Nissan Leaf Modules Li Conversion - Questions

- no cells were empty or near reversing.

Question: Do over-discharged Lithium-whatever cells reverse their polarity? I think not, that is a NiMH phenomenon...

Good question. I do not know if they can actually be reverse charged (i.e., that they will hold a charge with reverse polarity). All I wanted to say is that a reverse polarity will be applied to the discharged cells trying to reverse charge them. And will get damaged by being fully discharged (and probably even more damaged if tried to be reverse charged)...

(+-):(+-):(+-) ---> all cells have some charge left in them, the polarity of all cells is (+-)
(+-):(00):(+-) ---> the left and right cells still have some charge left, but the middle just got discharged empty
(+-):(-+):(+-) ---> now the middle cell is being reverse-charged by the adjoining cells as the pack continues to be used

I did find a reference, for what it's worth, about actually reverse charging Li cells: http://www.candlepowerforums.com/vb/showthread.php?316280-Reversed-polarity-18650-cell

Good work!

Didn't you get BMS connector harness and wires from Hybrid Auto? I did, and plugged these in the plugs on the on-battery orange BMS connectors, and routed the cables to under the seat where I terminated them with 9-pin R/C battery charger connectors. So, I don't have to open the bike to measure the battery voltage. Took some measuring to make sure I got the wires ordered right, but I think it is worth it the peace of mind - I was not too eager to open the bike just to measure the voltages every now and then. I can now either use my multimeter and spend a couple of minutes to measure each of the 36 cells or plug in an R/C voltmeter to see the voltages of each of a group of 8 cells at a time.

Oh, on my cells I did not need to modify any spacers. In fact, there were no spacers between the cells in my build. Do you have spacers in your pack? The cells are compressed against each other in my case, nothing between any two cells. There are spacers in each of the 4 holes for the threaded rods, which maintain the proper spacing between each 2 packs, so that the bus bars line-up perfectly. Compress the battery using the threaded rods just enough to line-up the bus-bars: the bus bars would not be long enough if there were spacers between the 18 packs.

I used a simple jig to keep the BMS wires in a position to order them before I soldered the 9-pin connectors. Just a couple of pieces of plastic trim I had lying around in the garage. And masking tape holding the BMS leads in place and to mark them before I soldered them to the wires on the 9-pin connectors. Make sure they don't touch, or sparks will fly (don't ask how I know :)).
.

The 2 wire harnesses that plug in the on-battery wire harnesses. The wires are spread out and taped to plastic pieces, ready to measure and number before soldering. 1" heatshrink tubbing is already slid over each of the 48 wires (only 36 are connected), ready for after I solder them for good.

Wires all ordered and temporarily connected to the 5 RC 9-pin connectors. With 5 cheap 8-cell RC voltmeters plugged-in simultaneously to measure all 36 cells at the same time. That's when I found out these $2 a-piece voltmeters are junk (or should I say, only accurate to 0.1V or so, not the 0.01 I was hoping). Still, they are consistent. So I can pick the best one and have an "offset map" of how its displayed values relate to the actual values measured, for a quick and accurate voltage check on the go: the v-meter is smaller than a matchbox so it can be stored in the bike somewhere (unconnected most of the time, connect only when needed).

The BMS leads routed to the underseat storage.

One v-meter connected under the seat to check voltage of up to 8 cells at a time (just unplug and plug in the next connector to check the next group of 8 cells and so on).

Got it. Yes, mine came with some metal segments along the edges of the cells too. They do not separate the center of the cells, and actually I don't think they play the role of spacers either. If I recall, the actual spacers are inside the holes for the threaded rods and these metal segments go around these spacers I mention. I was actually thinking of removing these metal pieces as they do not seem to play any role in my build structurally. When you mentioned "spacers" I was thinking about a full size plate separating every two modules. In some builds I've seen people mention they install 2mm plates between the cells. I do not have that and I think you do not either. The metal segments that you and I have do not cover the entire cell - they are just around 2 of the corners, right? And yes, on every other cell they do have threaded studs pointing up for mounting additional rails. I thought using these studs would be a good way to lift out the entire pack at a later point. I actually got the matching rails that are apparently part of the Leaf pack but they are longer than what fits in the VX 1 - can be cut to size and used as a jig to lift the battery up (or to compress it down to the frame for an even more bomb-proof installation).

I don't have a good photo of these, but I think they are visible somewhat here (they do not separate the cells in the middle and the spacers in the holes for the threaded rods are barely visible). Note that I have the 2013 cells, which look a bit different from the earlier versions.
.

.

Yup, reading some of the Leaf cell threads over at endlesssphere is worthwhile. I found some references there for tests and specs of these cells, which make me very comfortable using them in the VX1 - it appears that they can safely handle continuous discharge currents in excess of the maximum instantaneous current the VX1 can generate. At least on paper. My observation is that at under about 40A of current draw there is virtually no voltage drop. So in easy riding they barely break a sweat. And shouldn't be stressed much by bursts of full power application in the VX1, as long as they are not nearly empty.

On a separate topic, I wonder how regen braking affects battery life in very cold temps. Currents can be something like 30A (I've seen those show on my display) and probably more with regen braking at high speeds. 30A regen is 0.5c in our application. What would be a safe charging current at "cold" temps? And what is "cold"? 0C? Anyone know? I could not find relevant info online (though for some Li chemistries they say charging at reduced currents, like 0.2C for instance, may be permissible. So, while I may be fine charging at the wall at near 0C and under 12A, it might not be good to use strong regen on a similarly cold battery...

Anyway, mostly theoretical for me as I don't ride in the cold, but might be of interest to others who do. Hard to find data online, because the Leaf car battery has a heater - ours don't...

Since this thread somehow evolved to have the most info on the Leaf lithium conversion for the NiMh Vectrix, I'll add a couple more things to it, which I did not see covered so far.

Dust-proofing. When removing the cooling fans from under the seat, that creates an opening near the motor controller. I thought it would be a good idea to place a piece of foam to block the airflow and the entry of dirt and dust from near the rear wheel. It is pretty easy to size a piece of foam and stuff it in the plastic cover before reinstalling. No glue or fasteners were needed when I did mine - just wedge a strip of packaging foam I had lying around in the rear of the cover when I had the cover off and it will compress against the frame when mounting the cover.

Winter storage. Per advice I got from Hybridautocenter, the Leaf cells should be stored at no more than 40% charge. From Nissan they come at 3.8V standard. That is just under 137V for an 18 modules pack. The Vectrix is supposed to draw some small amount of current when it is off, so care should be taken to check that the pack voltage remains above some minimum voltage level, similar to how it is with the NiMh original pack. The difference is that, the NiMh pack does not get damaged (or gets less damaged) when over discharged, where I understand the Li Leaf cells will get damaged if over discharged. I'd estimate that for a 18 cell pack one should not allow it to go below 110V in storage to avoid trouble, though I've read some info that even down to 90V (2.5V per half module) won't damage it permanently. My pack seems to maintain voltage at rest very well. It has been sitting unused due to winter weather for over a month now and starting from around 137V when I last rode it, I barely see any decrease in the pack voltage, probably 1V if that (I got a revision 4 Runke, which is the latest, and a late version of the motor controller board; your mileage may vary, depending on which charger and controller hardware and software versions you got).

Lastly, the cheap R/C style mini 8-cell voltmeters I got draw about 15mA. They draw it from their ground and second positive terminals (i.e., only 2 out of their 9 terminals draw significant current). This means that, if left plugged-in, they will slowly discharge two cells out of the up to 8 they can measure. That will disbalance the cells. So, either put a switch to disconnect the v-meters, or just keep them unplugged when not needed. They will drain a 30Ah cell near empty in a couple of months, if left plugged-in.