Post Go back to editing

LTM8062 Magic smoke

I've been evaluating the LTM8062 Tracking battery charger for some time, and have a good understanding of how it functions. I built a small breakout based on the reference design on page 19 of the datasheet:

Where R1 = 1M and R2 = 380k for a 3-cell application. I've been getting good results, although EMI seems to be a little high. Here is a graph of the battery voltage of each individual 18650 cell over time:

I'm using old cells, so that explains the difference between the cells. Charging starts at ~1.5A and tapers off to about 200mA. After that, the charger modulates between charging and not charging every few minutes. All is well.

It's a shame there's no easy way to set the current, because 2A is a bit much for my application. Oh well.

Because I want to protect the cells, today I added a protection circuit which also provides balancing. One feature is that it disconnects the output when it first powers up. When you connect a charger, it connects the MOSFETs so the pack is connected. Smart.

I've disconnected and reconnected the cells a few times, and used the LTM8062 module to "tell" the protection IC it's time to connect. Perfect. Today, I went trough the same routine and the module blew up. I'm pretty sure there were no anomalies that could have cause e.g. a short. Inspection shows that something inside the module, between the Vina/Vin and GND banks has blown.

I've checked if the protection board pulls a large current with a bench supply, but it doesn't.

I'm guessing the sudden connection of the cells to the charger causes a current rush. The LTM8062 datasheet only speaks of reverse input protection and thermal foldback, not of overcurrent protection. I'd guess the IC inside the module has that anyway, but maybe not.

Freak incident or am I missing something? Any ideas before I go back to TI's bq24133 I was about to design into the product before I came across this module?

Thanks in advance!

Parents Reply Children
No Data