A8 Motherboard: Is the 74LS08 chip Really Needed?

[Level42]

 

In the end it came down to using a 74F08 when developing the 1088XEL, which took the last bit of flakeyness out of the system, and it would likely work just as well in a real A8 system.

 

- Michael

 

So what signals are buffered on the 1088XEL now ?

[+mytek]

 

So what signals are buffered on the 1088XEL now ?

 

PH0 going into the CPU, and PH2 coming out of it are both buffered by the 74F08.

 

Schematic of that section: CPU.pdf

 

- Michael

[+mytek]

With all the recent swapping of the Atari's 74LS08 for a 74F08, and the introduction of the Phi2 Fixer board based upon this chip - all in an effort to improve stability, I remembered doing an in-depth study of this back in 2017. That led me to this topic which came out of research being done during the development of the 1088XEL where minimization of components was key, and specifically the necessity of the 74LS08 was the focus at that time.

 

So here's a BUMP in case anyone is interested in where the use of the 74F08 really started to take hold as a substitute for the original 74LS08, and some of the other possible alternatives that led up to choosing it in the end. By far the 74F08 substitution turned out to be as simple as it gets to fix many a problem with modern upgrades in our old Atari systems.

[reifsnyderb]

I read through all your research on this with great interest.  Is there a reason you opted not to install a 68 ohm resistor in series with the Phi2 buffer?

[+mytek]

18 minutes ago, reifsnyderb said:

I read through all your research on this with great interest.  Is there a reason you opted not to install a 68 ohm resistor in series with the Phi2 buffer?

Well at the time I was really looking for a no-change drop-in replacement for the 74LS08 that would improve things. But I wasn't adverse to having to add parts if needed. However when I discovered that the 74F08 would do the trick on its own without creating big under and/or over shoot spikes I was sold, and got my drop-in replacement - no resistor needed.

 

Correction: In the beginning I was really looking at the need for any buffering, but after discovering that it was necessary, settled on the 74F08. Later still in the 576NUC+ since I also needed a few inverters for other parts of the logic this got changed to a 74F04 chip, with a couple of it's gates back-to-back for the buffering. For some reason on that same system I went with a 74AHCT14 for the crystal oscillator and CSYNC buffering instead of a standard speed HCT part (might have been a problem sourcing an SOIC version at the time).