The new gyros are retrofitted to the current board. I believe that it was Adam Bellchambers idea. Adam and SparkFun designed a single, flat daughter board that plugs right into the current board instead of the three LISY gyros. The entire assembly is a test prototype, complete with a soldered wire for a last minute design change.
We have asked SparkFun if they would be willing to consider upgrading the current board to use the Invensense gyros, as an intermediate step between the current board and "UAV DevBoard3". We have not heard back from them yet.
We have not yet done any RF testing on Invensense gyros.
In my opinion, not offering an upgrade board would be a very unpopular decision. Just looking at the gyros, I don't see anything so difficult about replacing them. It may not be "Solder 101" but it seems reasonably doable.
If a daughter card can be made to have pins that line up with the current gyro pins, then it would seem like they could just continue using the current main board, but with the new daughter board for gyros. That would keep them from having to modify the board, and would allow the rest of us the option of upgrading.
The other day I tried to de-solder a 4 pin AMP connector form my UAV Devboard. I gave up. It's really really hard to suck out all the solder and get all 4 connections to be free and ready to pull out a tthe same time.
Am I missing a good technique ? I don think it's possible to heat all 4 connections at the same time.
I'm not sure what 4 pin connector your referring to, but for something expendable like a connector, just cut the pins, and remove them one at a time. That's what I'd do with the gyros where possible. Solder wick, dipped in flux paste removes the solder pretty easily.
If you're working with something that requires multiple pins to be heated at once, a hot air rework station is the way to go. With solder paste, and a hot air station, even changing the CPU isn't all that hard to do.
Can anyone suggest a fix for this, I was running matrixpilot 2.02 and had to modify the options.file to work with 3 channels, elevator, rudder, throttle. All worked fine, then I edited the options file as I had left ailerons still enabled and now the ESC is telling me there is an abnormal servo input. It works absolutely fine straight from the Rx. I have the channel reversed on the Tx and have tried the software reversal in options.h but I am getting nowhere. My next step was to 'scope the o/p on ch3 from the UDB.
Any suggestions welcomed, the option file is attached
There is a quick way to find out what channel 3 output from the UDB is doing: connect a servo up to it, and power your receiver by plugging your ESC into a spare channel. Check to see how the servo responds to the Tx through the UDB. Then, connect a servo up to channel 3 of your Tx to see what the correct polarity is supposed to be.
Chances are you will find that either there are no pulses coming out on channel 3 (cold solder joint), or the polarity is reversed.
I note in your options.h file that you have the UDB polarity of the throttle hardcoded to 0. I assume you also tried 1?
A couple of other possibilities (long shots):
1. ESC not making good connection to UDB.
2. You have to satisfy both the ESC's and UDB's ESC arming sequence during power up. The way this is normally done is to power up your Tx first, and power up the UDB. Then, if there is an arming sequence for the ESC, you need to do it either as soon as the UDB powers up, or wait until the UDB records the trim settings, which it does at around 10 seconds after power up, indicated by a single "wag".
Let me know what you find out when you connect up a servo to channel 3 of the UDB, we'll then go from there.
Also, there is one more thing you could try that might give us a clue....
You mentioned that everything was ok when you first started out.
Save a copy of your options.h file, and then download a fresh copy of MatrixPilot, and see if your throttle works then. If it does, then we can take a closer look at the implications of the changes that you made to options.h. If that does not work, then we will want to take a closer look at the possibility that your ESC does not make a good connection to the UDB.
By the way, you might want to join the uavdevboard discussion group. There are around 100 members, and there are always several lively discussions going on.