Product Idea: USB Flash Drives

[barray]

An interesting fact is that the Rock64 v3 SBC has a bodged component on the underside of its PCB, which looks like a hand-soldered SMD transistor.  That must have been a last-minute fix to the board schematic, and some additional manual labor has been selected instead of fixing the board layout.

Not surprised at all - even the most mature companies can end up doing these last minute bodges. I've seen top-end hardware with sticky foam to overcome some dry solder joints for example!

It's a cryptographic device, and as such it's preferable to be "Made in Europe" instead of "Made in China".  That's why I've opted for the "homemade" approach.

Ah yes, I understand. This may be a rabbit hole, but there is a type of computation being developed by ?IBM? that means you can run code on non-trusted machines, and it's impossible to tell what is being computed. The idea being that you don't even need to trust that the silicon operates correctly in order to be able to use it - might be useful for your project. Checkout homomorphic encryption: https://en.wikipedia.org/wiki/Homomorphic_encryption (To be warned, I think the instruction set is still very young.) Combine this with some proof-of-work and you could be well on your way.

IIRC, JLCPCB provides an option to even source the required components itself, through their usual supply channels.  Some additional design details must be submitted, a few additional rules must be followed, and the BOM must comply to the parts selection available through the supply channels provided by JLCPCB.

It would be a good option to have someone do the entire manufacturing process, once the sales volume reaches a certain point. However, please keep in mind that it would require the testing of assembled devices to be offloaded too, which usually ends up as one big mess, requiring a staff member to be in contact with the manufacturer almost on a daily basis.

With all that in mind, the offloading would be, IMHO, a viable option only when/if the sales volume of the storage device outgrows the "handmade" approach.

Sure, it's just something that could be explored in the future. I really suspect the components we will end up using will be quite common though.

And yeah, testing would most definitely be required. At those sorts of numbers it would even make sense to have somebody in the Country where the parts are manufactured to do such testing, rather than shipping around the world multiple times.

I agree, some money has to be made, for a few reasons.  First, there will be refunds due to lost packages, "it doesn't work" complaints, and who knows what else.  Second, I simply cannot do everything myself, and the other person handling the communication with the customers, shipping of the devices, etc. surely wouldn't be enthustiastic enough to work for free.  Third, I eat quite a lot of food, which certainly isn't free.

Haha, and the price of coffee is likely to go up due to that ship blockage!

IMHO, beating an A1-rated Micro SD card should be good enough for the device in form of a "hat" for one of the BL602 boards.  However, I'd say that a commercially viable device would need to equal or beat an average low-cost USB 3.0 flash drive.

Yep, sounds right to me. I think with security features, people would even be willing to put up with a certain amount of slow-ness, even if we don't meet target speeds. You wouldn't want it as a primary drive, but transferring it from one location to another would still look attractive.

Speaking of the above-mentioned cryptographic device, I actually managed to reach Pine64 around the time before the pandemic started.  Unfortunately, they didn't find the device to be commercially viable and were not interested in manufacturing it.

Well, this is exactly why I want their input.

Also, somewhere on the forum I came across a discussion about having Pine64 to start manufacturing some kind of a device (unfortunately, I cannot recall all of the details).  Basically, the response from Pine64 was something like "bring the device to the commercially viable level on your own, and then we'll talk".

That sounds quite harsh - at that point what is Pine offering? If we do all the R&D and even make the manufacturing a viable process, then we may as well just go down the crowd-sourcing route (even if it is a pain).

Quite frankly, we're probably on our own anyway.

At this point I think that's a sad truth. I have no idea what we would need to do in order to get their attention.