One down four two go… I installed the the BMS in the trunk configured to monitor 12 cells. It seems to be working. I connected my laptop (with a USB to CANBus adapter) up to it and was able to read each cell voltage. I still need to mount the thermistors in the battery box and run the CANBus line throughout the vehicle.
I finished populating one BMS PCB so it can handle 12 cells. The BMS is designed so that I may use it on battery boxes with as few as four cells and as high as 20 (the front battery box has 20 cells).
I have a battery test stand that uses AA sized lifepo cells in regular old AA battery holders. This makes it easy to populate it with as many cells as I need for testing the BMS.
Here it is charging for the first time.
Each time I drive my car, I learn a little more. Here’s a run down of what I’ve learned…
1. The current shunt should be installed on the bottom of your battery pack, between the negative terminal of the pack and the negative side of the motor controller. I originally installed it on the positive battery lead, but realized the voltage reading on my meter wouldn’t work.
2. With the meter, I can read pack voltage, current draw and amp hours. It is interesting to see the effects of shifting and how much hills can suck your pack dry (I guess my car is heavy, I should probably get it weighed).
3. My batteries are not very well balanced. After sitting in storage for 2 years, 3 of the cells are just a bit higher than the rest. Once I get my BMS installed I will do a full balancing on the pack to hopefully correct this. Until then, I don’t feel comfortable doing a full charge (up to 4V per cell) or a full discharge (thinking I go down to 2.8V per cell). I’ve only been driving my car on short trips so far ( less than 20 miles).
4. Cold weather = voltage sag. It hasn’t been much more than 40deg F the last couple weeks. Taking short trips doesn’t give the pack a chance to warm up. With a relatively full pack I can easily get the cells to drop below 2.5v (avg) if a gun the throttle. It bounces back pretty quick after the acceleration. This was why I wasn’t getting the acceleration I expected out of the motor controller until I tweaked the voltage limits.
5. The Mes-Dea vacuum pump is a bit loud. I mounted it on a thick piece of neoprene and wrapped it in foam to quite it down. It kind of ruins that hole “turn the key and hear nothing” thing, but it’s not too bad.
6. Power steering will be nice.
My dad brought his trailer over and we hauled the Boxster down to the DMV. 45 minutes later, I picked up my new plates and drove the car home (Dad following me just in case something went wrong). Fortunately the trip back was uneventful. The car felt great on the road. Acceleration feels similar to the stock vehicle (maybe a little less). I’m going to play with the Zilla settings to try for a little more oomph. Shifts very smoothly even without pressing the clutch (note to future builders, you probably don’t need a clutch).
Over the holiday I’ve been able to put in quite a few days working on the car. Today I finished up the high voltage cabling, wired all the low voltage circuits (+12v, ignition lines, hall pedal…) and spent some time building an interface cable for the Zilla (since I misplaced the original). At the end of the day, I lowered the car off the jack stands and took it for its maiden drive.
Still on the To-Do list:
Fix sensor switch in door latch Re-attach the front bumper Wire up DC-DC to charge the 12v battery
- Mount charger
Vacuum brake booster
- Power steering pump
- Install cabin heater
Covers for the battery boxes Wire up the amp-hour meter
- Build and install the BMS
- Cooling pump for the Zilla
When I purchased my cells, they came with these copper bus-bar-ish connectors. They are bent in such a way that provides a little compliance between connections. This helps lower the stress on the cell’s binding posts.
They are basically four pieces of .02″ copper held together with some heat-shrink. They are sized to join two cells sitting next to each other with their wide sides touching. If the cells are position with their narrow sides touching, these connectors are too short.
(Note to cell manufactures – please locate your binding posts so that we can use the same length connector regardless of the orientation. Thanks)
For my last EV build I had to some slightly longer connectors to handle the narrow end orientation. Fortunately, I can re-use those for this project. However, I also need some special connectors for the front battery box. So, I ordered up a sheet of copper and started cutting…
I made a cardboard template as a guide. I located the holes in the template about 3/16″ further apart then the actual hole spacing to allow me to add the bend. Long ago I had free access to a waterjet cutter so cutting out shapes like this was a breeze. These days I have to pay for waterjet cutting, so it just wasn’t worth the expense for a few connectors. So, tin snips did the trick.
Over the last two weeks, almost every time I wanted to work on the car, it was raining. Since my car is outside, this makes it tough. So, this week I lowered the car back down on its four wheels and moved it into our covered carport. I hope it’s the last time I have to push the car by hand.
After the move, I got back to work. The first task was to mount the two 2/O cable glands in the front trunk. It turns out that removing the gas tank wasn’t such a bad idea after all. It made the task of running cables to the front trunk MUCH easier.
Next I mounted the front battery box and filled it with cells. This would have been much easier if the my cells hadn’t gotten just a little swollen while in storage. It meant that I had to compress groups of cells between two aluminum plates squeezed with several pipe clamps while I applied poly strapping to keep them compressed. Then I could fit them in the box.
I’ll be replacing the poly strapping you see in the picture with strapping that goes all the way around the box. With the idea that it will hold the cell in the box if the car every was to roll over. However, there is so much friction between the cell I would guess that they are not going anywhere.
There are a number of little things I need to do before installing the battery boxes. I am going to be installing a BMS board on each battery box and need a way to wire it to each of the cells. I decided on using simple DB25 connectors. They are inexpensive, but robust. They have good mechanical properties with their mounting screws and have decent current capacity (the contacts are rated to 3 amps). A little work with the Dremel tool and …
I also needed a way to secure the cells from bouncing once they are in the battery box. I decided to use poly strapping through some slots in the box. I will run the strapping all the way around the bottom of the box. I cut slots then ran thin strips of sand paper through them to remove any sharp edges.
The slots are a little hard to see, but they are there.