22 Oct
2011
22 Oct
'11
4:16 p.m.
The PPS-4 was actually a bit more
'sophisticated' than I
first thought. One other thing that was unexpected was
that they are powered eg. 0 and negative 9V. I wonder why it's like that...
The early 70's LSI was mostly PMOS.
Simpler devices were often a good match to a 9V battery in handheld devices where that was
the target. You wouldn't believe how much effort was put into putting LED drivers on
the same chip where they could (often they couldn't and the LED driver was off the
main logic chip.)
More complex PMOS logic often had special supplies for clock and enable lines. You often
find small switching converters (magnetics or charge pump) to generate the required
voltages.
A little later high-density NMOS also would use oddball voltages (both positive and
negative depending on application) for density's sake. e.g. 8080A.
Some good examples of how some handheld devices generated the necessary oddball voltages
at http://www.jacques-laporte.org/HP35%20power%20unit.htm
Tim.
23 Oct
23 Oct
3:44 p.m.
The PPS-4 was
actually a bit more 'sophisticated' than I
first thought. One other thing that was unexpected was
that they are powered eg. 0 and negative 9V. I wonder why it's like that...
The early 70's LSI was mostly PMOS.
Simpler devices were often a good match to a 9V battery in handheld devices where that
was the target. You wouldn't believe how much effort was put into putting LED drivers
on the same chip where they could (often they couldn't and the LED driver was off the
main logic chip.)
More complex PMOS logic often had special supplies for clock and enable lines. You often
find small switching converters (magnetics or charge pump) to generate the required
voltages.
A little later high-density NMOS also would use oddball voltages (both positive and
negative depending on application) for density's sake. e.g. 8080A.
Some good examples of how some handheld devices generated the necessary oddball voltages
at http://www.jacques-laporte.org/HP35%20power%20unit.htm
Tim. 
4:11 p.m.
I still don't understand the reason for the rev.
polarity, but it
It's similar ot the difference between NPN and PNP transistors.
Everythign flips upside-down :-)
seems that high would then be -9v?
Probably. You cna actualyl take the logic levels whichever way you want,
and provided you're totally consistent (and consistent in how you name
the gates), it'll all work.
There is nothing wrong with the statement that 'The 7400 is a quad
2-input NOR gate with a +5V power supply and logic levels such that '1'
is between 0 and 0.8V and '0' is between 2.4V and 5V'.
[Before I get an long string of messages telling me the databook says
that the 7400 is a NAND gate chip, I know that!.But notice I've also
swapped the logic levels round...]
I guess its because of the way the PMOS gates are
made.
Why are the LED drivers so difficult to make, is it because of the
'high' (a few mA) current ?
Baiscally, yes. The logic transistors work at a much lower current. Often
the separate LED driver chip was a bipolar (as opposed to MOS) device.
-tony
6:04 p.m.
On Oct 23, 2011, at 3:44 PM, Benjamin S?lberg wrote:
Why are the LED drivers so difficult to make, is it
because of the 'high' (a few mA) current ?
In the IC world, a few mA is pretty high current. The amount of silicon devoted to making
the I/O pads can actually dwarf the area devoted to the logic on a chip with a lot of
multi-mA I/Os. Even as recently as the '90s, it was a bit of a luxury to drive LEDs
directly from I/O pins (for example, I seem to recall the near-ubiquitous PIC16F84 only
had a few pins capable of that, the rest you'd want to slap a 2N3904 on the output to
drive your LEDs).
- Dave
4661
days inactive
4662
days old
3 comments
4 participants
participants (4)
-
ard＠p850ug1.demon.co.uk -
benjamin.soelberg＠gmail.com -
fraveydank＠gmail.com -
tshoppa＠wmata.com