Window Comparator Has Strange Outputs

[Scr1bbl3]

Hi all,
Found this windows comparator on the interweb and decided to implement it for giggles. Simple premise but it has some odd outputs. The voltage divider resistors are all 10K and the pullup is 5.1K so the trip points should be 1.66V and 3.33 volts with 5Vcc. I am using a BA4560P comparator (dual comparators)
When I am below and above the trip points of 1.66 and 3.33, I get 3V at the output.
Between 1.66 and 3.33, I get 5.
Can anyone explain this?
Thanks, Scrib

[John B]

BA4560P is a an opamp, not a comparator. It will not be able to swing completely to the supply rails and a pullup resistor isn't doing anything useful in this situation.

Try using a basic comparator like the LM393.

[wasedadoc]

Your comparators are not "open collector".  The 3 Volt output is when you have one pulling high and the other pulling low. A quick fix is to insert a diode on each comparator output before the pull up resistor.  Cathode to comparator output.  Anodes commoned and connected to pullup resistor. Or use open collector types such as suggested LM393.

[Scr1bbl3]

Thanks John,
I can't believe I did that. I just pulled the bag out of the drawer labeled comparator and just popped it in. Don't trust the labels and double check the part number.

[Scr1bbl3]

That'll do it - read the damn data sheet, right! Thanks for the feedback.

[Zero999]

Add a small diode, say 1N4148, in series with each output, with the anodes connected to R_L.

[MrAl]

Hi,

Adding a diode may not get the output close enough to 0v.

[Zero999]

Hi,

Adding a diode may not get the output close enough to 0v.

Define close enough?

The BA4560 is a bad choice. Its output will only be just under a volt or so of the negative rail to start with, so the low output will be around 1.5V with the diode and 10k pull-up.