- March 9th, 2015
- Write comment
I have made significant progress on the camera front.
The first camera, the Seek Thermal is available commercially, but only shipping to the US. So I have ordered a sample camera using a reshipper, I got notification left the reshipper on Friday so looking at a delivery early next week.
The second camera I am interested in is the Panasonic Grid-Eye 8x8. I was hoping to buy a USB evaluation board, which while being not quite the sensor I was after would have been close enough to decide on the product. Unfortunately Digikey won’t ship that product to Aus either, ITAR concerns I assume.
With Digikey unwilling to supply the chips internationally using a reshipper for the supply wasn’t really satisfactory so I contacted Panasonic about getting another supplier. I had made some prior attempts to contact them without getting anywhere so I went full court press with at least four simultaneous inquiry attempts. Calling the telephone number of Panasonic Industrial Devices Sales Asia (PIDS) in Singapore actually worked. I was passed to Ivan Khor who has been incredibly helpful, he reserved ten chips out of their next production run and passed me on to Braemac to handle the local distribution.
So the Panasonic chips are a bit more fiddly and will take more time but the solid link into Panasonic will probably make the effort well worthwhile. I also learnt that the evaluation board is actually a Digikey product, so I will have to knock together my own I2C interface.
A key part of what I want to do is replace the standard microwave interface with a touch screen, essentially mount a phone as the control interface. The big gain here is avoiding the limitations of a four character display, I think this is why so many of the advanced functions are so difficult.
Basically I want the following features:
- ARM processor capable of running Android
- 3.5″-5″ touch screen
- 1x USB host-mode
- 1x USB slave-mode
- Some GPIOs or additional USB
I think I could design a board to do this myself, it is essentially a standard ARM application note. However that is work, prototyping it is work, testing it is work and it being fully custom would be expensive. So the lazy me scheduled in the grand plan a week of trying to buy something else that someone else had done, then doing the design if that didn’t work.
My initial thoughts were to get the guts of a cheap tablet and maybe tweak the BOM slightly. I had heard that the Shenzhen pattern was that most companies were very specialised, different companies made the screen, the PCB, the chassis, the cardboard box, etc. So I went hunting on Alibaba for the PCB element of that chain. Unfortunately I misunderstood the way it came together. As far as I can tell the a company that builds the product does the design of the product selects the parts such as the screens, chassis etc. that they buy. There is also a range of standard pick and place manufacturers that assemble PCBs according to the design they are given. The first isn’t interested in just selling PCBs, the second doesn’t retain designs they build to order.
Fortunately I think I have stumbled on a solution though. The Banana Pi. Essentially a product which was inspired by the Raspberry Pi but uses a different, more powerful, ARM chip with a better feature match to what I am after, such as built in flash. It is also, like the Raspberry, open hardware and offered by a range of Alibaba manufacturers.
I will get quotes to modify the Banana Pi boards, just change the BOM to remove a few expensive parts I don’t need. I expect though that it will cheaper to buy the quantities I am after unadjusted as a fraction of a larger standard build.
Another interesting thing which happened this week is that Mark Rober came out with a marketing video proposing to do something similar to what I am looking at though focusing primarily on the thermal rather than usability angle.
It is actually reassuring to see that other people think this is a good idea and his website counting the number of people who are vaguely interested is also useful market research.
I’m not particularly worried about the competition. From the looks of his prototype and videos he is well away from a realistic design, let along product. Far more concerning to me is what the major manufacturers are doing in private and the risk that this may kick those research projects up into a higher gear.