We hope that you have all been keeping cool in this particularly long and hot summer! We are back again with the latest installment of the Sa saga!
For those looking for a quick summary rather than the full read, skip ahead to the end of this update.
To begin, a quick recap of where we last left off: We had finished up all of the components of the umbrella, but had run into an issue with our ribs sagging too much under the weight of our fabric. In order to solve this problem we had put out an order for some new prototype rib components made with high-grade ceramic composite materials, called zirconia and alumina, and had been eagerly awaiting to test the results!
In this photo you can see the new alumina and zirconia rib prototypes. Although their appearance looks identical in this photo, they actually have fairly different physical properties. However, the unfortunate news is that, despite our high hopes of success with these materials, both were not able to stand up to the rigorous testing we put them through.
While both materials did perform very well in general, and were indeed extremely strong, the issue came down to brittleness during rapid flexion. Ironically, while the rib can handle an immense amount of stress when applied slowly, rapid flexing, even in less extreme amounts, snapped the ribs. This unfortunately proved to be a deal breaker for us, since a sudden powerful gust of wind, or more likely, the rib getting stepped on or slammed in a car door might have the potential to break it.
So, with even heavier hearts than before, we went back to the drawing board and set out to prototype our pre-bent rib design. Just to refresh everyone’s memory on that design, it is one in which we would have the rib shape curved slightly so that once the fabric is applied it will give a flat appearance. The challenge with such a design was more from a manufacturing standpoint rather than a conceptual one. In any case, it was after our meeting about the ceramic ribs that we finally had some significant changes happen!
As we have mentioned in the past few updates, we had been collaborating with our team of engineers in Korea who would be stepping in from June in full force to assist with the remaining aspects of the Sa and the Sa Compact. We had several discussions about what would make the most sense in terms of moving forward with their cooperation.
Having both the Chinese manufacturer and the Korean team working at the same time could prove challenging for a handful of reasons. Of course, the language barrier and need for a translator makes things difficult, but beyond that, the need for frequent trips, management of the latest physical prototypes and 3D part models, as well as deciding who would oversee the tooling of the molds and assembly, required us to simply choose one team to manage the project, while the other maintained an advisory role.
So with that in mind, we made the decision to have the Korean team completely take over the project and to have the Chinese manufacturer advise. There are a plethora of reasons behind our decision, but it essentially came down to the fact that the Korean team’s superior abilities in engineering would allow them to be able to create an even better Sa from a technical standpoint, and their ability to prototype quickly within Korea promises even faster results!
Now, jumping back into the timeline, as soon as the Korean team fully took the reins on the project in mid-June, we had an important meeting and decided how things would be moving forward. First, and most importantly, we decided to change manufacturing of the umbrella primarily to Korea. We decided to do this because our Korean team has many resources in Korea with whom they can quickly interact with and get reliable results from, whereas working back and forth with China would require more time and effort.
Of course, we may still have our team in China handle some parts, and perhaps take over more of the process in the future, but the first Sa batch out the door will be made in Korea! Making this change will mean that this first batch will cost us a bit more to create, since there are no pre-exisitng umbrella assembly facilities in Korea and a specialized team must manufacture and assemble them. However, creating them in this way will ensure that we can closely oversee the process to ensure a smooth and reliable production run for the first batch of Sa.
We also decided in this meeting to change one aspect of our production strategy in order to speed up our time to market. This step was crucial to shaving off a few months on our ETA. Essentially, up until this point we had been making one-off prototype parts which would then, once finalized, be created with expensive molds capable of producing at least 100,000 units or more for mass production. Such molds take quite a long time to make, often 2-3 months per mold per part.
In order to avoid such a delay, we decided to change our strategy and use a different type of mold for our parts. Such molds are equally as high in quality as the mass production molds we had been considering, but can be produced much quicker and cheaper. The caveat, however, is that they can only be used about 3,000 to 5,000 times before they are no longer usable.
However, the distinct advantage is that such molds can be interchangeably used for prototyping and production! In other words, if we keep our first batch size within the aforementioned unit range, we can directly jump from prototype to production and shave off 2-3 months of time! Since our first batch is going straight to all of you, and that quantity is under that threshold, this strategy makes sense for us!
Beyond that, it also means our next prototype will be made to real spec! Whereas many of the one-off molds are only able to make approximations or substitutes for the real parts, since they often use alternate processes to achieve their results, these new molds will actually give us the real production parts to test with! So, that means that rather than another quasi-functional prototype, coming up next instead we will actually have a real deal, fully functional Sa umbrella, ready to hit the production lines!
So, with a very optimistic path ahead we happily handed over control of the project to our Korean engineering team! We will get into the new projected timeline a bit later in the update, but along with tackling the rib issue they also wanted to completely reassess the current umbrella design to make sure it was up to their standards. This whole process took about a month and a half, but was well worth the time spent. I went to Korea shortly thereafter and we had a week of meetings to go over their findings and to finalize decisions on moving forward.
First, they noticed a handful of parts that they felt were not as strong as they should be in the older prototype design. To solve this, they went back in and strengthened a bunch of components across the entire umbrella. Then next, they came to the conclusion that the magnetic closure system, while being fine in principle, did not provide enough outward magnetic force to function as desired.
Although that news was less than great, it is certainly fantastic that they were able to catch that ahead of time. Likely the Chinese manufacturer simply lacked advanced knowledge when it came to implementing magnetics, since this project would have been their first time working with such technology. In any case, at our last meeting we were able to workshop the magnetic closing tech and we came up with a much better solution!
Instead of a central magnetic switch, we were able to revert back to something closer to the original design, which simply used a loop of magnets embedded in the panels themselves. The change we made, which enabled us to utilize this solution, was actually to relocate the structural component originally along the outermost edge of the panel to a position where it is perpendicular to the rib and meets at the tip of the panel when folded.
If you will remember back quite a while now, we had intended to essentially use a “frame" of structural elements to create the planar design of the Sa panels. If you imagine the umbrella closed, but not yet wrapped shut and thereby leaving a triangular shaped folded panel, these “frame” components would be the rib, then components along the bottom and top sloping edges. However, rather than using two edge parts, we can also generate this same planar design by putting a single component perpendicular to the meeting point of the bottom and top edges and adjacent to the rib. Since this design also allows the component to wrap evenly upon itself, it also allows for us to succeed with our magnetic design!
In any case, it was a huge relief to succeed with that unexpected magnetic issue! However, the best news of all is that we finally succeeded with our rib design! The Korean team came up with a fantastic solution for our dilemma. Basically, using the same principal as the pre-bending design, they were able to compensate for the fabric weight and sagging by creating a three-staged rib.
In this photo you can get a sneak peak of this design. Basically, the rib is split into three parts, all short enough so as not to sag over their own length and made just long enough so that we can manufacture them with a high success rate. A single long rib would require another manufacturing method and/or be prohibitively expensive or unreliable.
These three components are then attached with connectors which are able to fix the angle between each section. By doing this we can compensate for the material weight as the distance from the supporting element becomes greater. So, for example, the first connection point might compensate for 1 degree of angle since it is closer to the support point and experiences less sag, while the second connector might compensate for 3 degrees to account for more displacement. These connectors will appear flush within the material and the rib will give a completely flat appearance when the fabric is applied!
The Path Ahead
So there we have it! The Korean team really did live up to their reputation and all of our lingering problems have been solved! Within our short two months together we were able to solve all of the lingering issues and put together a path towards an even better, faster result! So, with that, we are now moving forward to prototype aka put together the final Sa umbrella by the next update!
Since we have now changed our production strategy to use the prototype molds for our first production batch, we first need to rigorously test and then improve our 3D model to make sure it is completely up to spec for production before sending out for our such new molds. That should only take about 2-3 weeks.
