Lpc2000

Good points! I was planning to use a Zener to keep the voltage below 
4.5V when the load gets REALLY low (below 1.5 mA, in our case). 
However, I'm more and more comfortable with using two MAX1684 (one 
for 3.3 and one for 3.6); I'd use a 3.6V for the radiomodem so I 
don't need to care much about load regulation, to which the 
radiomodem is very sensitive if the input voltage is only 3.3V.

About the MAX availability, there's stock in Digikey for MAX1684 and 
MAX1685, as well as in MAXIM's online store, NewarkInOne and AVNet. 
Our device will be produced in low quantities, and in the short term, 
so I hope availability doesn't become a problem.

Guille


--- In lpc2000@yahoogroups.com, Bruce Boyes <bboyes@s...> wrote:
>
> >> I'm currently experimenting with a MAX1684 for the main circuit 
(<150
> >> mA, 100 mA typical when full active) and a linear regulator 
(LM317)
> >> for the radiomodem (<5mA typical 98.3% of the time, 160~200mA 
average
> >> 1.6% of the time). I'm trying to trick the LM317 with large R1-R2
> >> values (~8KÙ) to keep power consumption low. The radiomodem will 
work
> 
> Careful there. Most versions of the 317 won't maintain regulation 
without a
> minimum load, so check the data sheet carefully. We just brought up 
a new
> design where the tech misread the schematics, used setpoint 
resistors which
> ere too high, and the result was the 317 output was equal to the 
input (30V=
> )
> due to too-low output current, and this smoked some 3.3V parts on 
the outpu=
> t.
> 
> Also before you design in Maxim, be sure you can really get them, 
and there=
>  is
> a second source if possible. Maxim has a history of long and/or 
unpredictab=
> le
> lead times - we are periodically bitten by this. Some Maxim parts 
have uniq=
> ue
> functions which we need, but it's frustrating...
> 
> Linear Technology has some great, non-commodity regulators which we 
use in
> quantity. Efficiencies are good, support is good, and so is 
availability.
> 
> Regards
> 
> Bruce
>