Share this project

Done

Share this project

Done
Photo original
A 0.80mm thin flexible wristwatch with an E Ink display housed in a single piece of stainless steel.
7,658 backers pledged $1,026,292 to help bring this project to life.

Recent updates

Photos

D316c836a99e2208b8dfb8915cb22a1b original
818f2fb868be1d4319468757260abc54 original
Effc51446053f07b220c9df39f2bd249 original
6659a40733a3d455a8b10eef515fd519 original
7be4f4d1860c69da5415931a15c9efbe original
39095372f0c7641762a47fb8c1761d23 original
12d862ea8f5e2fcd0033897181afd9ff original
C83efc3d024a3083ede572ead3caed60 original
Bdd296939af41a2407c86e409ef57cb1 original
71b614755da302ed33484f8fe180f2c3 original
9648ef0eb0f90ccbbbadde5d24ec2d48 original
6d3243f78281c839c2f31b9d09b80a94 original

Finances Update & Where We Go From Here

67 likes

Hi again, we wanted to add this Financial Summary into last update, but it took about six hours to distill all the technical details into something that could be digestible, and it took the better part of today to get all these things we’re about to share organized in a way that’s easy to follow. 

This update will show where our money has been spent to date, over the last two years. As many of you are asking about refunds, we wanted to open the books so you can see where the money you’ve contributed has been applied. 

The money has mostly gone into development, tooling, and purchasing the parts that go into making the watch and charger. We have included detailed pie charts below showing money in and money out. 

We are also publishing this information to show where we have not spent money. We realized immediately after the Kickstarter campaign ended that we had to run a tight ship and put every penny toward parts, development, and tooling, so we went to extreme lengths to keep our other expenses as low as possible.

Finances | The Details 

Money raised: 

The chart below shows all the money we raised. There’s a large chunk of money from all the Kickstarter backers and some additional funding post-campaign via pre-orders. You’ll also see $100k of additional outside funding we raised.

Money Used:

The money spent was separated into into two groups. The first group, money from Kickstarter and pre-orders, was used exclusively for tooling, parts, and development costs of the CST-01. The second group is the additional investment, which we used for our operational expenses. We split our expenses up this way to ensure that your contribution was unequivocally used on the specific manufacturing costs of the CST-01. 

The first chunk of money went to Kickstarter and Amazon fees. This was for hosting the campaign and processing payments. This was roughly $100k. The money we received from the campaign was $923,634. 

The Validation, R&D, Shipping & Start Up Costs bucket covers things like creating programming and test fixtures, experimentation of adhesives and materials, test tools for proving out the encapsulation process, sample batteries for evaluation, sample metal bands, and evaluation circuit boards. 

The largest slice was spent on parts for the watch and the associated tooling. Big-ticket items are the E-Ink display, batteries, battery-management ICs, microcontrollers and clock crystals. Tooling includes injection mold tools for the plastics, dies to form metal bands, diecut tools for trimming circuits, ultrasonic weld tools, and a custom machine for applying adhesives and encapsulants. Here is a break-down of all parts, tooling, programming and certification paid for to-date:

Running Lean, Keeping Operating Expenses to a Minimum: 

Our operating expenses came out of the additional 100k investment. We’ve broken this out below. 

Certain numbers stand out, for instance, you’ll see that “Compensation” is roughly around $18,000 for two years. This number reflects the net salaries paid to both of us over the two years we’ve been working on CST. This number is low because we not only invested a lot of our own savings into the company, but we paid ourselves the minimum we needed to scrape by month-to-month once the remainder of our personal savings was exhausted. 

Some things to note are what we’re NOT spending money on. When we could do things ourselves, we did.  This includes all of the CAD & specification drawings for the watch, the electrical engineering, a large chunk of the programming, the electrical testing and calculations, making our own programming and testing fixtures, procurement of parts, and identification of outside vendors (for steps that our manufacturer didn't have the capability for). We do not have an office (we worked from home or on-site at our manufacturer). We did not buy computers or software. We did not hire a staff. That has certainly helped as paying for even a small office seems like an unnecessary expense.

