DSSR20 failure/intermittency

[Nicky Ivy]

My setup has 2 DSSR20s, in parallel. Peak current through each is about 15A. I've lately been noticing one of them not switching on and pushing current through, no red light. I tried switching the panels that feed each DSSR, and both fed power through the good DSSR, but not the bad one. I tried redoing some wiring, which had no effect. I switched out the bad DSSR for my spare, and now it's working on both DSSRs.

What is the most likely culprit here? Nothing visually looks wrong with the board. I can try measuring certain things, or touching up certain components, or even reflowing the whole board if that's safe to do.

[Dacian Todea]

Are you sure the DSSR did not worked and was just the LED that was not illuminated ? It is possible that is a bad solder joint on the LED and in that case the DSSR20 will work but there will be no indication. If LED is not the problem and so no charge current goes through the DSSR20 then most likely there is a bad solder joint of the small 4 pin device that is close to the LED. 

A closeup photo of the board my help to see if that is the case.

The LED and that 4 pin device are the only two parts that can be responsible for not having an LED illuminated.

[Nicky Ivy]

The SSR definitely would not deliver power. I did take a close look at the DSSR itself - the solder on the LED looks a bit marginal. I tried touching up the solder on the LED and reconnecting the suspicious DSSR. It initially delivered the same power as when the panels were plugged into the other SSR, (the LED was on but very dim), then a few minutes later stopped delivering power. Every few minutes the DSSR gets some power for a second or two but still does not continuously deliver power (see graph screenshot I attached)

[Nicky Ivy]

Realized after I posted that the graph I posted is mislabeled power, it's showing current.

Here are links to photos of the DSSR itself, since groups didn't seem to like me attaching directly.
https://drive.google.com/file/d/12L0xofWE8z3p2LIQ_X5zQij7mrYpCdMh/view?usp=sharing

https://drive.google.com/file/d/1DZ4V3Joxovpe_9wsWAj8R87PIGeHzlmv/view?usp=sharing

https://drive.google.com/file/d/1aUZXQw0eGU0IR2aCOpWCm2_t-tUO5Dzz/view?usp=sharing

[Dacian Todea]

It makes no sense for the LED to be dim. LED can only be ON at the designed brightness or OFF since LED is powered through a resistor from a fixed 5V supply and if the supply is not 5V then LED will not be allowed to be ON.

Is the DSSR20 installed inside or outside ? any chance water or condensation got to it ?

Maybe is because of the litign and angle the photo is made at but take a closer look at the two 5.1kOhm resistors that are connected in series (they are in the opposite corner of where the LED is) and they just look like both are soldered on only one side but may be just the shade and light from the photo. They power the entire circuit so if one of them is not soldered or makes just intermittent contact that may also be the problem.

During production I only see the LED that may not solder properly for less than 1% of the boards and I need to manually correct that but never seen any of the resistors not soldered properly before so maybe is just shade.

If that is not the problem I will ask you to send it to me to check in to more details and I will pay for shipping costs.

[Nicky Ivy]

I did some more touching up on the LED soldering and now it seems to be working fine, for hours since installing it. But that was the only thing I changed between it being intermittent and then it seemingly working fine. Is it plausible that the LED not soldered properly causes the rest of the board to not work properly? I would think that wouldn't be possible but you never know.

I'm going to leave this board in the loop for now for the next couple of days and see if it is actually fixed.

[Dacian Todea]

No the LED will just not work but not influence the rest of the circuit. So charging will still have worked even without LED being soldered.