We're looking at 1ms resolution. How does this sound: The timing is basically the following: Each car just passes the beacon and announces that it has finished a lap via RF to a beacon. That beacon notifies the host PC. The host PC keeps track of lap times for all cars. Since the time taken to announce the passing of the beacon and notify the host PC will be constant for each car, the timing should be accurate. I have a timing class for the PC that is capable of timing easily to 6 decimal places so no problems with 1ms resolution there. Another benefit of this approach over the IR method will be that the micro can be in low power or sleep mode for most of the lap, and wake up / transmit when the beacon is passed. Good for battery life for the cars that have their interiors stripped and have no cigarette lighters for power supply... Since there are a lot of different types of cars competing, the on-car hardware may not be the same height for each car, so I would need a tall, narrow beam of IR to be sure of triggering each car. Another advantage of using RF over IR would be that the in car hardware could announce a low battery condition, which would be handy to allow replacement before the battery dies. I've also just discovered that futurlec have started operations in Australia! Their prices (in AUD) are about 1/3 of my local distributor for micros :) -----Original Message----- From: pic microcontroller discussion list [mailto:PICLIST@MITVMA.MIT.EDU] On Behalf Of Russell McMahon Sent: Wednesday, 17 March 2004 2:44 PM To: PICLIST@MITVMA.MIT.EDU Subject: Re: [EE]: Identifying cars > Make a giant sized magnetic barcode and stick one on the side of each car. > A 2 foot by 3 foot barcode should be readable from several feet, no? Railways use these. Readable from a distance ______________ Cars have receiver which sees a beacon (eg laser line). Car then transmits it's ID and a local clock AFTER passing line for several iterations with random pause between. Could be done with eg IR. Could have several trigger points on track. Needs uP in car. Needs timing in car. Needs receiver (one per trigger point) Not overly complex. As long as car clock is stable enough for period of race the absolute clock value is irrelevant - it self calibrates after the first reading. Tlap = Clock_now - clock_then. 100 f/s = 10 ms resolution for 1 foot timing. = 1 mS for 0.1 foot timing RM -- http://www.piclist.com hint: To leave the PICList mailto:piclist-unsubscribe-request@mitvma.mit.edu -- http://www.piclist.com hint: To leave the PICList mailto:piclist-unsubscribe-request@mitvma.mit.edu