We both live in Chicago. When we have to travel to San Jose for CST, we organize our schedules so that we can find the cheapest airfare and lodging possible. We found that hotels are too expensive to stay at when we have to be in San Jose for weeks at a time (even the cheapest hotels can be over $100 a night per room). So, AirBnB always comes out as the cheapest option. We sort options by price and select the cheapest possible accommodations. 

Most recently, Dave has been spending almost every other week on-site. Because money is tight, Dave took the cheapest possible option for accommodations to the extreme. He rented a van from AirBNB with a futon in the back that he parks in the manufacturer’s parking lot. It’s $35 a day, so he gets transportation and somewhere to sleep. Again, since we’re mostly working when we’re on-site, these sparse accommodations don’t really matter too much, and it has allowed us to be at the manufacturer’s site more often in recent months.

What This Means for Delivery of Watches 

As you can see from the numbers here and in the last update, continuing production at this time is a challenge. We have the majority of the parts needed, but are struggling with the yield and the costs involved to assemble them into final units.  At the moment, we don’t have the reserve funds to make up for the yield and labor costs. We are currently investigating options that may allow us to get units out to the backers and continue to produce, but the details of these conversations are still in the early stages. We will be shipping as many watches as possible and are working with the manufacturer to see how many watches can be shipped.  

We hope that this update helps to communicate that we have been doing everything in our power to get through these issues and get these products out to you folks. We’re deeply sorry for what looks like may be a negative outcome and it pains us to write this post. We assure you that we will continue to look for ways we can get watches out to you and we will ship every last watch we are able to. 

-Dave & Jerry

Production Update

83 likes

Production numbers and speed of production are much lower than estimated, initial production was initially estimated at 1,000 - 2,000 units per week from our manufacturer. The production line is far from running at full efficiency as we encounter issues and have to install additional steps for testing. That estimate is currently at 200 units per week. As we work on the line training operators, actual production including yield was only 130 over the past 2 weeks. 

We kept the first batch small in order to see how the watches did in the field. Unfortunately, we had a higher percentage of failures and returns than we were comfortable with. In order to prevent shipping more unreliable products, we went back through the steps of the production process to find where those errors were and added more testing to ensure that we ship robust units. This process added more delays to our goal of getting units out the door. We continue to work on our process in order to get our yield and costs into a range that will enable us to ship to our backers with the amount of parts we have on hand. 

This is very disappointing to us and we know incredibly frustrating for all of you. Below is an explanation of what’s happening to date as we continue to make all efforts to get these numbers into a more respectable range. A brief summary shows where we are getting an incredibly low percentage of shippable product. We are getting failures of parts and then additional failures as we go through assembly steps. This all adds up to 44% of all the products not being ‘shippable’ as they fail functional or cosmetic tests at various points in the assembly. 

The details on production and yield challenges. 

E-Ink circuits:

We are getting 80% yield on these parts out of the box. We have tried many approaches to rework the failed circuits. None of the rework approaches have yet managed to get the failed circuits to work. 

The reason that 20% of circuits are failing is that too much pressure was applied to the battery management chip during assembly, This effectively crushed and destroyed the chip. The units were not tested to our electrical specification, and as a result, we were shipped boards with 20% failed parts, which not only adds 25% to the per-part-cost, but also adds cost in testing to weed out the failed units. 

As these parts are one of the most expensive parts of the watch, we have been investigating ways to avoid eating this 25% cost increase. We have successfully tested and documented a process for removing this part but have been unsuccessful in finding a vendor willing and capable of removing the parts and attaching replacement parts. The battery management chip is a custom bare die part cut from a thinned silicon wafer (In short...It’s tiny and a real challenge to place and attach) We are working on other ways to regain that 20% loss, but we may have to realistically live with the loss. 

The Batteries - Attaching and Encapsulation. 

The battery supplier has been incredibly proactive in attempting to fix issues and sent an expert from France to San Jose to thoroughly test and document the challenges with these batteries. Tests covered thermal, mechanical and visual inspection guidelines. 

Visual inspection show that some batteries have visual defects (bubbles) that indicate that they may not be electrically sound in the long term. Pull and stress tests were conducted on the battery tabs to see where these tabs would fail or fall off. The results of this in depth investigation mean we now know we have an 80% yield (20% of the batteries fail after we put them through tests and inspection) This yield issue effectively makes the cost of the battery per product also 25% higher. 

