FEDEVEL
Platform forum

Naveen-Krishnan , 08-16-2018, 08:56 AM

Hello everyone,

I am devasted after checking the PCB what I designed.

In my card there are 3 power supplies 5V, 3.3v and 4v. The 4V is short circuiting with Ground only on PCBs after soldering all the components. On the bare PCBs without any components there is no short circuits.

I thought there is some problem with the decoupling capacitors in 4V power supply, so I removed them. After removing I found that the short circuit started happening in 3.3V also, whereas before it was not present. I don't know what to do now. Can anyone help me out please

robertferanec , 08-16-2018, 11:30 AM

How do you power your board? Can you measure how big current flows into your board?

Naveen-Krishnan , 08-16-2018, 12:57 PM

Hi Robert Feranec,
Thanks for your reply. I powered through DC power supply.
Yes I can mesure it. In the DC power supply it was showing 150mA.

The bare PCB without components is not showing any short circuits, which means the design is correct ?

The problem could arise from bad or blown capacitors or soldering ? Am I right ?Please correct me if I m wrong.

Regards,
Naveen

mohsin_qau , 08-16-2018, 01:17 PM

Can you share your power schematics and PCB Power section layout?

MuhammedHesham , 08-16-2018, 05:51 PM

i met something like this .. i found out that anew component was damaged .. i bought another one and its works fine .. i don't know why short circuit can be made when removing decoupling capacitors in 3.3 V

robertferanec , 08-17-2018, 02:28 AM

- 150mA is not much, but still try to see if there is no component getting warm or hot (use your fingers or use alcohol - e.g. see Louis Rossmann's video https://www.youtube.com/watch?v=t5fICjcaJ3E )
- I am not sure how you measured the short circuit (e.g. if you used diode measurement or resistance measurement), but I would use DVM (Digital Voltmeter) to measure resistance between individual powers and GND. Be careful, sometimes it may look like a short circuit, but it doesnt have to be (e.g. if there are protection diodes - including the diodes inside chips, or some chips have very low impedance termination, ....). It may also depend how you connect DVM, try to swap the probes (if you measured VCC by RED probe and GND by BLACK, try to measure VCC by Black and GND by RED) - just to make sure, that what you are measuring is direct short circuit e.g. you measure resistance 0.1OHM
- another option is to check Power OK signal of the regulators (but this depends what kind of chips you are using, some chips do not have power ok pins)
- or I would start building another board, but step-by-step ... fit the first regulator, measure if everything is ok, then another one, another one ... etc

PS: The problem can be something completely different and not power supply (I am not sure what everything is on your board and what is connected to your board)

Naveen-Krishnan , 08-17-2018, 03:14 AM

Thanks everyone for your reply.

Robert Feranec, I will try with alchohol whether some components are heated or not.

I would like to add in some more information.

There are jumpers in my design, please see the picture attached :

- In the 3.8V the pad which is connected to the rest of the BOARD had a short circuit with ground (taken out of the box, without powering)

- In 3.3V the pad which is connected to the rest of the BOARD did not have any short circuit initially.

- I tested 3.3V by connecting the jumper in 3.3V.
1)I used DC power supply. My settings in the DC power supply was 12V and minimum current.
2) I connected the voltage input to the PCB 12V jack as you see in the above picture. Immediately the voltage dropped to 6.6V and the LEDs started blinking.
3) I increased the current in the DC power supply, so that i get the required voltage what I set initially.
4) There was a noise in the DC power supply, some kind of vibration and then the voltage increased to 12v. At this time the PCB was consuming 12v 150mA
5) All the LEDs was glowing, although I designed the LEDs to glow using transistor. I did not program the transistor whatsoever and it was glowing. I don't know for what reason the LEDs were glowing.
6) I checked the input core voltge, DDR voltage all were good at this time ( the inputs to this was 3.3V).

-I tried removing decoupling capacitors from 4V to check for shorts. But couldn't find any.

- At this step I connected 3.3V jumper and 4V jumper. Connected the DC power supply. Both 3.3V and 4V stopped working.

- I thought because there was a short in the 4V, the 3.3V power line is not working. Hence I removed the 4V jumper to check that 3.3V was working like before.

- But 3.3V stopped working, 3.3V was short circuitng with GND from now.

I stopped my tests and asked my manufacturer to get the board tested using X-rays to find if there is any short circuit.

Anyone please help me. I had a sleepless night yesterday.

Regards,
Naveen

robertferanec , 08-17-2018, 04:43 AM

Was your power supply working ok when both jumpers where not connected to the rest of your board?

I do not know what is on the rest of your board, but in some cases if not all powers are connected, some components can be damaged ... e.g. if you only connected 3V3 and not 4V, there could be current leaking between powered and not powered part of your board and this could possibly damage components ...

To damage a power supply should not be so simple. Try to leave 3.3V and 4V disconnected from the rest of the board and replace the power chip. See if the power will work.

PS: Maybe there is manufacturing issue. I can see solder balls on the PCB. Also, double check if pins of the power supply chip are correctly soldered (maybe try to go around of the chip with soldering iron and re-heat the pins).

Naveen-Krishnan , 08-17-2018, 06:36 AM

Thanks for your reply Robert Feranec.

Yes when the jumpers were not connected the Power supply was working perfectly.

I think the Power supply is working, why because only the 3.3V pad in the top, which supplies to the rest of the PCB alone is short circuiting.

Both the bottom pads of 3.3V and 4V are not short circuiting. When jumper not connected the 3.3V and 4V are present perfectly on the bottom pads which is the output of the Power supply.

I have given the boards to the manufacturer, they will return with the X-ray results and flying probe results. They told me that its easy to find a short circuit with these 2 tests.


I have some ideas please help me out whether they are feasible or not please
1. I have a 33pF decoupling capacitor at the load side(Top pad of the jumper) of 4V. Because of such a low capacitance there could be a short circuit ?

2. When I connected the DC power supply to board, while the 3.3V jumper is connected. The voltage in the Dc power supply dropped immediately to 6.6V, when I tried to increase there was vibration and ticking sound in the DC power supply. Could there be a short during manufacturing or bad component (transistor or capacitor)? Also during Power supply measuring 6.6V, the LED were blinking during the 1st test. But during 2nd test, 3rd test ..... the LED blinking went off.

Naveen-Krishnan , 08-20-2018, 06:30 AM

Hello everyone,

Got a lead for short circuit between 4V and GND. The footprint of the module is wrong.

There was some pins coming in random order. Please see the picture.




I did the footprint in standard pin order like QFN, SOP, etc. But I was wrong, my mistake. I have asked the manufacturer to put Kapton tape on the wrong pin and resolder it again.

Regarding the Short circuit in 3.3V no lead, will update you once I find. I think as Robert Feranec mentioned, because of the short in 4V there was a back flow leakage of current and it might had damaged some components.

Thanks.

Regards,
Naveen

Anyways thanks for your support.

robertferanec , 08-20-2018, 08:16 AM

Thank you for letting us know.

Naveen-Krishnan , 08-21-2018, 03:46 AM

Hello everyone,

My PCB arrived and 3.3V and 3.8V are working now on another PCB perfectly.

A component on the 3.3V line would have been damaged for sure on the other PCB where there was a short.

Hope this post could help someone !

Cheers ! Bye.

Regards,
Naveen