Pete Schaefer wrote:
Frank:

I do have quite a clue about EMI, and clearly have a better clue than you
about systems engineering in general. Here's a question for you:

Why bother?

For the same reason that we build planes instead of buying them.

If you're trying to cert bluetooth for aviation, maybe with the thought of
selling some other bluetooth product that you think you can make a big chunk
of cash with, then, yeah, maybe going off to play with bluetooth on your
airplane makes sense. Personally, I can't see it. Maybe it makes sense for
using it to reprogram boxes on your airplane, but to go to the extent of
making it useful/safe in flight....nah. Way too much effort for too little
return, given that the inclusion of an RS-232 port is so freakin' easy.

I agree totally. except maybe about RS-232. I suggest that some sort of
current-based rather than voltage-based signalling would be more
noise-immune. Bring back 20mA current loop! :-)

If you have some other goal in mind, maybe some other sensing or data fusion
tech (e.g. may you have a huge array of air data sensors for some advance
stall detection method), then you have to look at whether or not the tech
risk buys you something that you can't get otherwise. Why bluetooth rather
than the 1/2 dozen other wired data communication protocols (e.g. 1392, .
422, 232, 485, CAN, etc.) that are out there?

The TWO key ideas in the proposition (wasn't mine, BTW) were wireless
comms and self-powered. If you can do BOTH of those, then wired comms
and power is best. Wireless comms has been pretty much solved. For
example, I saw a projection in an engineering mag that within 25 years
wireless will replace wired as the cheapest comms technology to the home.

But without "self-powered" it is pointless. Are there any devices out
there that can turn (e.g.) vibration into *useful* amounts of electricity?

I get rather frustrated with people who get really !@#$ing enamored with
technologies for implementation and loose sight of what their goals are.

Hmmm... so what are *my* goals? Maybe I want the geekiest plane on the
block? I suggest that one or other of your 4 beers has introduced some
patronisation (as well as abrasiveness) into your system.

All
too often, risk variables get introduced where none is warranted, resulting
in zero or negative value added.

I'll choose what risks are warranted on *my* project, thanks.

I beat on my guys daily about issues like
this. (Kelly Johnson (...yeah, I work at that place.....) had a lot to say
about where it was acceptable to take project risks..too bad so much of it
never got captured in "the rules").

This is fine where *you* get to set the goals.

So, back to Blue Tooth....

Why bother?

*With* self-powered devices, it gives options not available via wired
systems.

For data collection, I've already got a half-dozen options in my hip pocket
that I know will work just fine with very well understood EMI issues that I
know how to mitigate. What's my goal? Blue tooth airplane or getting the
data for some other purpose?

I have very little room for Geek Factor on any airplane that I'll ever
build. If it can't buy it's way on (I'd lump Blue Tooth in here), then !@#$
it.


Pete

P.S. I just got done with a 5 hour drive and am tired as all !@#$. I've got
4 beers in me to diffuse stress. Please forgive my abbrasiveness. I'm not
really that bad of a guy. I just don't want people to pursue ideas that
will get them killed.

Hey, I prefer to talk straight too. And also don't want people to kill
themselves. But if people don't experiment with Experimental category
aircraft, then nothing is going to change.

Frank