single op-amp window comparator

with a rectifier in front of the comparator you could do that, but you loose the ability to randomly choose the thresholds.

--
ciao Ban
Bordighera, Italy

Yup, you can do it with a couple of extra diodes. It's not as accurate ( and needs to oversome the Vf of the diodes ) but may suffice. Don't have a schematic handy though.

Graham

Not really. You loose a lot of accuracy because of the diode drops but for moderatly large values of random, you can get any values you want like this:

If you make this circuit and its top to bottom flipped version, the Vx voltages are determined by R1 and R2 or Vin depending on Vin.

! / \ R1 / \ ! Vin -!

you loose

but

like

Vx

whatever

center

The inputs to the opamp go where? Vx is?

NT

but

like

Vx

! / \ R1 / \ ! Vin -!Further resistor dividers on the Vx and Vx' signals can give you whatever

center

i cant seem to understand the circuit...where do i connect the positive and negative terminal of the op-amp to this circuit...and does this apply to any comparator op-amp.tnx

[....]

The top to bottom mirror image goes to the other op-amp input.

--
--
kensmith@rahul.net   forging knowledge

Modified drawing

**************** Mirror image along here **************

You make a mirror image along the horz line.

R1 and R2 are much lower values than R3 and R4

--
--
kensmith@rahul.net   forging knowledge

thanks Ken, wasnt familiar with that one.

NT

LTC1042

Modified drawing

You make a mirror image along the horz line.

R1 and R2 are much lower values than R3 and R4

--
--
kensmith@rahul.net   forging knowledge

what if i would to have a window comparator for 3-6 volts how can i
compute for the values of the resistors

a simple hack version comes to mind.

using an OP-AMp from a +&- rail supply to get a

0 volt output into a mini bridge rectifier to convert the signal to always + or - what ever.

by setting setting one input of the Op-amp to a voltage and using the other to monitor the external source voltage, one can then adjust the gain using a loop back into the - input and adjust the + input for the comparator center point. the idea is to produce no voltage or low voltage at the output as the reference voltage starts to get near the set voltage.. basicly we are looking at a simple OP-amp circuit that you can set the + input using a dual rail supply.

the output can drive some schmitt trigger or mechanical relay etc..

like i said, just a hack in thinking of course.

do you have any example...i really need a circuit for this...tnx

Lee, you should ask for a single resistor solution. :-)

Rene

Thats easy:

T \\---------------------- ---O ! --------------------------- Vin ------/\\/\\/\\----VMiddle

When the voltage is far enough from Vmiddle to be either voltage, it heats up the thermstat and it closes.

--
--
kensmith@rahul.net   forging knowledge

[...]

I think the reason you are having trouble is that several of the values are just "pick one" so you don't see an obvious "this resistor has to be

10K".

Assume R1 is 10K.

Assume R3+R4 is 100K.

Find R2 such that D1 is carrying the same current as D2 with the input voltage at the positive trip point.

Do the same for the mirror image version.

Calculate the voltage at the junction of R2 and R3 (and R2' and R3') at the two trip points.

Now you can write equations for what the inputs of the op-amps will be in each case.

The two inputs to the op-amp just become equal when the voltage passes the trip point.

This means that you end up with two equations that have R3, R4, R3' and R4' in them.

You also have R3 = 100K-R4 and R3' = 100K-R4'

This means you have 4 equations and 4 variables. If the gods are smiling on you, there is a solution that does not require negitive resistors.

--
--
kensmith@rahul.net   forging knowledge