Op-amp comparator circuit for detecting voltage outside 2V-3V band with 5V supply

Hi again, happy New Year and all that...
I'd like a simple way to determine if a voltage  is outside of a certain band.
Say the band is from +2V to +3V.  I'd like a circuit with an output that goes high (5V) if the input voltage is =<2V or =>3V.
Ideally I'd like two separate outputs, one for =<2V and the other for =>3V.
OpAmp(s) will no doubt figure in the answer somewhere, but I get a bit confused when using them with a single (5V) supply.
I've googled, but there's a lot of noise caused by a BAND called HIGH VOLTAGE
Thank-you,,Graham.

You are looking for a "window detector". Most wire the comparator outputs together to detect within the band. However you can separate the outputs and invert the outputs directly, or invert the inputs of both comparators to provide the outputs you require. In reality by separating them, you are just looking for two independent comparators and you just need to decide if you want the output to be active low or active high for the detection that you want.

Sounds good, thanks v much.

Graham Aubrey beat me to it here, but I'd refer you to the excellent Electronics Tutorials site for their article on Comparators which includes a bit on Window Comparators (which is an alternative name for Window Detectors).You can use Op-Amps for this but a comparator is like a lower-gain op-amp and is usually a lot faster (this may be a factor in your requirements?). An ideal IC to use for a window comparator is the  LM393 dual comparator.  It can work from 2-36V power, and the inputs work rail to rail.Here's a circuit from the Electronics Tutorials article: The LM393 has an open collector output so you can wire the outputs together as shown above. You want to detect voltages between 2 and 3 volts so instead of 3 identical resistors R above, you'd use (for example) 2K for the bottom R, 1K for the middle R and 2K for the top R. The TI Data sheet above also shows what they call a "Limit Comparator" in Fig. 28.Your comment at the end is a bit cryptic (maybe you weren't using the right search term) but it prompts me to say that you need to protect your inputs against voltages greater than VCC.  If you're likely to encounter those, put a zener diode (say 5.2V) from your input to ground, and a resistor in series with the input  (maybe 1K or 10K - the input currents are in the range of nA so you can make this quite high).You say you want to detect if your input is OUTSIDE a certain band (=<2V or =>3V).  The above will detect if the voltage is inside the band.  to make it detect the other way round (outside) just swap the + and - inputs of each comparator, or put a logic inverter gate on the output.It's a fairly simple circuit, and one which every electronics person should have in their arsenal, but if you're not clear on anything let us know.

That's really great,  thanks again to both of you. Those links are excellent too. 
I thought the solution would probably be something simple and elegant, which it is.
My cryptic remark re googling was meant to be an eye roll at the musical band 'High Voltage' dominating the search results. Annoyingly, "band", "high" and "voltage" are similar to the keywords I was searching with! :)Thanks a lot,Graham.

One additional thing not widely known about the LM339/LM393 comparators. The output (even though it is open collector) is current limited so that you can drive a large capacitor directly without a current limiting resistor. It doesn't sound like Graham is trying to do this, but for the general edification of the viewing public, it is handy to know. 

Hi.  I'm using an MCP602, which I happen to have, and in spite of not having open collectors it still drives a BD139 in OR fashion through a pair of 470R resistors.But the problem's not there.  Below is what I'm using for testing.
...which I hope is similar enough to the circuit given above.  
I thought that, the gains of the opamps being very high, turning one of the 10K presets would at a certain point switch the relevant LED full on or full off.Unfortunately not!  As the preset is turned to near the "turn off" point, the LED dims, flickers, then flashes fast and dim before extinguishing.
(Also, I'm using a metal screwdriver on the preset, and surprisingly (?) (as it's all low impedance) when it is touched to the turning bit, that affects the LEDs too.)OpAmp B seems worse than A for this, for some reason, on two different chips.
I didn't bother putting the caps on the second preset, as it didn't make any difference to the first one if they were there or not.Using a switch mode adaptor or bench psu, the result is the same.There's obviously yet more that I'm not getting.
Thank-you for any help.Graham.

GrahamFrom your description it seems to me that they are oscillating.  It is possible that your implementation just oscillates because of the layout or decoupling, but then it would probably oscillate  for other input voltages as well. It would help if you could look at the output with an oscilloscope.That being said, I think the effect is as a result of using an op-amp and the cure is probably to add hysteresis to each op-amp. Try googling  “op-amp comparator hysteresis”.This site looks promisinghttps://www.electronics-tutorials.ws/opamp/op-amp-comparator.html
Because hysterics works by adding or subtracting a little voltage to your reference voltage, you may find that it interacts with the resistance divider of the pot. If so you may want to buffer the voltage derived from the pot with a unity gain buffer. 

Hi, ok, thanks again for all the help.I get it now about hysteresis.  If you're effectively trying to have zero hysteresis, as I was, I see that it's going to cause all sorts of problems.But regrettably I'm STILL a bit confused about the suggested LM393. Having read the datasheet, I still can't really see what (apart from the open collector) it does that an op-amp wouldn't, and vice versa?

PS  Don't forum threads normally get bumped up to the top when a posting is added?

Graham - OK, an op-amp has very high gain but is relatively slow.  A comparator has far less gain but is very fast compared to an op-amp.  Probably not a big deal for your application though.  A comparator is designed for comparing levels and giving an on-off indication (though they do have a linear region as you've found).  They also often have an open-collector output designed for wire-or applications so you can have several comparators driving one output - not something you can do with op-amps without using extra diodes.  Comparators are usually designed for single-supply and can often work down to lower supply voltages than op-amps.I found a good article on the differences here.You can use an op-amp as a comparator, they're perfectly adequate for a lot of applications, but you can't really use a comparator as an op-amp.re your observed jitter on the output, when you adjust the presets you're probably introducing mains hum voltages which (near the trigger point) will cause on-off triggering.  As Aubrey says, Hysteresis is the way to overcome this.

No, unfortunately.  Once you've had your 15 minutes of fame, that's it!   I (and other experts) will often check back on posts we've responded to to see if there have been updates.  Looking in the recent comments often helps as well.

It also helps if you click on “Reply” rather than a plain message. The poster of that response will get a notification, just like you should from this response.

Spider-Gwen Costume. Tons of Spider-man fans have a love for the ongoing comic book series, Spider-Gwen, which features an alternate world in which Gwen Stacey, the neighbor of Peter Parker, is bitten by a radioactive spider and dons the suit to become Spider-Woman.ene