Monitoring power supply voltage
Archiver
Posts: 46,084
Hello all. Does anyone have any experience with monitoring the power
supply that is powering the STAMP? I would like to hook an AD
converter up to the power supply so that I can determine if the
battery is good, dead or needs replacing. Is the ADC overkill? Could
I just use a cap, resistor and RCTime?
Thanks in advance,
--Jeff
supply that is powering the STAMP? I would like to hook an AD
converter up to the power supply so that I can determine if the
battery is good, dead or needs replacing. Is the ADC overkill? Could
I just use a cap, resistor and RCTime?
Thanks in advance,
--Jeff
Comments
Look at
http://www.emesystems.com
This site, from Tracy Allen, is a gold mine of tips and tricks for the Basic
stamp.
It is a "must be seen" for all serious Basic stamp users.
Note that I have no relation with Tracy Allen but I use this message to thank
him
for the very interesting site.
Regards
ECO
Original Message
From: <bs2web@y...>
To: <basicstamps@yahoogroups.com>
Sent: samedi 5 mai 2001 09:21
Subject: [noparse][[/noparse]basicstamps] Monitoring power supply voltage
> Hello all. Does anyone have any experience with monitoring the power
> supply that is powering the STAMP? I would like to hook an AD
> converter up to the power supply so that I can determine if the
> battery is good, dead or needs replacing. Is the ADC overkill? Could
> I just use a cap, resistor and RCTime?
>
> Thanks in advance,
>
> --Jeff
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>
get. The Stamp has a certain threshold on its logic pins (I think it is
about 2.4V on the regular Stamp -- you can test it to be sure). Below the
threshold is a 0, above is a 1.
So say you have a 9V battery and you want to know when it gets to 7V. You
could arrange a voltage divider so that 7 --> 2.4
That's 34.29% (about). So if you put a 10K resistor between the pin and
ground, you'd need a 19.156K resistor between 9V and the pin. Say you use a
20K resistor since you won't find 19.156 in your junk box.
At 9V that would give you 3V -- enough for a solid 1 reading.
At 7V that would give you 2.3V - A zero.
So you'd get a zero somewhere just above 7V. You could juggle the resistor
values around to get whatever threshold you wanted. A pot is infinitely
variable, but they aren't very temperature stable necessarily. 1% resistors
would be better -- depends on how crucial things are.
If the 10K resistor is R2 and the 20K resistor is R1 then:
Vout = Vin*R2/(R1+R2)
Why? Assume the Stamp pin has no impedance (its resistance is much higher
than R1 and R2). Then the current flowing through the resistors must be:
Itotal = Vin/(R1+R2)
And the voltage seen on the pin is the voltage across R2 (which is grounded
on one end). So:
Vpin = R2 * Itotal
or
Vpin = R2 * Vin / (R1+R2)
But it is customary to write the resistors together to form a ratio:
R2/(R1+R2) and then multiply the ratio by Vin. Math-wise this is the same
answer either way (12*3/6 = 3*12/6)
More on things like this at http://www.al-williams.com/wd5gnr/basiccir.htm
One thing to look for is that you might get a "bouncing" between 1 and 0
right at the threshold. You could clean this up in software. Or you could
use a comparator with hysterisis. You'd set up another divider to get your
threshold voltage using 5V (which is regulated) and compare the two
voltages. Because of the hysterisis, you wouldn't get the bounce. You could
also design your input divider so that the low voltage point was at 5V and
compare to your regulated supply directly.
You could use an A/D. You could also use the charge rate of a cap, although
really that's the same thing as the voltage divider (measurement against
your threshold) so I don't know what the benefit would be.
Regards,
Al Williams
AWC
*Mention Code M7SHIP on or before May 7 2001 and get FREE standard US
shipping on any order! http://www.al-williams.com/awce/
>
Original Message
> From: bs2web@y... [noparse]/noparse]mailto:[url=http://forums.parallaxinc.com/group/basicstamps/post?postID=-J4NnP-PbCLd0345pN8C0sE8ng5kaQc44JRVd8ydJWmR76hPab136sdC8MwFyD7TY0-S3k4yEQ]bs2web@y...[/url
> Sent: Saturday, May 05, 2001 2:21 AM
> To: basicstamps@yahoogroups.com
> Subject: [noparse][[/noparse]basicstamps] Monitoring power supply voltage
>
>
> Hello all. Does anyone have any experience with monitoring the power
> supply that is powering the STAMP? I would like to hook an AD
> converter up to the power supply so that I can determine if the
> battery is good, dead or needs replacing. Is the ADC overkill? Could
> I just use a cap, resistor and RCTime?
