# Comparator circuit for temp indicator

Discussion in 'Electronic Basics' started by dwbauer, Mar 26, 2006.

1. ### dwbauerGuest

I have two circuits at http://bellsouthpwp.net/k/e/kelmark2180/therm.html to
show what I am thinking. I am trying to make a lead acid charging
temperature monitor. The circuits are from existing url's that I found by
searching Google. My question is repeated on the page as follows:
I want to set up a temperature monitor for charging lead acid batteries. I
think the sensor can be imbedded into the negative battery post. Distance
from the sensor to a LED indicator would be around 15 feet. I don't think
there is a need to have exact temperatures. Likely a range within 10 degrees
will suffice.

I would like to have the output from the temperature sensor light up a red
LED when the temperature exceeds 130° Fahrenheit in place of the voltmeter
in the first circuit (1). Since the first circuit is supposed to scale the
output and provide 10 mV/degree, could a proper comparator circuit be made
to indicate when the voltage exceeds 1300 mV or 1.3V? (Circuit B)

I have no clue how to calculate what resistor values would be needed in the
comparator circuit (2). I am assuming there will need to be a trimmer to
vary when the LED would turn on, maybe at V - REFERENCE? Would I need to
supply the comparator voltage from a different source and what would that
be?

Any help is certainly appreciated; Don

2. ### James ThompsonGuest

....
Look into the lm34 temp sensor ic's, they look like a transistor. Its
output is a voltage that is linear to temp change.

3. ### Andrew HolmeGuest

One input of the comparator can be connected to where the voltmeter is in
the other circuit; and the other input can be connected to the wiper of a
trimmer. The trimmer can be connected between the power supply rails (like
RV), or you could connect the top end of the trimmer to the +5V regulated
output from the LM7805. The trimmer could be anything from 10K to 100K;
it's not critical. You might also connect a large resistor (>1M) between
the comparator output and the (+) input to provide positive feedback for
hysteresis.

4. ### dwbauerGuest

and give it a try.
Thanks Again; Don

5. ### John FieldsGuest

Try this: (view in Courier)

+12V>--+---[78L05]--+-------+----------[750R]--+
| | | | |
[0.33µF] | [0.1µF] | |
| | | | |
+------+-----+ | |
| | [HLMP4700]
| | |
| | |
| +------+-------+ |
| | | | |
| | | +---|--[1M]--+-[750R]--+
| [3740R] | | | |
| | | | | |
| +------|---+--|+\ |
| | | | >------+
| | [LM34]---|-/ U1A
| | | |
| [1000R] | +-------+
| | | | |
| | | +--|+\ |
| | | | | >-+
GND>---+-----+------+-------+--|-/ U1B
LM393

Use the 12V from the battery to run it, wire it up on a little piece
of perfboard and keep everything pretty close together. Instead of
running a 15 foot set of leads to the sensor and building everything
around the LED, wire everything around the sensor and run the leads
to the LED.

I'd use a twisted shielded pair for the LED. with the shield
grounded.

6. ### John FieldsGuest

---
Better yet:

+12V>--+---[78L05]--+-------+--------------------/ /--+
| | | | |A
[0.33µF] | [0.1µF] | [HLMP4700]
| | | | |
+------+-----+ | +------+---/ /---+
| | | |+
| | [1K5] [1µF]
| | | |
| +------+-------+ | GND
| | | | |
| | | +---|--[1M]--+
| [3740R] | | | |
| | | | | |
| +------|---+--|+\ |
| | | | >------+
| | [LM34]---|-/ U1A
| | | |
| [1000R] | +-------+
| | | | |
| | | +--|+\ |
| | | | | >-+
GND>---+-----+------+-------+--|-/ U1B
LM393

7. ### Anthony FremontGuest

AIUI the OP wants it to trip at 130F. Wouldn't your divider be a bit
low (I calculate ~1.055V) or does leakage thru the LED pull it up?
Couldn't the OP use a 5K linear pot in place of the 3.74K resistor to
fine tune the trip point?

8. ### John FieldsGuest

---
No, you're right. It should be 1300 ohms. I did the math properly
but transcribed the 1300 onto the post as 1000. Good catch, thanks!

Yes. Matter of fact, he could replace both resistors with a 5k pot.
That might even be preferable if he wanted to make the trip point

+12V>--+---[78L05]--+-------+--------------------/ /--+
| | | | |A
[0.33µF] | [0.1µF] | [HLMP4700]
| | | | |
+------+-----+ | +------+---/ /---+
| | | |+
| | [1K5] [1µF]
| | | |
| +------+-------+ | GND
| | | | |
| | | +---|--[1M]--+
| | | | | |
| | | | | |
| [5kR]<---|---+--|+\ |
| | | | >------+
| | [LM34]---|-/ U1A
| | | |
| | | +-------+
| | | | |
| | | +--|+\ |
| | | | | >-+
GND>---+-----+------+-------+--|-/ U1B
LM393

It should be sealed, though, since it's probably going to live
pretty close to H2SO4 vapor. :-(

9. ### Anthony FremontGuest

No biggy.
I was playing around and with 5% resistors, the trip point could vary
between ~120F - ~139F. Sounds like some kind of adjustment would be a
good idea. 1% resistors would narrow that down to something that's
probably negligible (127F - 131F).
Not a good place to live. Is there any worse place (on Earth) for
electronics than in an automotive environment?

10. ### John FieldsGuest

The 78L05 has an output voltage which can vary from between 4.75V to
5.25V, so with an output voltage of 4.75V and worst case
resistances for the 3740 ohm and 1300 ohm 1% resistors for a low
reference out we have:

4.75V
|
[3774R]
|
+----1.208V ~ 121F
|
[1287R]
|
GND

At the high end we have:

5.25V
|
[3737R]
|
+----1.365V ~ 137F
|
[1313R]
|
GND

Which with the OP's request for a 10% machine would be fine.

Unless he actually meant +/- 5%, in which case the pot would be
needed to adjust out the variation in the 78L05's output.

11. ### Anthony FremontGuest

"John Fields
Oops, I didn't even think about the voltage variation.
Ah, I've heard it's nasty in there.

12. ### dwbauerGuest

Thanks for all the expertise.
Actually, this will be used to monitor charging on a large 12 volt 1200 amp
hour battery bank with 150 amp alternator and a 120V 70 amp charger when on
shore power. Realizing that we don't get all those amps in reality but it's
a start.
I'll get busy and order parts.
Thanks again for everyone's help and the time for circuit drawing; Don