Thermionic Idler wrote: ↑Wed Jan 11, 2023 5:44 pm
Nick wrote: ↑Tue Jan 10, 2023 10:06 pm
Did you not try my suggestion of changing the value of the pull up resistors? I still think 47k is wrong even in the case of single input single output. Would/might have avoided a new board.
I have not, but there was a genuine reason for that - I can't get my side cutters in, the resistor networks are simply packed too close to other components on the board for me to be able to snip them off.
I will look further into this though, before the re-do, and ask a question on the AMB forum about why 47k was picked - I do know that Ti Kan and Brian of Linuxworks spent some considerable effort refining this design empirically to work reliably, and so they must have arrived at that value for a reason. And I've had that driver circuit in place and working on AMB-issued PCB's for years and it's been flawless - the only issues originated from the Arduino and those were my fault.
The approach I've tried to take is only change what needs to be changed and don't second guess anything, cos these guys know a lot more about this stuff than I do! Look what happened when I deviated from the design and tried to connect one TTL to two darlingtons.
I would ave just soldered the four resistors on the back of the PCB leaving the 47k resistors in place. Adding 1k or 2k2 or 4k7 in parallel with 47k is as close to be good enough. That would be good enough to get the four lines workign or at least see if I was talking tosh.
My thoughts as to the size used are based on the spec sheet, the input current is specified as 0.93ma. So just being very sloppy, a 4k7 pull up would give
( 5v - 1.2v ) / ( 4k7 + 2k7 ) = 0.5ma
1k pull up would be
(5 - 1.2) / ( 1k + 2k7 ) = 1ma
The 1.2v is the two forward biased junctions in the darlington. The 2k7 the internal resistor.
The 47k they have specified would give
(5 - 1.2) / (47k + 2k7) = 76ua it seems to me, when you add the 100ua that seems to be in the most the PCF8574 can source, that gives you at most 0.18ma. Use that .18ma and that looks like it would give at most 100ma collector current on the output. Probably enough to drive the relay but its all a bit close to the edge IMHO.
Little known fact, coherent thought can destructively interfere with itself leaving no thought at all, that’s why I prefer incoherent thought.