# Window comparator design using LTC1040

I am trying to use LTC1040 as a window comparator(Since I have it) with upper limit of 39.2V and lower limit as 24V. The problem is it has two op amps with 4 inputs each and I can't understand whether I can use any one op amp or should I use both. If it can be achieved with one op amp I would like to know how the circuit can be made.

Please let me know. It would be of great help. Thank you

• Reading the datasheet: "LTC1040 [...] Single Supply 2.8V to 16V / Split Supply ±2.8V to ±8V". So, your 39.2V upper limit seems a bit beyond the capabilities of this opamp, unless you reference its supplies to some other ground. – dim Jul 20 '18 at 11:32
• "unless you reference its supplies to some other ground" could you please tell me how I can do that? – Prem Arjun Jul 20 '18 at 11:46
• A comment won't allow me to explain it as it should. You could ask another question for this, specifically. But I don't think it would be a good idea anyway, because it would reduce the allowable input signal range (your signal won't be allowed to go down to 0V anymore). A better, simpler way would be to reduce the input signal range using a resistor divider. – dim Jul 20 '18 at 12:04
• ... But you need to take a step back and rethink about the whole solution: 1) Is this opamp really appropriate? This unusual 4-input topology and the specified input range makes it harder to use. 2) If LTC1040 is really appropriate, how to constrain the input signal range to what the opamp allows? What is the input range anyway? what supplies are available? What are you trying to do? Right now, this looks like an XY problem. – dim Jul 20 '18 at 12:05

You need 2 comparators and OR logic to make a window comparator. Both exist in the LTC1040 .

Power Supply and same CM limit must be <=16V with a suitable reference and both input and thresholds scaled into that range.

Although the dual differential Op Amps are redundant for this application of Vbat monitoring.

# All you need is a dual comparator.

and precision resistors to meet your tolerance error spec,.

simulate this circuit – Schematic created using CircuitLab

• don't forget hysteresis – JonRB Jul 20 '18 at 19:30
• Yes but it was not specified 1% positive feedback to both affects both levels so it could be 10M 1nF in series but some oscillation with input dynamics needs to be better Spec’d to minimize oscillations – Tony Stewart Sunnyskyguy EE75 Jul 20 '18 at 19:44
• Hello @TonyEErocketscientist I have few questions related to LTC1040. I have written an Email(which is in your profile) with the subject "Question on LTC1040 comparator" to you. Please respond to that. Thank you. – Prem Arjun Jul 23 '18 at 17:56
• Try to ask here instead – Tony Stewart Sunnyskyguy EE75 Jul 23 '18 at 18:44