| ??? 02/28/08 15:56 Read: times |
#151602 - Being paranoid is a good attitude... Responding to: ???'s previous message |
Brian said:
I'm sure I've been way too cautious and overly paranoid from the start of this project but considering the amount if time and work I have in the little prototype area of my eval. board, I've been trying to play it safe.....thus my wayyy to large a value in the current limiting resistor department. Being a bit paranoid is entirely proper when doing electronics. Brian said:
As for the pull-up resistor I'm using with the sensor, the value of 470 ohms was recommended by the manufacturer of the sensor. Should I go against their recommendation in this case? The question is, in which context they recommend this pull-up. If the sensor is far away from microcontroller, cable capacitance and pull-up resistance form a time constant allowing the output signal only rising rather slowly. In order to achieve fast edges the pull-up is then reduced to rather small values. It's exactly like the old open collector TTL-chips, if you have experience in this field. The disadvantage of a small pull-up is the big current it causes whenever the output transistor is turned on. This can cause a lot of interference. In an ABS project where sensor and micro are far away from each other it might be better to use an additional driver at the sensor and an additional receiver at the microcontroller. The advantage would be a reduce of source impedance and by this a reduce of injected noise (into the cable), and finally a smaller current consumption. So, you could put a Schmitt trigger gate of 74HC14 directly at the sensor, using a pull-up of 4k7, unless the data sheet intervenes. At microcontroller side another 74HC14 gate could act as a receiver. Now you can also place some filtering in front of receiver in order to remove noise or to protect against ESD, etc. Could you please tell us which sensor you use? Then we might give a schematic for your interface. Kai |
