Sure, I agree, many considerations. My point was, that making it transparent, the user can see what goes into it, so the business-model kinda revolves around that principle. Open-vendor, Open-biz, Open-tail (retail)
Another argument for “inventing” a trade-mark or model would be to put a self inflicted tax on the final price, which should go back to the source (repo). Maybe just a symbolic 1-3% of the final price. I think most users will totally get and support that?
I think we may have veered into interesting territory that warrants it’s own thread…
I came here to say that it looks like the B-G431B-ESC1 board looks pretty close to what this is all evolving towards. They are $25 CAD from Mouser… it’s one problem is that it doesn’t have SPI and there is not spot for an angle sensor, and it’s hard to reach everything.
I think that if we got rid of the daughter board to reduce cost with minimal downside,
-made the pins easier to access (2.54 mm pitch and ideally a row on either side so it can be used with a solderless breadboard),
-break out some of the general purpose pins, found a good magnetic sensor to match it with,
-put a spot in the middle of the board for the sensor (only good for low current stuff due to magnetic noise, you can leave the area unpopulated and use and external sensor if desired).
-Some mounting holes (2.5 mm is good). The most common pattern for gimbal motors is 19 mm bolt circle, 2.5 mm screws, 4 holes in a square. Sensor has to be in the middle.
That would be a good board to standardize on. Should be possible to make it cheaper than the board at Mouser, without the angle sensor?
What else?
It appears to have CAN even, for whoever wanted that.
Then add up your time. Another 100%. I estimated to really make it worth my while the Mosquito will cost $50.
Again we are coming full circle because no one will sell it. Also it’s using standard G431 which is a beefed up version of G031. What makes you think the problems you see on G031 Lepton won’t also be present on the G431 XYZ yet to be designed / tested / sold board?
Here in EU it’s EUR17.60, but with a maximum order quantity of 2. So not the best thing to base a product on…
I think the B-G431B-ESC1 is a cool design, especially impressive for its level of integration. I think one can learn from it.
But I also feel it’s warping people’s perception a bit of what is possible.
Firstly, it’s actual current capacity is far lower than one would assume from its specs and MOSFETs. While the FETs can take it, the board itself is thermally not able to handle high current. And in addition the board is sensitive to BEMF and spikes, so when you do use it with higher current you can easily break it.
Secondly, while I have no facts to back this up, my assumption is that ST “subsidizes” this board, or at least sells it at no or minimal profit.
If you look at all the components on there, and take into account they’re using quality stuff (not the cheapest components from LCSC), then it’s hard to understand how they can hit this price point even at volume.
I think they’re kind of giving this board away so that people then buy their MCUs and other chips.
But these two things together build an expectation in people that you can make tiny high power boards at low price points… I’m not sure that’s really possible though… at the very least, it isn’t easy, and the B-G431B-ESC1 itself doesn’t deliver on that promise either.
ST sells them at a massive loss, and chalks up the loss as advertisement. This is an 8-layer dual-sided with blind/buried vias, and an STM32F103 daughter board, so try manufacturing this and you will probably look at minimum $100 per board manufacturing only. Just the bare PCB is about $23 per board.
Good luck making anything like that at $20 per board assembled and sold as an end product.
Also, JLC doesn’t have the capability to make blind/buried vias, you need to go to a high end manufacturer, and your cost will probably double.
Runger, I pm’d you asking about the lepton G4, I know this board isn’t very good about telling you when you have a message. Just wondering about things, I am thinking I would like to work on this as a hobby project for a bit, whether I would end up using it or not I’m kind of stoked about finally making this happen as I have envisioned as I believe it has become clear it is the way to go for a lot of the community to advance.
