once the molds are created, we do a production run, where the full quantity of parts will be produced. For each run, we occupy an injection molding machine and the manufacturer only has a limited number of them, so our run has be scheduled and interleaved with all of their other customers.
Also during this time, our manufacturer will be assembling our 3 circuit boards (2 go into the wearable Fitbit. 1 goes into the base station/charger). In parallel, we are designing and building test equipment to functionally test each board as it comes off the line. As each board is produced, it is placed into the test equipment where a series of tests are automatically run that measure voltages, currents, frequency, etc, to make sure that each component has been properly attached to the board and that there are no shorts or disconnects anywhere.
A key check-in date to see if we’re on track will be the week of June 29th. That is the week that we expect all parts to be completed and on-hand. For some of the parts, we’ve got our manufacturer to commit to 3 working shifts a day to meet our demand in the short time required. Once we have all parts on-hand, we will begin assembling all the pieces into Fitbits.
We do expect snaufus to occur and hopefully have padded our schedule accordingly. We’ve already have a couple incidents happen a few days ago. The big scare was that our battery supplier delivered our batteries last week, but most of them were about 1mm too long, which prevented them from fitting inside the Fitbit. We did get a few early samples from them, which were all within our requirements, but somehow the rest of the batch that were produced were not up to spec. Here’s an image I sent them detailing the problems.
The supplier is scrambling to redo all the batteries and I actually just got word from them an hour ago that the new batch of batteries will be ready by June 4th, so this won’t cause any delays.
We also had to re-tweak our wireless radio a bit recently to improve our range as we found that later sample units did not have the range as our earlier prototypes. This required us to get some additional parts, which we’re currently sourcing and I don’t think this should cause any delays, either.
This is the end of yet another long post. Please be gentle in the comments and I hope you all hang in there.