> -----Original Message----- > From: Brent Hilpert [mailto:bhilp...@shaw.ca] > Sent: 27 August 2017 09:00 > To: r...@jarratt.me.uk; General Discussion: On-Topic and Off-Topic Posts > <cctalk@classiccmp.org> > Subject: Re: DECstation 220. Another Impasse > > On 2017-Aug-26, at 11:09 PM, Rob Jarratt wrote: > >> You have some of your reasoning reversed regarding comparator behaviour. > >> Comparators are typically: +in > -in --> output transistor off --> > >> output > > TRUE / > >> HIGH with pull-up R. > > > > > > I have double-checked the datasheet > > (http://www.ti.com/lit/ds/symlink/lm339.pdf). It says "The output NPN > > sinks current when the positive input voltage is higher than the > > negative input voltage and the offset voltage". My interpretation is > > therefore the opposite of what you say above. Is it possible that > > different versions of the LM339 work differently? > > > No, the writer needs to be shot. OK, fired. Well OK, demoted. > > Move down from section 8.3 where the comment you quote is, to section > 9.2.2.1: > When both IN– and IN+ are both within the common-mode range: > – If IN– is higher than IN+ and the offset voltage, the output is > low and the output transistor is sinking current > – If IN– is lower than IN+ and the offset voltage, the output is > high impedance and the output transistor is not conducting > etc. > > They seem to have given up on including circuit examples in the datasheet but > if you look at older datasheets you can work through examples to see the in/out > relation. > All the standard comparator circuits like an oscillator or adding hysteresis with > a feedback R to the +in wouldn't work if they functioned in the manner you > were reading. > > Another way to look at it is that a comparator is essentially an open-loop op- > amp and as with an op-amp the output follows in like direction to the +in and in > inverse to the -in. >
Thanks for pointing that out. I have to say that it seemed logically strange but I thought these datasheets were all pretty accurate, that one is a pretty big mistake, although as you point out it is correct elsewhere. The peaks of the G176 output are about 300ns long. I can't tell from the datasheet if that is long enough, but that may be because I don't understand datasheets all that well. What I can see is that it can take up to 1.3us to respond, but that isn't quite the same thing (I think). I have added an addendum to my post now which explains what the G176 on the other board does. This is what I added: "For comparison, on the original board, where I get a corrupted video display, the G176 outputs rather different signals, with a signal on pins 1, 2 and 3, the signals seem to vary in amplitude from 1V to 2V. I can't see the length of the peaks though because the signal is too irregular, unlike on the board in question. The difference in behaviour could be either cause or effect, because, I assume, the BIOS is telling the G176 what to do and that may depend on whether it has sensed anything at all. Or it could just be a faulty G176." Regards Rob
