??? 11/15/05 23:07 Read: times |
#103773 - Clarification Responding to: ???'s previous message |
Erik said:
However, re my first post, the OP did the "usual" in attempting to do 485 without reading "10 ways to bulletproof". That should be MANDATORY before even writing the first line of code ot drawing the first line of the schematic if 485 is involved. Whoa there. You've jumped to the wrong conclusion. Before I even considered posting here I did a search and read up on previous 485 threads. I read National AN-1057 and also AN-847 which deals exclusively with the 3 resistor method of failsafe biasing. I thought whis was obvious when I said in my original post: Chris said:
so my research indicates that I should use fail-safe biasing I was always talking about the 3 resistor method of failsafe biasing, not any particular 485 driver IC "special feature". None of these application notes mention what I was considering, which is driving the line in the SPACE condition only, so that is why I posted here. The reason I wish to consider this technique is so that timing constraints are eased, and the MCU does not have to control the transmit enable line. Also, because this effectively forms an AND bus, I get defined results if two devices are trying to "transmit" simultaneously. This then allows a method of slave discovery/enumeration using a form of collision detection. |