Compounding this issue is that even when the batteries get past the inspection, these ‘good’ batteries go through the production and assembly are attached to the E-ink circuits, These circuits are then encapsulated in a flexible resin. Additional bending and stress tests are conducted after this step. 10% of these assemblies fail those tests. The result is that those entire failed assemblies become scrap.  This points either refinement needed on battery protection or a more stringent battery tab pull test required.

We found that failures in the shipped units may have been due to a battery attachment problem. We tracked down that there were a batch of units where the incorrect solder was used on the line causing unconnected cold solder joints on the battery. (The solder had the wrong melting temperature and could not "wet" fully to attach the batteries). We have added confirmation steps to ensure the correct solder is used moving forward. 

Assembling to bands 

The final few steps are to attach the encapsulated circuits to the bands and then bend test the final product. We get an additional 77% yield here (23% of the final products fail at final testing.) The primary point of failure at this point is delamination of either the battery tabs from the battery or the battery tabs from the circuit board. Once this happens, the only salvageable part of the assembly is the metal band.

What does all the above mean? 

We get a cumulative yield of 55%-56%. with all the testing that is necessary this is very time consuming and requires real human attention at every step of the process. In cold cash terms it balloons the parts and labor cost to around $300. The low yield effectively doubles the production costs. This is obviously something we need to improve in order to be able to deliver rewards to everyone.  If we kept the line running with the yield we are seeing, we would not be have enough parts to send to all the backers.

We understand your frustration and hope that by pulling back the curtain to talk about the problems we have encountered, we will be able to communicate what is keeping us from getting these watches out currently.  We will continue to be as transparent as possible in future updates.

- Dave & Jerry | CST

We are shipping

194 likes

Hello Backers,

Last week, we worked with our manufacturer to finish building, testing, and shipping our first batch of products!

We'd like to share a few photos of this final process:

Final Testing:

Ready for packing:

 Scanning base station serial numbers:

Boxing up:

 Ready for delivery:

We are delighted with how the product has turned out and are excited to be getting them onto your hands.

At this stage we are being careful to start slow to make sure our process has all of the kinks worked out of it.  For those of you who are curious how we are going about the shipping queue, we are shipping by both order number and the SKU that we are currently producing.  We have started with the black, raw titanium, medium watches for this shipment, but will be changing the line over to cover the different color and sizes in subsequent orders.

Now would be a good time to make sure your address is correct in the Kickstarter system.  If you are unfamiliar with how to update your address, we wrote a brief tutorial here:

https://www.kickstarter.com/projects/1655017763/cst-01-the-worlds-thinnest-watch/posts/752244

Thanks again for the patience and and support.

-Jerry & Dave | CST

Progress Update

91 likes

 Hello Backers,

We are back from another couple of trips out to the manufacturer and have some progress to share with you.  

New Steel Tools and Cavities

We found that some of the yield issues we brought up in the last update were due to a tool in our molding machine flexing.  We modified the molding process and made a new tool from steel that does not flex, so all of the units show much more consistency than previously.

The photos at the top of this post are from encapsulation tests we have been running.  We are currently addressing some issues we are seeing with the encapsulant on the small size units and running bending stress testing to make sure it does not have any issues before we ship the product.  These tests are currently in progress.

The Batteries

We have a documented test procedure, and have tested a batch of batteries with this procedure to make sure the batteries we use are good and mechanically sound.  Dave personally inspected the batteries that pass this test and found a small subset of the batteries seem questionable.  We are waiting to hear back from the battery manufacturer for their recommendation on how to resolve the issue.

The Bands

The issues with the bands look like they are becoming resolved.  We have received a batch of bands from the manufacturer and while we currently need to test every unit to determine whether it meets spec, we are seeing that we are getting a good yield of bands and should be able to get to 100% in spec in time.

Moving Forward

The last two hurdles are making sure the units pass bend testing and resolving the battery issues. Once we are past this, our plan is to start getting these out the door to you.  

Once again, we were delaying the update until we could say "we started shipping!" since it looked like it would be just a week or two out for the last few weeks.  We apologize that our update cadence is slower than many of you would prefer, so we will be revising our approach to keep you informed irregardless of whether the milestone we anticipated was met.  We will be back visiting the manufacturer again next week and we will post a brief update after that trip as well.

