Radion G4 went swimming ! - Need advise.

[GregHeath]

I recently bought a known Radion G4 sold as faulty as it went swimming in a marine tank, I spent £70 so not overly concerned if its blown beyond repair but this has me stumped, I only wanted to use it for a small frag tank I am setting up, the rest of my setup also runs on Radions so these are not new to me.

I don't know if the unit was on or not when it went swimming.

Symptoms: Initially no power at all. After clean with contact cleaner and a light brush initial red colours on power indicators for 1/2 second then solid green lights on the power indicators, LED's do not light and not detected on USB.

The positive:

The power supply works and provides 36.1 volts as expected.
The driver board had no salt creep and is generally very clean.
The LED's all work when powered individually with a 12v DC supply.
No components missing or loose.
All resistors check out OK, get identical readings across components of same value.
All transistors check out OK, get identical readings across components of same value.
All diodes check out OK, get identical readings across components of same value.

The negative:

The LED boards had minor salt creep on the connectors and on one of the small resistors but it all cleaned up nicely.
I can't test / am unsure how to test the IC's.
When testing the contacts on the switches I don't seem to be able to get conductivity even after multiple cleans.
Fuse was loose but intact, pinched the connectors and now sits fine.
There is a single click from what I can only assume is a relay but without any markings when power is on.

So as you can tell, I am pretty stuck, could use some advise on next testing steps, if FixReef is here as I have seen some of his posts then his advise would be appreciated. Below are the photos.

 

[_AV]

Just from the pictures posted I can see that at least 20% of the pins on the LED cluster connectors no longer have any solder on them. Which is likely due to corrosion. I'd start there.
The solid green status LED supports the theory too.

 

[AKG]

Op you would do well to find someone good at soldering with a nice fine soldering iron and go back over the board heating up the solder and resoldering then checking again after the board has spent a day in a food dehydrator set on low. Imo.

 

[GregHeath]

Just from the pictures posted I can see that at least 20% of the pins on the LED cluster connectors no longer have any solder on them. Which is likely due to corrosion. I'd start there.
The solid green status LED supports the theory too.

Thanks for your input, I thought that too when I first looked at it but then I realised not all the contacts have tracks and I put that down to the fact the clusters aren't fully populated, so maybe it was penny pinching on behalf of Ecotech.... but I can reflow those solder joints myself and so I'll give it a shot.

I'll come back in a few days with some results.

Cheers,
Greg

 

[_AV]

From the pictures I can tell there was already an attempt to reflow some pins, but clearly not all. The oxidation layer on the pin and the pad is significant at this point, solder won't stick without prior cleaning. Even if it looks like solder is covering the surface, the contact is not going to be solid. Without a solid contact, it won't be able to handle the current needed to run those LEDs at full intensity.
You can also see that most of the test pads around the connector have a very dull copper color. That's the sign of oxidation caused by exposure to water. You would expect the same state for the pins and the pads next to them.
The main board seems to be in a really good shape. I rarely see the USB connector this shiny on a G3 light.

 

[GregHeath]

Quick update on this.... I really struggled to get anything to remove any oxidation from these pads and pins, tried baking soda and distilled water, contact cleaner, isopropanol, I even tried a gentle going over with my dremel but nothing removed it.

I decided to look under the hood and I removed a tiny section covering one of the pin tracks and discovered that the tracks are all laid in what appears to be copper and so by my understanding this would be the natural colour anyway ?!?

I decided to have one last go at cleaning, this time I used vinegar and much to my surprise it started fizzing away almost immediately on the contacts, so I left them to soak submerged for 30 minutes and I did the same on the driver board (contacts only - not the board and no components).

Immediately I got some LEDs light up, and so rinse and repeat, last thing I did was tighten the pins slightly in the LED board so they grip better on the driver pin outs - I will assemble later and give it another test - fingers crossed.

oh.. and... anyone know where I can buy those connectors on the LED board - just in case I do need to reflow and I damage the connector?

Cheers,
Greg