This is a multi-part message in MIME format.

------=_NextPart_000_0028_01C61ABE.8EFC0FB0
Content-Type: text/plain; format=flowed; charset="iso-8859-1";
	reply-type=original
Content-Transfer-Encoding: 7bit

> I have learned a lot from your messages on this topic (I'd always 
> assumed that "saturated" was pretty-much a
> single constant state, and it never occurred to me that there could 
> be such a change in Vce).
>
> However, this:
>
>> (10:1 = 3.32 BITS FWIW).
>
> is one of the daftest things I've seen you write!  :-)

That's as maybe :-).
Entirely correct though :-)
( ie 1 / log10(2) = 3.32 ...)
(to 3, excessive in the context , significant figures.)

I was attempting to forestall the inevitable comment about ratios when 
my two preceding 'bits of accuracy' related comments were compared.

>> My first attempt used a single floating transistor but this 
>> (stupidly)
>> included the base drive in the lower divider resistor. Usually 
>> that's
>> reasonably irrelevant. Here it's fatal. So I now have two 
>> transistors
>> in a classic bipolar high side drive.

> Are you going to publish the circuit, for those of us lacking in 
> classical education?  :-)

Possibly useful, so, OK.
Very simple but still needs a little design.

For battery ~= 12 V.
RBatt3 could be eg 100k - value non critical.
Rbatts ensures that base current of QBatt1 is about (12-Vbe) / 10k ~= 
1 mA.
Ic of QBatt1 is about 12/78k ~= 0.15 mA so Beta =~ 0.15/1 = about 0.1 
ish in this instance.

Rbatt5 & Rbatt4 divide the 12 volts to put it in the ADC conversion 
range.
68k/10k chosen for my application. Alter as needed.
The combination has impedance < 10k so meets ADC conversion speed spec 
(which is not overly important here as voltage only varies 
verrrrrrrrrrry slowly).

QBatt2 is shown with NO base pulldown resistor - turns off fine when 
if uP drive is active low BUT if the uP is fully powered off it should 
have a large value R to ground (1 M or 100k OK) to ensure transistor 
stays off when it oughta.

Output UVBAT is never outside processor allowable range (by design) 
and there is no applied voltage and no current drain when off.

> To my rather
> simplistic understanding, Gain (Beta) is Collector current / Base 
> current.  To get this < 1, Ic must be less
> than Ib.  Where does the "extra" Base current go?  (I was never any 
> good at transistor electronics - digital
> is my thing :-)

See diagram. Battery supplies Ie = Ic + Ib.
Output divider sees just Ic while the rest exits stage left (here) via 
RBatt2.

Battery load is of no import here as main load is >> divider load.


> Is significant settling time needed between turning on the divider 
> and measuring the voltage?  (You don't
> actually say you're measuring battery voltage, but this sort-of 
> falls out from the fact that you're connecting
> it to an ADC).

I haven't done any actual ADC measurements yet, but no. The input will 
come up to voltage in a small time compared to processing speed. (10k 
and 1 NF has a time constant of 10 uS. Actual pin capacitanmce will be 
<< 1 NF so N Tcs settling allowed. Also, once on it can (& will) be 
left on while uP is operating so it is fully transparent.



        Russell McMahon

------=_NextPart_000_0028_01C61ABE.8EFC0FB0
Content-Type: image/gif;
	name="adcsw40.gif"
Content-Transfer-Encoding: base64
Content-Disposition: attachment;
	filename="adcsw40.gif"
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------=_NextPart_000_0028_01C61ABE.8EFC0FB0
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist

------=_NextPart_000_0028_01C61ABE.8EFC0FB0--