Troubleshooting PCB with LT1017I comparator gives unexpected output

Thread Starter

jipristt

Joined Jun 4, 2017
20
Hello,

I have a rather simple circuit that became quite a puzzle when printed on a PCB and after spending lot's of time to figure out what is wrong I really got obsessed and I'd like to ask for some help to figure it out.

Here is my circuit, is an LTC1017I comparator that takes 2 input voltages and produces an EN signal of 5V if both voltages are disconnected or less than 5V. Vps, Vsw and Vbat are all on 16.8V.

When tested on bread board the circuit works as expected!

upload_2019-7-19_12-13-7.png

After breadboard testing, I made the following PCB but I keep getting a 5V output on EN pin. Also, don't mind the components on the right side, I did not solder them during the troubleshooting, but if you are really curious about the whole board I can share the project with you.

While testing I tried cutting the EN trace between the two leads of U1 (OUTA and OUTB signals of the IC) and also disconnected R4 and measurements show that OUTA is always at 5V even when Vps value is at 16.8V while OUTB performs as expected.

Further, some extra info, there are no shorts with 5V plane to EN pin, continuity test verifies that all components are connected correctly without shorts and get the right voltages.

If you have any ideas on what might be the problem or something that I could test please let me know!

Thanks,
Christos

upload_2019-7-19_12-18-15.png
upload_2019-7-19_12-21-38.png
 

kubeek

Joined Sep 20, 2005
5,794
Even though the datasheet doesn't explicitly state that (which is odd), you should not be applying any voltage higher than the supply voltage directly to the input pins. You need at least a resistor in series with the input, for current limiting and to make a voltage divider.
Your comparator is most likely destroyed now.
 

Thread Starter

jipristt

Joined Jun 4, 2017
20
I altered the board to give 16.8V on the supply and now it works without any issues. Thanks for the suggestion!

But there is one question remaining, why the same circuit on the breadboard works with the 5V supply?
 

TeeKay6

Joined Apr 20, 2019
573
I altered the board to give 16.8V on the supply and now it works without any issues. Thanks for the suggestion!

But there is one question remaining, why the same circuit on the breadboard works with the 5V supply?
@jipristt
@kubeek is certainly correct; unless the datasheet specifically says an input may exceed the power supply voltage, you can very safely assume that it should not...ever. Perhaps this info would be useful to you: Simulators use models that do not always mimic the real device. Models can have errors; models can fail to implement certain features of a device (e.g. input voltage limits)--models are a simplified view of the real device. Simulation is a wonderful and powerful tool, but it is not perfect and it is not a substitute for clear thinking, nor a substitute for understanding your circuit, nor a substitute for close reading of a datasheet. Blindly assuming simulations are perfect can get you into a lot of trouble. As for your breadboard working...sometimes devices will operate--at least for awhile--under conditions far outside of their specifications. However, to rely on such operation is foolhardy; only operation within the spec limits is guaranteed.
 
Top