and I won't toss it for historical reasons. After all, it was the "bible" back when there was no SGS. I wasn't even aware that SGS made TTL. They made quite a bit of CMOS logic, of course. I compared with the Fairchild book, and they, more or less, agree. The NSC book is down here in the heap somewhere ...

There is a pullup, but it's on the anode side of a Schottky diode, the cathode of which is the input, and it's shared with the other inputs on simple gates. However, it's absent from the PNP input type of LS logic. Its bias apparently comes from the emitter-base junction current. There are at least two types, after all. Fairchild describes two input types, one, the one with the shared pullup resistor, is the diode-cluster input, while the other is the PNP input. The PNP-input stage has a pullup on the emitter of a PNP, the base of which is the input. The only apparent difference is that the PNP takes the place of one of the Schottkies in the diode-cluster input. Though there are not any direct pullups on the LSTTL inputs, there are currents that flow to the input through a number of diodes. The result would be that the external logic only has to drive the input low, as the inputs are weakly pulled into the positive region. The effect, of course, is that of a pullup, but then, so is that of the direct emitter input on standard TTL.

RE