Next, we need to contact our now Korean affiliate manufacturers and arrange for them to make our new prototype molds capable of producing our now smaller initial batch size. Some of the parts however, like the hydraulic cylinder and magnets, we may still be getting from our Chinese affiliates. This whole process should be completed in about a month.
Then finally, we expect in early to mid October, we will have the first real, fully functioning and made to spec Sa completed! As in two months from now, and in our next update! However, the Korean engineering team expects that we will need to make some very small adjustments at that stage. This is very typical, as not everything can be predicted completely accurately via computer model that will matter in the real world. It’s at this point in which we can do some fun wind tunnel tests also! In any case, any changes we make will be small, like strengthening magnets and springs, rounding edges, thickening parts, but all in all, nothing major or very time consuming to adjust. For that reason we expect it will take as little as two weeks and as long as two months depending on what needs to be done.
So, given that, we can expect to begin production as early as November, or as late as the year’s end. However, unlike previously, we do not need to wait 2-3 months for tooling massive molds, but rather can begin production straight away! It takes about three weeks to accumulate and prepare all of the raw material necessary, then it should take about three weeks to coordinate the assembly line process and begin manufacturing for all of the umbrellas. We anticipate we can produce about 100 pieces per day, and we will be shipping umbrellas the second they come off the production line straight to all of you. Given that, the umbrellas should finally be getting sent out to all of you within this upcoming Winter season, ideally closer to the year’s end.
It’s also worth saying that the new Korean team we have been working with has been nothing but professional and has been consistently realistic and punctual with the schedule they have laid out, especially when compared to our Chinese counterparts who were perhaps a bit too eager to please and overly optimistic with their estimates. That being said, I do firmly believe that the time estimates given in this update from our Korean team will hold true and that the Sa will finally make it’s long awaited debut as mentioned. I really want to thank you all again for sticking with us throughout all of this process and for your ongoing patience and resilience! It is much appreciated!
We left off the past update having just sent out for new rib prototypes made with the high-grade ceramic composite materials alumina and zirconia, to see if they would be a suitable solution to overcome the sagging we had experienced with our older prototypes. Although proving very resilient and overcoming our sagging issue, the new prototype ribs were not able to handle the most rigorous of testing, so we went back to developing our pre-bent rib design, which is a design in which the rib is curved in such a way so that the weight of the applied fabric will compensate and leave it with a flat appearance.
From mid-June our very talented Korean engineering team stepped in and took over the majority of development, with our Chinese team continuing on in an advisory role. Beyond that, we have decided to move production for our first batch of Sa to Korea so that our team can closely supervise the process. Thanks to their advice we have also been able to shave off about 2-3 months of production time by switching the type of part molds we will use, going from a larger capacity to a smaller capacity mold, while not sacrificing quality in any way.
With the new team’s help we were able to solve our rib issue once and for all! Based on the pre-bent rib concept, the solution we came up with is a three-segmented rib which uses connectors to offset the angle as the distance is further from the support point. Furthermore, upon a reanalysis of the design, the Korean team made a few adjustments to strengthen the build quality of the umbrella and noticed an issue with the magnetic closure’s strength, which we were able to come up with a solution to.
Moving forward, now that the design is complete, we have about 2-3 weeks of 3D model testing and prepping before we send out for our production-level prototype molds, which will take a month to create. This puts us in October for a fully assembled, functioning, and up to production spec Sa! Although we can begin mass production from that point onwards in theory, our engineers expect that we may need to make small adjustments to refine the design at that time, such as strengthening springs or adjusting wall thicknesses and so on. This should take as little as 2 weeks or up to 2 months depending on what we need to accomplish. Then finally, once we begin manufacturing it will take about 3 weeks to prepare the materials and 3 weeks to organize the assembly line. We anticipate we should be able to produce about 100 units per day once we begin, so we should have the Sa all shipped very quickly thereafter. In other words, the umbrellas should absolutely be shipped by this Winter, ideally around the year’s end. Thank you all again for your patience and endurance throughout this process!