Today we have a new subject to tear down- my own Sonicare toothbrush. At the end of a brushing cycle it made a short “Beep” noise and went silent. No reaction to button press or to putting it on a charger. No lights, sounds or anything. Oh well, might as well take it apart.
At first glance the thing feels permanently sealed and non-repairable. Even the user manual basically states that. They do however mention that you could open it to remove the battery for recycling prior to disposal. You do that by inserting a large flat screwdriver into the slot on the bottom and turning it to unlock. You then press on the metal shaft to get the “guts” out of the case:
Apologizing in advance for a lack of one single picture of the insides, let’s proceed to detail shots
The whole bottom of the assembly was full of crud and water, no wonder things stopped working.
The first step is to de-energize things seeing how much water is around. So I desoldered the two pins of the battery and pulled it out. This brush uses a Lithium Ion Manganese cell from Sony. On removal, the voltage on it was rather low- 2.5V or so, but I decided to let it rest to see if it recovers from whatever short is saw. I briefly looked around for replacement cells, and did find a vendor on ebay selling them, but was not able to confirm the chemistry of his cells. Lithium Ion comes in many flavors, and this particular application warrants the safer Manganese ones.
After scrubbing everything as much as possible with alcohol and letting it dry for a few days, I reconnected the battery externally, added current and voltage monitoring and tried to put it on a charger:
Battery voltage started slowly climbing up and the current looked reasonable- 32mA is about right for a 800-ish mAh cell that charges in 24 hours. At least this indicates the battery is somewhat behaving and the charging coil is still alive. Looking at the coil pins directly, I see 2V P-P square wave at about 80 KHz. So far so good.
After letting it charge overnight, I disconnected the ammeter, and tried running the brush. It happily started. So it appeared most things were functional and we just needed to see why charge light did not come on while on the charging base.
Poking around CR7 (a dual LED) with a scope I could see the top left pin seeing a square pulsetrain of about the right frequency for blinking. The right top pin had solid 3V. The bottom pins were floating. They seems shorted together on the board and going somewhere on the microcontroller through a trace that used to be there. It appears corrosion completely removed it along with the test point. As a quick test I connected the bottom side of the LEDs to battery + and the blinky light returned. From now it was pretty simple- repair the trace and we are back in business.
The repair consisted of scraping solder mask until I could find some resemblance of a solid trace and then soldering a piece of wire from that to the LEDs.
With functionality restored, I turned my thinking at how to make sure this brush lasts for a while longer. First I cleaned and lubricated with a silicone grease the rubber o-ring on the bottom and the white gasket on the top of the assembly:
I then reinstalled the original battery, made sure thinsg still worked and decided some conformal coating would be good at this stage. Not having any handy, I turned to the next best thing- a can of liquid tape:
I tried to cover every surface, avoiding the LEDs and the button. I then poured some more on the charging coil wires as that’s where most of the moisture ended up. I don’t know if at this stage this will help, or possibly make things worse by locking in whatever corrosion has already started. So if you are following along at home- the usual disclaimer applies. Proceed on your own risk.
After a few days of drying, I’ve assembled the toothbrush by inserting it back into the housing and turning the bottom a bit until the latches clicked. The slot in the bottom cap is in very soft plastic, so some of it got torn up during removal. Oh well.
So far so good- the brush is back up and running again. I don’t know how long it will last until something else gets corroded, but at this stage I took whatever precautions I could to prolong its life. Philips,on the other hand could have spent a very insignificant amount of money and coated the electronics in production, greatly increasing reliability. For something almost guaranteed to get water inside, it’s something that really should have been done. Of course that would mean handles lasting way past the warranty and fewer units sold, so why bother..
Electronically, the design is relatively simple- we have AC waveform coming in through the charge coil, rectified and converted to DC to charge the battery at a low rate and to feed the microcontroller. The Micro (PIC16) runs all the indicators and controls the vibration motor.
If you attempt this on your own, you’ll be working with unprotected Lithium Ion cell, which has enough energy to cause a mayhem. So be warned and proceed on your own risk