>
> Thanks in advance,
>
> --Jeff
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
to measure the battery voltage can be found at
http://www.emesystems.com/BS2rct.htm
>Jeff
>Look at
> http://www.emesystems.com
>This site, from Tracy Allen, is a gold mine of tips and tricks for
>the Basic stamp.
>It is a "must be seen" for all serious Basic stamp users.
>Note that I have no relation with Tracy Allen but I use this message
>to thank him
>for the very interesting site.
>Regards
>ECO
>
>
> > Hello all. Does anyone have any experience with monitoring the power
> > supply that is powering the STAMP? I would like to hook an AD
> > converter up to the power supply so that I can determine if the
> > battery is good, dead or needs replacing. Is the ADC overkill? Could
> > I just use a cap, resistor and RCTime?
> > Thanks in advance,
> > --Jeff
I've been wondering myself how best to monitor the input voltage to the
stamp. The RC time example on Tracy's site works very well(surprise!). One
thing that has been bothering me about using RCtime or a simple voltage
divider is that it seems it will increase the current requirements of the
project?. I've just finished trying out using a transistor to switch the
voltage to the voltage divider on and off. This gets me back to the nominal
sleep current of around 25 microamps. This seems to work well, when taking
into account the voltage drop from the transistor. The voltage drop from
the transistor will change slightly due to temperature, but maybe not
enough to significantly change the incoming voltage to the voltage divider?
What I'm trying to say is, how best to measure the stamps input voltage,
without increasing the current consumption while the stamp is sleeping?
Thanks
---Steve
At 10:09 AM 05/05/2001 -0700, you wrote:
>Thanks for the complement, ECO. Jeff, th article about using RCtime
>to measure the battery voltage can be found at
> http://www.emesystems.com/BS2rct.htm
>
>>Jeff
>>Look at
>> http://www.emesystems.com
>>This site, from Tracy Allen, is a gold mine of tips and tricks for
>>the Basic stamp.
>>It is a "must be seen" for all serious Basic stamp users.
>>Note that I have no relation with Tracy Allen but I use this message
>>to thank him
>>for the very interesting site.
>>Regards
>>ECO
>>
>>
>> > Hello all. Does anyone have any experience with monitoring the power
>> > supply that is powering the STAMP? I would like to hook an AD
>> > converter up to the power supply so that I can determine if the
>> > battery is good, dead or needs replacing. Is the ADC overkill? Could
>> > I just use a cap, resistor and RCTime?
>> > Thanks in advance,
>> > --Jeff
>
>
>
>
>Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>I've been wondering myself how best to monitor the input voltage to the
>stamp. The RC time example on Tracy's site works very well(surprise!). One
>thing that has been bothering me about using RCtime or a simple voltage
>divider is that it seems it will increase the current requirements of the
>project?. I've just finished trying out using a transistor to switch the
>voltage to the voltage divider on and off. This gets me back to the nominal
>sleep current of around 25 microamps.
>...
>What I'm trying to say is, how best to measure the stamps input voltage,
>without increasing the current consumption while the stamp is sleeping?
> > http://www.emesystems.com/BS2rct.htm
The example at the above URL shows a 100kohm resistor and 0.068uf
capacitor as the timing components.
Increase the resistor to 1 megaohm (1% film resistor) and decrease
the capacitor to 0.0068uf, and the timing will be the same. And
current "lost" will be reduced to 7 microamps. (assuming a 12.0 volt
battery and that the RCtime pin is turned around to be an INPUT or a
HIGH OUTPUT before your program hits the SLEEP command. [noparse][[/noparse]12-5]/1M =
7 u-amps)
I think the following is also true. I would have to check it again
to see if I am remembering right. The 7 microamps actually is _not
lost_. If the pin is a high output or an input, that current flows
directly into the Vdd power supply and actually decreases the current
that has to be supplied by the 5 volt regulator. So the sum total
stays at 25 microamps.
-- regards,
Tracy Allen
electronically monitored ecosystems
mailto:tracy@e...
http://www.emesystems.com