Thank you all again for your patience and support.  

-Dave, Jerry & the CST Team

Progress Update

130 likes

The first samples made with all final manufacturing processes:

We just returned from another 2 week trip to California to work closely with our manufacturing partner.  Above are the first units produced with 100% of the parts and processes that will be used in final manufacturing and they look great!  The issues that remain are the bands, batteries and yield targets which we'll dive into below.

The Batteries

As we started to assemble our test units, we found that the batteries have some issues where some units have tabs that aren’t well adhered. The good news is that the battery supplier is extremely proactive in fixing the issue and we are working closely with them to document a testing procedure to weed out and replace problematic units. 

Close-up showing a tab not well adhered

Dialing in the Process and Increasing Yield

Throughout this journey, our bill of materials has increased. From issues with E Ink displays, the original battery partner closing shop, and the move to Nitinol bands has left us with very little room for scrapping parts that don’t come off the line perfectly. With sample in-spec bands, we were able to assemble 7 working units. However, this was from a batch of 16 circuits. The main reason for the low yield is due to the aforementioned battery tab issue. We are addressing the issues in order to get the yield back to a reasonable number before we can pull the trigger on high volume manufacturing. 

The Bands

We’ve been working for weeks to get a schedule from the supplier for when we will be receiving bands in California. Delivery dates and quantities were in a state of flux, so we had been waiting to update until we had parts in-hand. Unfortunately, as of right now, the bands have still not arrived. 

So, what’s been happening?  At the point we approved the first article parts we were told that 200 parts a day was a safe estimate of what would be produced.  It took us several weeks to track down why we were not receiving parts as promised. It appears as though the supplier was having some challenges meeting that estimate due to machining tolerance creating a lot of ‘scrap’ parts that were outside specification. It turned out that the bands were not being measured according to our instructions, so many “scrap” parts were actually good, we also looked at their measurements and found ways to loosen tolerances without impacting the design in order to help them reach yield targets. 

As of yesterday 200 in-spec bands were due on the loading dock in California, but we just received a note that the 200 units that were supposed to ship were not shipped since the vendor had missed a process step (cleaning) and the parts were out of spec, so they were returned to the vendor. 

Our manufacturing partner has installed daily updates and check ins and have also mobilized a team to be at the band manufacturing site to oversee production in order to resolve the issues in getting these made. We will continue to update you as we get more information about the shipment. 

The cautious approach and the careful delivery schedule from the supplier is likely due to the following reasons: 

-The raw material costs for Nitinol sheet is incredibly high, running ‘failed parts’ and rushing them through the line is a costly error, running them slowly will get the supplier a better yield and less scrap parts...ultimately they will get more ‘good’ parts that pass inspection that we have paid for. 

-In ‘mass manufacturing’ it would be normal to run 2,000 or so parts through the line to check and measure them to make sure consistency was reached, in our world that would be around $50,000-$100,000 in parts to ‘check’ the system..or roughly 10% of all backing Kickstarter backing money to ‘check’ the system. 

-Internal systems are in place that check FAI (first article parts) and lead to PPAP (Production part approval process) these processes are internal to the supplier and it has taken sometime for those internal processes to be opened up to us in order for us to successfully collaborate on arriving at a solution. This supplier has invested a lot of time and effort so far in getting these processes fine tuned. 

-Chinese “Golden Week” was the 1st of October to the 7th of October, this means the lines ramped down and came back up upon return.

We have been hand assembling circuits to bands and continue to refine and prepare all the other steps of the process for the arrival of these bands. Fingers crossed we will be receiving these bands at 1,000 to 2,000 units per week, meaning about a month of production time.

Moving Forward 

Having two people at CST handle the design, packaging, engineering, CAD, process development and documentation for a product has been quite a handful. In some ways this has been good for us allowing us to weather these numerous delays by staying small, but it also means that sometimes communication suffers and we apologize for this. The manufacturing partner is now providing more support to get us across the finish line and dial in these details to increase yield. We are systematically addressing these last few issues around yield and the bands. We understand that you are anxious to get the watches and we cannot wait to get these out to you too!

- Dave, Jerry and the CST team