SI1145 Corrosion

For other supported Arduino products from Adafruit: Shields, accessories, etc.

Moderators: adafruit_support_bill, adafruit

Please be positive and constructive with your questions and comments.
Locked
User avatar
primexandy
 
Posts: 89
Joined: Sat Sep 06, 2014 11:09 am

SI1145 Corrosion

Post by primexandy »

Hi,

I have had an SI1145 sensor within a plastic box outside for 6 months now and it has just stopped working. I think it is down to some sort of corrosion / salts that has built up all over the solder joints...literally everywhere there is a white power. The board hasn't got wet but it will be down to the moisture in the air.

Is there something that I can spray the board with to protect it (obviously not over the sensor)?

Edit - it is causing a short somewhere which pulls down the 3v to 0.5v (over a 3 meter cable). I will of course take it down and try and clean it but would prefer it not to happen

User avatar
adafruit_support_bill
 
Posts: 88088
Joined: Sat Feb 07, 2009 10:11 am

Re: SI1145 Corrosion

Post by adafruit_support_bill »

Can you post a photo? Galvanic corrosion usually involves a combination of dissimilar metals and moisture. What kind of wire and solder were you using? Lead, tin and copper are relatively close in the galvanic series, so common soldered joints are usually fairly stable.

Flux residue can make a joint more prone to corrosion. Cleaning after soldering should help.

User avatar
primexandy
 
Posts: 89
Joined: Sat Sep 06, 2014 11:09 am

Re: SI1145 Corrosion

Post by primexandy »

I'll have to post a photo tomorrow as it is on a roof outside (and it is dark in uk). The white powdery stuff is all over "your joints" not mine. It's the default solder. My solder joints have the flux on and aren't bad. It is all other joints to all components (ie, nothing to do with me / my solder). The sensor is in a simple plastic box and hasn't got wet (no rain in). It has obviously been subjected to normal outdoor air, which is the only possible moisture it has seen.

I have ordered another one so I can just swap asap.

It is possible that the white powder isn't causing the short and that a component has failed but I'll have a look when it is day light when I can take it down.

The SI1145 board is soldered via the header pins to a breadboard. The breadboard it then soldered to the thickest copper mains cable wires I could find (to reduce resistance). The solder I used is just "standard" stuff in a little tube from the local hardware store and doesn't say what the lead/tin composition is.

Anyway, we can wait until I get a photo for you....

User avatar
adafruit_support_bill
 
Posts: 88088
Joined: Sat Feb 07, 2009 10:11 am

Re: SI1145 Corrosion

Post by adafruit_support_bill »

Hmmm. I've heard of one case of strange corrosion a year or two ago. It will be interesting to see if yours appear similar.

I don't believe there were any shorts or other performance problems in the previous case. Just white powdery corrosion.

User avatar
primexandy
 
Posts: 89
Joined: Sat Sep 06, 2014 11:09 am

Re: SI1145 Corrosion

Post by primexandy »

Below is the photo I took...had started to desolder the headers and then remembered the photo...

Image

Above shows 6 months of being outdoors in a plastic "take away" container. No water has ever got in.

I have now replaced the board and the new one fired up perfectly. This board didn't show up on the i2cdetect tool of the raspberry pi which was caused by the 3.3v being pulled down. (the 3.3v wasn't taken from the pi's 3.3v rail, it was taken from the 5v rail and a resistor divider was used to produce 3.3v)

The white powder is clearly shown. There was no white powder on my joints (even though they were messy :) ). Iterestingly the Macro shows the sensor chip very nicely

User avatar
adafruit_support_bill
 
Posts: 88088
Joined: Sat Feb 07, 2009 10:11 am

Re: SI1145 Corrosion

Post by adafruit_support_bill »

That looks like "creep-corrosion". It can occur without moisture, and can cause short circuits.
Lead-free soldering processes are more susceptible to this type of corrosion if exposed to sulfur in the atmosphere. An air-tight box might help. Not sure if there are any coatings that will prevent it. I'll see if any of the other engineers have any ideas on this.

User avatar
adafruit2
 
Posts: 22144
Joined: Fri Mar 11, 2005 7:36 pm

Re: SI1145 Corrosion

Post by adafruit2 »

try conformal coat. you'll have to apply it after soldering.
adafruit uses only RoHS processes, which is good for the earth but boy lead is an awesome stabilizer and is very much missed :)

User avatar
primexandy
 
Posts: 89
Joined: Sat Sep 06, 2014 11:09 am

Re: SI1145 Corrosion

Post by primexandy »

do you think the board will come back to life if cleaned? and if so what cleaner would you try. Money is an issue. The conformal spray looks like £10 a spray tin, a new SI1145 is £11 and an air tight box will be same (and more hassle and no guarantee it'll be air tight with cables etc and I'd have problems making a hole for the sensor). I'd prefer a cleaner if possible.

I might just continue as I am and hope the 2nd board will last another 6 months so I have a year's worth of data and then do a different sensor.

Next question, and probably more importantly. Since I am measuring the sun etc and want to keep the same box (since the clear plastic is reproducible for free each time I have a take away) will your other boards behave the same? The raspberry pi is also outside, ie in a damp shed under the sensor and it has no issues. Is it just your method of production at "fault" or have I just been unlucky with this particular board from this batch?

Would appreciate your thoughts as I don't want to keep seeing this problem.

User avatar
adafruit_support_bill
 
Posts: 88088
Joined: Sat Feb 07, 2009 10:11 am

Re: SI1145 Corrosion

Post by adafruit_support_bill »

From what I understand, creep corrosion can occur wherever there is any exposed copper on the board and sulfur in the air. Some types of 'no clean' soldering flux residue leave the surface more 'active' than others. Further cleaning of the boards may help.
Conformal coating seems to be the recommended preventative method.

This link implies that the corrosion itself is cleanable - but seems to be missing some important details on how: http://www.thefreelibrary.com/Is+creep+ ... 0212852929

User avatar
primexandy
 
Posts: 89
Joined: Sat Sep 06, 2014 11:09 am

Re: SI1145 Corrosion

Post by primexandy »

Thanks.

I'll try cleaning it with what ever the modern equiv of what I used to use was (trichloroethelene 1:1:1 spelling?) as that used to clean anything. I haven't used that for 18 years.

I will wait and see what this second board does and if it goes the same why I'll coat it next time.

Locked
Please be positive and constructive with your questions and comments.

Return to “Other Arduino products from Adafruit”