Olin Lathrop wrote:
>> I understand all this.
>
> No, you don't.

Yes I do, it just wasn't a concern at the time.  ;-)  You were going on
about the zener efficiency, and I was trying to make the point that the
zener wasn't even conducting hence no loss of efficiency there.

> You should do some research on "switching power
> supplies".

I'm about to.  I've been aware of their efficiency for some time, I just
never built one before.  This should be interesting.

>> I was afraid this was where the discussion was
>> heading.  Since I have a 10K resistor in series with the line, I can
>> only draw a maximum of 3.5mA.  Since I'm already drawing close to
>> that, the input voltage to the PIC is already below the 6V zener's
>> knee. AFAIK, the zener isn't even conducting except when the circuit
>> load is smaller, so where is my loss of efficiency?
>
> In the resistor, of course.  A 10Kohm resistor at 3.5mA dissipates
> 123mW, which is the wasted power in this design.  The load only
> receives 3.5mA at 5V = 18mW.  So you're drawing 140mW from the phone
> line but only using 18mW of it.  To get the power delivered to your
> PIC from a 40V line only requires 440uA in theory, not 3.5mA.  A
> switching buck converter that is 80% efficient would require 550uA,
> still a long way from 3.5mA.

Gotcha.

> By the way, I also think that sooner or later the phone company will
> catch up with this an insist you remove the load from the line before
> re-enabling your phone service.

We will see.  ;-)  I bet they don't especially if I can get it down to
about 500uA.  At any rate, that was not my goal as this is only an
experiment, not something that will be left on the line permanantly.
I'm only powering the LCD on "stolen" on-hook current for my initial
tinkering.  In the end, the PIC will be the only device powered this way
and it will be sleeping at the time.  Current should be so low that any
efficiency to be gained from an SMPS would be offset by the cost/number
of parts.  I still plan to experiment with SMPS.

One question:  Is it feasable to build an SMPS that will accept an input
voltage ranging from 9 to >50V yet still provide regulated 5V output?

>> Ideally I would have used a 5V1 zener, but I didn't have one and hey,
>> it's a PIC and it can take it.  ;-D
>
> Maybe that PIC on that day at that temperature.  There are no
> guarantees out of spec.

I picked up a 5V1 zener yesterday, that should be a bit better.

michael brown

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu