>1)Running a non-DCC engine on a section of DCC track
>is prone to causing overheating in the engine.

Well, I cannot see what it is you are attempting to achieve if you are 
running an analogue loco with feedback. How is the feedback circuit going to 
be powered when the loco is stopped?

And only a handful of DCC manufacturers have implemented the analogue loco 
operation under DCC, as it is an optional feature. Any overheating in the 
motor is dependant on what sort of motor it is, - and you certainly would 
not want to be using a coreless motor in this mode.

>2)There is no practical method for an engine to send
>data back to the controller using DCC.

Yes there is, the NMRA does have a proposal for doing exactly this.

>3)Where's the fun in that?

Well, I could say the same about your apparent attempt at re-inventing the 
wheel. There are more useful things to do in life, by building on the 
already existing facilities.

However another method you may wish to think about, if still interested in 
pursuing this, is the chip that is used in the Marklin system for their mfx 
locomotives. The data stream is sent back as some form of higher frequency 
carrier, and then in the control unit they demodulate it using a standard 
RDS receiver chip. This then attaches to an SPI port on the micro in the 
control unit, with the RDS chip recovering the clock and data and applying 
both to the SPI port, which is therefore acting as a slave device. I do not 
have any information on the RDS stream format, but based on this I assume it 
is some form of NRZ that allows the clock recovery to be done.

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist