Saturday, September 1, 2012

Heat: Not So Good for Chargers Either

A couple weeks ago, on a Monday morning, we were in off peak hours and I figured it would be a good time to top off the batteries before I ran the day's errands.  You may remember that I recently destroyed my eXpert Pro meter by re-applying the pack voltage to it and not removing 12V power first.  By the way, DON'T DO THAT!  Any way, I'd received a new one, but I hadn't yet installed it.  Without it, I have no idea how much current I'm pushing to the batteries, so I need to keep an eye on them to be sure I'm not over charging them.  The fact is you learn things when you watch what the batteries during charging.  Over time, I've learned quite a bit, and by watching them closely I can pretty much tell what's going on and predict when the charger will terminate.

At any rate, I know which cell will start to rise above 3.45V first, so I put a meter on it and started the charger.  For me, 3.45V is the top of the charge.  There simply isn't enough energy put into the cells above that point to justify the potential risk and damage to the cells by getting them to the manufacturer's 3.6V.  Ten minutes later, I walked out to see the voltage on the cell had climbed from it's initial reading of 3.26V  up to 3.380V.  I know by experience that means I'm about 35 minutes away from 3.45V when the charger should terminate it's charging sequence.  So you can imagine my surprise when I walked out 25 minutes later and found the battery at 3.364V, and the charger still running.  The voltage had dropped, yet the charger was still running.  I stood there for a moment, completely confused, it just didn't compute in my head.  Kind of like walking into your favorite BBQ place and seeing that it's full of vegans.  Not that there's anything wrong with it, but it just doesn't make sense.

Figuring that the charger must have just gotten confused (after all it does have a simple logic board in it), I figured I'd just reboot it.  I turned it off, waited 4 or 5 seconds and turned it back on.  The moment I turned it on, there was a succession of 3 pops, with the last one being quite loud and producing a bright orange flash and rush of hot gas shooting out the charger's vent port.  This was followed by me yelling "Oh shoot!", or something similar but perhaps more colorful.  I turned the charger back on... nothing.  Wasn't too surprised by that.  I checked the breaker on the house and had not tripped, so whatever failed in the charger failed in such a way that it simply consumed all the power coming into it and blew up, rather than shorting.  Or at least that's my take on it.

In either case, there was no doubt that I was going to need to remove the charger and send it back to Manzanita Micro for repair.  This was the second time I've had to do this.  The first was entirely self inflicted when I mistakenly shorted one of the batteries to the chassis with my multi-meter probe, while the car was charging.  That blew up the AC rectifier and melted the end of my probe.  I had no idea what happened here, but I boxed it up, got an RMA number and away it went to Washington.

It arrived in their shop on Friday and they sent me a note Monday afternoon that it was repaired, tested and ready for shipment.  Remarkably fast turn-around.  I called and spoke with Clarice, their office manager, to ask if they knew what caused the failure.  Clarice has seen enough repairs that she can recognize the likely cause of the failure just by looking at the parts that were replaced.  In this circumstance, both IGBTs failed as well as a capacitor.  She said she'd honestly never seen anything like it, but she'd ask Rich when he got it.  Rich thinks that one of the IGBTs failed and when it did, it took the other with it.  He thinks heat may have been a contributing factor.  This is where things get interesting and where there may be a lesson that all of you can learn at my expense.

As regular readers of this blog are aware, I live in Phoenix Arizona, which is slightly north of, and roughly rock throwing distance from Hell.  It's not uncommon for the inside of my garage to be 105°F.  If my wife pulls her car in after getting home from work, as it ticks itself cool, all the heat that was happily stored in all of that steel, gradually works it's way into the garage raising the temperature well over 110°F.  It's lovely.

Here's the thing.  Two years ago the car lived through it's first summer as an EV.  A significant portion of that summer saw the car dis-assembled because I had sent the motor back to Netgain because the balancing putty had fallen off.  Plus I took the opportunity to redo the battery layout.  So when it was back on the road in late July and I was charging it on a regular basis, I would come out to find the charger's yellow light flashing at me.  I went back to read the manual to find out what that meant, and I couldn't find a reference to what that was all about anywhere.  I figured it was just an oddity with this charger.  What it really was, was the charger warning me it was over heating.  I blissfully ignored this warning for the remaining portion of the summer.

As it happens, the car was taken apart for a good portion of the following summer to again fix a balancing putty issue on the motor, and to add air conditioning to the car.  In fact, I didn't get it back on the road until the first week of September, just in time to tow it out to EVCCON 2011.  But even then I would see the flashing yellow when charging, but by then I'd figured out what it was so I would dial the current back until the light stopped flashing.

That's been my modus operandi this summer.  I'd turn the charger on and dial in between 21 and 22 amps.  I'd poke my head back into the garage in 20 minutes or so and if the charger was over heating, I'd turn it down.  About 90% of the time, it was over heating.  I'd turn it down to 16 amps or so.  In retrospect, I think this behavior, and the initial instances of ignoring the warning, damaged the charger over time.  Heat is the enemy of all circuitry and when the charger was overheating, it felt quite literally like a blow drier firing out of that little vent port.  So my advice to you is don't to this.  I realize that most of you don't live in such hot climates, but for the few that do, pay attention to your charger.  Heat kills.

At this point, I intend to charge at 16 amps or so during the hotter months to protect the charger.  It's remarkable how much cooler the unit runs at that current level.  The problem is the charger was configured at a higher current level.  That means if I charge at a lower current level, there's a risk that the charger will overshoot the voltage I've set and consequently over charge the batteries.  Re-tuning the charger for a lower level is possible, but such a huge pain in the butt that I don't want to do it for the 1 month it's necessary before cooler weather arrives.  But I would like to be able to charge at any level I like and not risk over charging the cells.  I do have a solution in the works which I will be implementing and writing about soon, so stay tuned for that.  In the mean time, I'm going to have to watch the end of the charging curve very carefully to protect the batteries.