Single-Board Universal Module

I Can't get the sky switch to work. I'm sticking with the peregrine switch which works great. And no, I haven't even thought about the HID option. I'm not sure I'll ever develop it, unless I see real interest. The current issue with the board is the following: The board is supposed to enter programming mode when powered up from USB, but not when powered via 5V from the PPM or Devo connector. That is done via diodes separating the input supply from the 5V reference used by the LDO. The USB supply is hooked to the BOOT0 pin of the MCU with a pull-down resistor attached. So when power is supplied via USB, the BOOT0 pin goes high, and the MCU enters programming mode. When powered via 5V from PPM/Devo connector, the BOOT0 pin should be held low via the pulldown resistor. But it doesn't work reliably. Disconnecting the jumper isolates BOOT0 from the USB supply, and the MCU properly enters execution mode. I just don't yet understand why the circuit isn't working as expected. Possibly there is too much capacitance on the USB supply and the resistor can't dissipate the current quickly enough. I'll put a scope on it and maybe see what is going on.

I've looked at manual P&P machines. I'm actually holding out for an automated one (something like the FirePick) though nothing on the market yet. But that won't help at all here. As I've said before, I am not willing to sell these myself (at least not within the US). The risk is just too high unless I can get it licensed, and I can't see how to justify the cost of licensure. So our best bet is to try to find someone to produce and sell them without my involvement.

Thanks SadSack for that low-budget P&P example. I might do something like that. It looks easy and cheap and I probably have 90% of the stuff I'd need for it around already. It would probably hold me over for quite a while.

need a low voltage-drop over the diode, which is why I used Schottky's in the 1st place. The board needs to operate with a voltage as low as possible. With the Diode and the LDO, that leaves me little margin in the design. The Diode specs list reverse current at 1-5mA. The boot pin must be under 1.6V (probably better under 0.7V) at power-on to ensure it doesn't trigger the programming circuit. using the Shottky is not going to work well here no matter what I do. However, I only need the low Vf drop on the PPM/Devo supply side. On the USB side, I am guaranteed 5V, so I can just replace the USB-side diode with a general-purpose diode with Vf ~1V and <100uA reverse current and probably be good to go.

I think that is overkill. I looked at it with my scope, and I'm only getting 2V on the BOOT pin. The current solution would actually work fine for me with a 1k resistor, but I don't have much confidence in it at all temperatures. For the next iteration, I'll likely use this LDO:
www.digikey.com/product-detail/en/BA33BC...C0FP-E2CT-ND/3663735
with this diode (only for the USB side):
www.digikey.com/product-detail/en/RB160M...B160M-40CT-ND/926504

That should allow a lower input voltage (on the ppm/devo side) and only allow 30uA current through which can easily be handled by a 10k resistor.
I've been working on the Android app and PPM output this weekend. All of the settings are implemented. I just need to get binding coded, and I'll be ready to go.

both cost and board area are critical for this board. I looked at dc/dc converters, but they are a lot more expensive and do not generally suppot as much range in Vin

my current solution will work passed 15V. I don't know if I'll ever need it, but users have been known to use 7.2V batteries which could get passed to the LDO. I'd use a 500mA LDO if it actually gave me a benefit. I haven't seen one that does though. I haven't measured the power yest (really should), but I think I estimated max current at something like 230mA.

That is a good find. I'm not sure I want to add a dollar to the board cost just to save power in a non-typical usage case. In almost all operating modes, I can rely on the input power to be 5 to 6V. Allowing variance in supply voltage is important, but I'm not yet sure I want to optimize for it.

The diodes are there to isolate the 5V USB from the other external 5V supply. That is needed for 2 reasons: 1) allow the oard to go into programming mode when powered by usb but not by normal power without setting any jumper, and 2) allow programming without reverse powering the transmitter via USB. I like this design, and think I can make is work, so I'm panning to keep that aspect. Still Waffling on the LDO vs DC/DC though.

There are no royalties or rights to be had. The design is entirely open source. But I only have limited contacts. Finding someone to sell them will be critical. China seems most promising, but really anyone outside the US will probably be fine. I could do it myself for anyone not within the US, but my market is probably mostly here, and I'll need at least some volume to bring costs down.

I really need to measure input current. If my 230mA number is accurate, I may really want the DC/DC supply. none of the LDOs I've looked at could get up to 12V (which is my target maximum Vin) before hitting their power dissipation limits. To find a part that would do so, I'd end up at the same cost as the TI part. The TPS62172 should be able to run down at around 3.5V and still have good stability and will be able to run at 12V at my estimated current. Even from China, I can't get a great price break, so it will add over a dollar to the board cost. I haven't worked with these buck converter ICs before. How risky is the 2.5MHz switching likely to be? The recommended filter (2.2uH + 22uF) looks like it'll have very low output ripple.

where do you see that? It looks more like 10 to 20mV to me from Fig21. The datasheet already specifies an LC filter. You would add a 2nd filter in series with the spec filter to further reduce noise (if needed)?