I have a project that needs thermal monitoring in two places. When the temp(s) reach the set point, it needs to drive a high signal into the shutdown pin-10 of an SG3525 PWM which is currently held low with a 1k resistor to ground.
A comparator will do the job (and also gives me hysteresis) but how would I go about driving a single comparator input with two thermistors? Can I simply place a diode on each thermistor? Or should I use a dual-comparator and parallel the outputs?
A comparator will do the job (and also gives me hysteresis) but how would I go about driving a single comparator input with two thermistors? Can I simply place a diode on each thermistor? Or should I use a dual-comparator and parallel the outputs?
On my irs2092 based class d amplifier I have PIC micro doing house keeping functions.
It holds off 2092 on power up until VCC is valid.
If VCC goes below threshold it holds the 2092 in reset through an opto-coupler.
I also have a PIC monitoring output signal. If it sees DC for more than 500mS it disconnects the speaker relay.
This could be expanded to monitor temperature too.
Some of the small PIC's have A2D too.
It holds off 2092 on power up until VCC is valid.
If VCC goes below threshold it holds the 2092 in reset through an opto-coupler.
I also have a PIC monitoring output signal. If it sees DC for more than 500mS it disconnects the speaker relay.
This could be expanded to monitor temperature too.
Some of the small PIC's have A2D too.
I know that paralleling them won't work. Here's why.
Condition one:
NTC 1 - 10k (ambient)
NTC 2 - 4k@50c (reached trip point)
Ohms total = 2,857 (set comparator for this value)
Condition two:
NTC 1 = 6k (38c, OK!)
NTC 2 = 5k (45c, OK!)
Ohms = 2,727 (comparator tripped)
In the second example, neither source of heat reached the threshold. Yet the thermal protection was activated.
It may work with a diode in series with each thermistor but I'm not sure.
Last edited:
I'm one of those people that hates software like a tube head hates silicon. I know an MCU would be capable but I don't have any experience with them nor do I want any. lol Thanks though.
I assumed ambient temperature was the same for both NTCs. If not , yes, you are right.
TI temperature switch.
TMP709 Temperature Switch - Texas Instruments | DigiKey
http://www.digikey.com/product-detail/en/TMP709AIDBVT/296-30279-1-ND/3060756
Data sheet
http://www.ti.com/lit/ds/symlink/tmp709.pdf
TMP709 Temperature Switch - Texas Instruments | DigiKey
http://www.digikey.com/product-detail/en/TMP709AIDBVT/296-30279-1-ND/3060756
Data sheet
http://www.ti.com/lit/ds/symlink/tmp709.pdf
Last edited:
Quad comparator costs almost the same as dual (or single) comparator, so you can use one comparator per sensor, use the open collector outputs for OR'ing, and another one to invert since it seems you need it. With the remaining one, you can do something like a power good signal or just leave it alone.
What's OR'ing?
If I use a dual-comparator, Can I use a single voltage reference for both and have either output shift the reference? I really need to fiddle with this in simulation to see what's going on. I get lost trying to imagine everything at once. I'm horrible at chess.
I'll doodle up something tomorrow and post for review.
Those comparators have open collector outputs. So it it outputs a zero, the output is pulled to GND, and if it outputs a 1, it is left floating. So you can wire the outputs together with a pullup resistor.
If one or both comparator outputs a zero, the common output is pulled down to GND. That's an OR function.
If both comparators output a 1, then the common output is VCC. That's an AND function.
Depending on the way you wire the + and - inputs you can do all sorts of logic.
If you want to enable the device only if the temperature of both heatsinks is below threshold, this means :
(Temp1 < Th) AND (Temp2 < Th)
So you wire one comparator per sensor, and set them to output a 1 if the temperature is < threshold and a zero when overheating. Connect both outputs together, add pullup resistor.
So if one overheats, the common output is zero. If both are OK, output is VCC.
Next you add another comparator to invert that since it seems your PWM chip has active high shutdown.
That's what I figured. Op-amps were the dawn of the (practical) computer age.
I might be able to have either output kill the oscillator without the need for the extra inverting stage.This.
Don't know what format that is.
Last edited:
Thank you nec3. That's exactly what I was wondering. I'm very limited on space and that would be ideal.
Now I need to figure out how to drive the shutdown pin or if I can connect one of the error amplifier inputs to the comparator output so that a normally open output doesn't interfere with the SG3525. If I use the shutdown pin, I'll still need to buffer the signal since the comparator uses feedback for hysteresis and when the output is open there is still a voltage present.
Now I need to figure out how to drive the shutdown pin or if I can connect one of the error amplifier inputs to the comparator output so that a normally open output doesn't interfere with the SG3525. If I use the shutdown pin, I'll still need to buffer the signal since the comparator uses feedback for hysteresis and when the output is open there is still a voltage present.
Last edited:
As i can see in the data sheet, Softstart, Comp, Shutdown, EA input all may be used.
When use Shutdown pin, which seems easiest to me, i think, you will not need a buffer. Just design comparator as to give when T=High>> OUTPUT=1; when T=LOW>> Output= 0 (less than 0.6V) and connect its output to Pin:10 directly. There is a 5 K inside IC at Pin10.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.