Re: 5V vs 3.3V RAM modules (TP701C)

[skat@pacific.net.sg (Sam King )]

At 12:45 AM 4/22/96 -0400, you wrote:
>First of all, bear in mind that the memory socket is designed for 3.3V circuits
>Secondly, for a 5V IC to work, it REQUIRES 5V!!!!!
>Third, according to Ohm's law, V=IR, where V=voltage, I=current, R=resistance.
>So for the 5V memory board to work, the thinkpad MUST output a 5V to the 5V
>module, or else, there will not be enough voltage for the module to
distinguish >0 and 1.  (I think anything > ~4.2 is considered 1.  Don't
quote me on this.  I >am not too sure).  So the current output is now 5/R
instead of 3.3/R.  So the >IBM tech is probably right......

actually, it's the other way around. logic 0 is ranging from 0 to something
like 0.7 volts (it really depends on the circuit) but the threshold for
logic 1 is actually quite low, definitely less than 3.3

bear in mind that the 5volts is a rating on the memory module. it's clearer
for notebook manufacturers that way. i guess they could quote it in watts
(power consumption) but i don't think that any notebook manufacturers ever
made notebooks with currents that could fry the memory chips, and it was
easier for the public to follow (never mind about confusing!) if they just
labelled the voltage rating.

so, the chips don't NEED 5volts, but they will draw more current given less
voltage, and (this part i'm not sure) that being the case, circuitry via the
power supply to the memory module are likely to be fried too... had this
problem with a robot last week. we put in another sensor without changing
the resistor. it was added in parallel so we thought the voltage drop would
be the same, but we forgot that more current was drawn, and so we blew the
motherboard :(

sam