On 8/28/22 02:08, Henner Zeller wrote:
On Tue, 23 Aug 2022 at 23:36, Brian K. White <[email protected]
<mailto:[email protected]>> wrote:
Heads up though, Steve has me convinced about that (A)* idea he
mentioned in an earlier post.
I've made a new version based on that idea but it needs some testing to
work out how hard (A)* should be pulled, a dead short or a resistor of
what value? A simple short would make the pcb simpler but right now I
made room and got an 0805 footprint crammed in there.
I did some measurements and attempted to pull down the A* signal with a
resistor (I didn't want to start with a dead short).
Here is the A* (or, RDWR* as I call it) signal without any pull-down. It
goes nicely 5V rail-to-rail
no-resistor.png
And here, I pulled it down with 120 Ohm, and we _still_ get about 2.8V
voltage
120ohm-pulldown.png
That means that this draws about 2.8V / 120 Ohm ≈ 23mA and we still have
not pulled down the signal enough to get even close to a TTL threshold
voltage for 'low' (< 0.8V).
I didn't want to go further (we would probably need more than 40mA to
get even close), as this likely exceeds the rating of the chip. To pull
down A* it would take significantly less than 100 Ohm.
yyeowwww
I have not found a historic datasheet of the 40H367,
Prepend "TC". I only tried that because I looked at the parts list in
the service manual after I couldn't find it either a few days ago.
https://www.datasheetarchive.com/pdf/download.php?id=b30ebc8d72908297a24a7edc1595ebaf432084&type=M&term=TC40H367
Steve said he measured it around 8.5, but ramping up so that maybe it's
only 4 or so average over the whole pulse. Maybe that 8.5 was itself
only a measurement artifact of the briefness of the pulse.
I'll test what the PG Design does sometime. It doesn't conflict with the
internal ram while in bank1, but in any other bank it must be
suppressing the internal ram somehow.
Maybe my first excuse to play with the unnecessarily fancy...
https://github.com/bkw777/ScopeShuntX
but a modern equivalent would be a 74HC367.
The datasheet for that chip (
https://www.ti.com/lit/ds/symlink/sn74hc367.pdf
<https://www.ti.com/lit/ds/symlink/sn74hc367.pdf> ) lists in the
'absolute maximum ratings' (read: don't do this if you want your chip to
survive...)
35mA as continuous output current and the output clamp current as 20mA.
With a pull-down of 120Ohm, we are already above 20mA.
So for my personal playing with the system bus, I will _not_ consider
pulling down A*, but rather the pulling up RAMRST signal. Not only does
it only require about 1.7mA for overpowering T9, but also we only put
strain on a discrete component that is much more resilient. From both
somewhat-out-of-spec choices, pulling up RAMRST seems much less
dangerous than pulling down A*.
Also, I still don't see an advantage in using A*: The relevant reset
One thing would be the internal ram stays chip-disabled all the time
instead of just output-disabled. That's up to 16 ic's all turned off.
But any advantage from that would be overshadowed if it's really burning
that much through A*.
But I think Steve might have had something else in mind about using the
ramrst signal for something (besides power).
--
bkw
