I did an lt spice sim here, where I have 2 lines in parallel, and series termination. you said this becomes ratty, but my simulation is fine. you simply forgot to correct the series resistor value... agree?

Sorry, afraid I can't.
You picked the pathological case of the two lines being exactly equal length (being driven exactly in the middle of a single line).
Here's what happens when they aren't (which would likely be the real condition):
Yellow traces are for equal length (your simulation), green traces are for T2 being 2/3rds the electrical length (33.3ns) of T1 (50ns).
The signal rather ping-pongs between the two unterminated ends.

 

Sorry, afraid I can't.
You picked the pathological case of the two lines being exactly equal length (being driven exactly in the middle of a single line).
Here's what happens when they aren't (which would likely be the real condition):
Yellow traces are for equal length (your simulation), green traces are for T2 being 2/3rds the electrical length (33.3ns) of T1 (50ns).
The signal rather ping-pongs between the two unterminated ends.

View attachment 146353

ok I get it ! I was ignoring electrical length of the line................. thank you very much Cruts. So I'll have to use thevenin at both ends. Do you think a 40 inch long bus will be alright with 74HC and some approximate termination ? I think I will split the bus into 3 pieces as well.

 

ok I get it ! I was ignoring electrical length of the line................. thank you very much Cruts. So I'll have to use thevenin at both ends. Do you think a 40 inch long bus will be alright with 74HC and some approximate termination ? I think I will split the bus into 3 pieces as well.

I think termination at both ends will be okay, even with a 40" line, as long as you can drive the double termination.
Splitting the bus would reduce the risk, of course.

 

I think termination at both ends will be okay, even with a 40" line, as long as you can drive the double termination.
Splitting the bus would reduce the risk, of course.

I will be using 244 buffers to drive the lines... They can drive up to 35mA maximum. I will have to use large enough resistors not to get too low a voltage drop or too high a current. What value do you think would be ideal?

my calculations give that for a 4V HIGH state, a resistor of 90 Ohm for thevenin, would give a current of 33mA for 30 ohm output impedance of the driver...

this is for single end termination, which should be good enough. Anyways, that's not too bad! I would now double or tripple this resistor's value, terminate at both ends, and get something decent... right ?

 

I will be using 244 buffers to drive the lines... They can drive up to 35mA maximum. I will have to use large enough resistors not to get too low a voltage drop or too high a current. What value do you think would be ideal?

I would start with about a 300 equivalent for each end and go from there.
That means the drivers have a 150Ω equivalent load.
If the reflections are too large, then the polarity of the reflections will tell you which way to adjust the termination resistances.
If the reflection is smaller than the initial value, the termination resistance is too low.
If the reflection is larger than the initial value, the termination resistance is too high.

Simulation below with a 244 unterminated driver and 300Ω line impedance:
Yellow is termination too low (200Ω eq) and green is termination too high (400Ω eq) for T2 [V(out2)].

 

I would start with about a 300 equivalent for each end and go from there.
That means the drivers have a 150Ω equivalent load.
If the reflections are too large, then the polarity of the reflections will tell you which way to adjust the termination resistances.
If the reflection is smaller than the initial value, the termination resistance is too low.
If the reflection is larger than the initial value, the termination resistance is too high.

Simulation below with a 244 unterminated driver and 300Ω line impedance:
Yellow is termination too low (200Ω eq) and green is termination too high (400Ω eq) for T2 [V(out2)].

View attachment 146366

Fantastic..........very good indeed. This pretty much solves my original problem. I think I will also split the bus into 3 parts because that is easy enough to do with just 3 buffers and 3 extra control signals. This coupled with thevenin termination will give me a clean enough signal and no worries! Should I also add a 100nF capacitor across each line from Vcc to GND right at the terminations? For both terminations?

Anyhow. Damn! You have helped me more than I would expect from anyone! Thank you very much Cruts! I will show you my finished project when it's done if you're interested, it's a 16bit 8086 compatible CPU/Minicomputer built solely from 74HC!

 

Should I also add a 100nF capacitor across each line from Vcc to GND right at the terminations? For both terminations?

Yes.

I will show you my finished project when it's done if you're interested

I am. Please do.
I'd be interested to see pictures of the bus signal.

 

Yes.
I am. Please do.
I'd be interested to see pictures of the bus signal.

Yes! Till the next time comrade.