Please help as I am running out of option (My first PCB doesn't work)

Thread Starter

Khisraw

Joined Nov 14, 2020
56
Hi All,
I am posting this hoping someone can help. It is my first PCB.

I have designed this circuit below. On the breadboard, it works as expected however after I manufactured the board and soldering the components, it doesn't work. There are two sensors on the board PIR and proximity sensor. The proximity sensor works fine but the PIR sensor doesn't work. I have checked every connected more than 10 times but can't figure out where it is going wrong.

  1. The PCB has two layers, the top and bottom layers are GND.
  2. U1 is LM290
  3. On pin 7 of U1 I get a different value to pin5 of MCU

The rest all seems fine. Can you take a look and see if I have missed something? Or any suggestion for testing?

Thanks in advance.
 

Attachments

dl324

Joined Mar 30, 2015
16,846
Since you have a working prototype, it should be straightforward to troubleshoot. If you no longer have the breadboard, troubleshoot using the typical methods; or recreate the prototype.

In any case, a schematic would be helpful. Along with a better description of what isn't working.
 

Thread Starter

Khisraw

Joined Nov 14, 2020
56
roubleshoot using the ty
I attached the schematic below. The problem is that the value in Pin7 of U1 and Pin5 of MCU should change as the PIR detects an object but the value of these two pins reads as(Pin7 of U1 ~330mV and Pin5 of MCU as ~290mV).
 

Thread Starter

Khisraw

Joined Nov 14, 2020
56
Hello there
:)
at a glance of your schematic Your LEDs backwards if breadboard is true to the schematic it never worked.
Thanks, that section doesn't have an issue. On the PCB, I have corrected the problem, and works fine but the PIR is the problem. It doesn't detect anything :(. On the breadboard, it functions perfectly.
 

Audioguru again

Joined Oct 21, 2019
6,674
Somewhere in this thread you said the opamps are LM290 which do not exist. You mean LM2904 which is almost the same as a low grade LM358. It is too noisy for your circuit that has a gain of 10000 times. PIR circuits in Google use a low noise dual opamp which are biased at half the supply voltage.
Oh, your schematic has no supply voltage.
 

MisterBill2

Joined Jan 23, 2018
18,176
The diagnostic procedure would be to start checking the PIR section at the output of the PIR sensor, and following the signal towards the MCU connection.
What I offer as a guess is that one of the capacitor connections that should not be tied to common has accidentally been tied to the common at one of the solder points.
Did the TS design this circuit, or was it designed by another? The one who inderstands how the circuit works should be able to verify quite rapidly where the flaw is located.
Another method, if a bare board is available, would be to use an ohm meter to verify the connections match the net-list, and that nothing is grounded, (tied to common) that should not be.
Now I am guessing that the circuit designer and board designer are not the same person.
 

MisterBill2

Joined Jan 23, 2018
18,176
The problem is in the circuit area between the PIR sensor and where the signals join. So at this point the TS needs to do some signal tracing. I already presented suggestions. Check the supply voltage at the PIR device and then look at the output signal from the device. If the TS does not understand the circuit then there is a big problem. Somebody understood it well enough to design the circuit, but evidently there is either an error on the PCB, or a failed component, or a wrong component. With no feedback as to the capabilities for testing and checking the best is to wish a lot.
 
Top