Dear Russel, At 02.11 2005.02.12 +1300, you wrote: >>This circuit really seems to ensure full swing (I've been >>able to successfully test this aspect) but, will it give >>maximum speed? This is may concern. > >The circuit you drew with 2 transistors should work to 100 kbps plus provided you change the 100R to 1k to 10k. >The problem you had before with low voltage swing was due (probably) to attempting to drive too high a current with the opto for the avail;able drive. Typically you get an output current of 10% of the input current maximum. )Check data sheet for your opto). It's not much better than that.. in fact. But I didn't really understand this point of the data sheet, hence my failure. >The following circuit (use fixed width font) should work and uses only 1 transistor. I use something similar at 100 kHz+ as a FET gate driver. > > > > RM > > > ,-----------------.---------------- DTR > | | (low, i.e. +12V) > | | > \ 1k - 10k | > / C > \ ,---- B NPN (BC547) > / | E > \ | | > | | | > | 10K | | >5 ------*----------* *---------------- RX > 4N25 | | >4 ------. |-|<|--. 1N4148 > | > | > | > | > | > | > | > '--------------------------------- RTS > (high, i.e. -12V) Ingenious, brilliant! I've been trying ways to shut off the transistor when the opto was saturated, but didn't find a way. It's in your schematic above. :) Speaking of pure performance and high/low simmetry, do you think that the PNP/NPN circuit may perform a bit better than the circuit above? Number of components-wise they're fairly similar anyway (1 more transistor versus 1 more diode and 1 more resistor). What I liked most of the PNP/NPN approach was its simmetry. Greets to NZ, TPM -- http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist