Batteries


The nicads have been leaving a coating of KOH all over everything and really needed better ventilation. The holes for the fans were already there, but the vents were very small for this type of flooded battery pack. So I clamped the lid to the mill and CNC’d nice round 2 inch holes for the exit vents.

Here is the exit vent for the nicad pack. It uses a 3 inch plastic cap anchored with 4 screws and nyloc nuts to 2 inch plastic pipe adapters, sealed to the battery box lid with silicone.

Here is the exit vent for the nicad pack. It uses a 3 inch plastic cap anchored with 4 screws and nyloc nuts to 2 inch plastic pipe adapters, sealed to the battery box lid with silicone

This is the cover for the 80mm fans and the inlets on the Nicad pack. These covers are just simple sandwich boxes from Target. The fans were also sealed to the battery box lid with silicone.

This is the cover for the 80mm fans and the inlets on the Nicad pack. These covers are just simple sandwich boxes from Target. The fans were also sealed to the battery box lid with silicone.

The video chips have arrived. The LCD is handling the info just fine.

So far the Lithium and the Nicads are staying within 2 degrees C of
each other under driving conditions. The lithium might be a couple of
degrees higher during charging.

The bench test with only one cells shows everything is ok. I need to
connect up two more cells on the bench to see that it works as if it
were in the truck. Once bench proven, I’ll install everything into the
truck.

The only part I’m waiting for is the video output chip. It should be
here in the next week. Then it will be time to bench test everything,
then install it in the truck.

So far the Thundersky’s are still doing fine. I’ll know more when I
get the BMS installed and can monitor the voltage of the cells under load.

There are now two temp sensors in the pack. One on the center
Thundersky cell and one on the nicads. This will allow for a thermal
comparison while under load.

The BMS boards for these Thundersky cells are now built and
programmed. I found a software bug that needs some more exploration.
The software is easy to tweak so it can be dialed in to any
differences between brands of Lithium cells and/or drivetrain/charger
setups. The only hold up is that some parts for the master board have
not arrived nor has the LCD. Testing as a complete system will
hopefully happen by the end of the week.

This system seems like it could be made to work with lead acid. The
3.5″ LCD gives a lot of info.

The UK boards arrived today. The parts will get ordered soon. The 3
Thundersky cells are installed and running in the pack.

So about 11 nicads were pulled and the 3 Thundersky 40ah cells were
installed with a temp monitor on the middle cell. So far it’s running
cool on neighborhood and highway drives.

The plastic panels are machined. Now it's time to see how few nicads can be removed to fit the 3 Thundersky cells in.

The plastic panels to prevent the end screws from damaging the nicads are machined. Now it's time to see how few nicads can be removed to fit the 3 Thundersky cells in.

The last of the bench testing of the 3 Thundersky cells took place
today. At a 40 amp and 55 amp load they are close to 100% of their
rating. This also implies a low Peukert factor.

The cells are grouped together from Thundersky with extrusions at each
end. The screw heads stick out enough that it would surely puncture a
nicad. So tonight I drew up a 1/4″ thick plastic panel that will fit
over the cells ends and cover the screw heads. Tomorrow I hope to
machine a couple of them. Then about 12 nicads have to come out of the
pack to make way for the group of 3 Thundersky cells. Then I have to
make up some thick spacers to make sure the Thundersky’s and the
nicads are packed in tight.

I plan to install the Thundersky’s discharged about 10% so it will be
hard to overcharge them for the first drive/charge test. Drives will
be with temperature monitoring while driving aornd the block, then
around the neighbrhood, then on the freeway.

The BMS boards are only bypassing .50 amps. I’m going to modify them
to bypass something like 1-2 amps for a margin of safety. I sent the
designer a note asking for a part number of the transistor. I’d prefer
a fet. But the trigger voltage available is only around 4.0 volts.
Maybe a logic level fet would work ok.

These cells will be a good test of the various BMS systems that are
here already and some others that are on their way.

Peter in the UK has a running Lithium BMS. It uses a slave board on
each cell and a master board. He uses an LCD to communicate with them.
Peter has been doing this awhile.

I’ve ordered slave and master boards that I will populate with parts
and get the software loaded into each board.

This looks very promising. I suspect it could also be used with lead
acid with just code changes.

It’s simple to install.

The first round of baseline testing of the cells I received from a
scooter vendor for testing, show that they can achieve their nameplate
rating of 40ah at a 1C discharge rate, down to 2.5v. They typically
put out 2-5% more than that. All testing was done with a constant
current load of 40 amps. Recharging was constant current at 10 amps to
4.25v max.

The temperature of the exterior of the cells in my garage were 103F.
Ambient estimate was about 75-80F. I’ll track the ambient more closely
next round. More testing will make clear if temperature produced is
the limiting factor in using these cells in an EV.

During discharge I noticed that the voltage fell much faster when it
hit 2.8v. So I suspect this will become more important as the loads go
up in future testing.

Next round of baseline testing will be discharging at 60 amps constant
current down to 2.5v. After that it will be time to put them in the truck.

« Previous PageNext Page »