Redox Batteries

Here's an interesting concept. "Refueling" a battery by circulating the electrolyte out of a depleted battery for off-board recharging, and filling up with a fresh charge. This enables the charging process to take place without the presence of the battery the whole time.

More from The Kneeslider, Wikipedia.

Driverless Neighborhood Vehicles

No one has gotten the styling right but, as I've said before, the concept of driverless cars in certain areas or neighborhoods is bound to happen.

Saturn Door Replacement

Last week I had a mishap involving an electric vehicle and a friend's car, resulting in me punching a hole in the rear door of said car ('93 Saturn sedan). Luckily no one was hurt.

To fix the damage we went to a junkyard and pulled an identical door from a salvage car to replace it. Since early Saturns had mostly plastic body panels (doors, fenders, trunk lid, bumpers etc) bolted to steel frames the whole panel swap should've taken about 20 minutes.

However, once we removed the panel we found the door beams had been bent, and the window could not retract fully. Since the interior color of the salvage door was different and the wiring harness was cut at the junkyard, we had to change the frame and interior panel, while keeping the original wiring.

Because of the steeper learning curve it took us a good 4 or 5 hours (and we weren't in any particular hurry). The design of the door was very interesting and quite clever in its simplicity. This was all before side airbags so it was just a matter of removing the exterior panel, then the interior panel to get to the wiring. Once that was removed five bolts was all it took to remove the door. If it didn't have power windows or locks the entire door change would probably take half the time.

Both the original and replacement doors had no corrosion and all the fasteners worked as expected, even the plastic wiring clips. The door hinges had some play which allowed for adjustment and fitting. We also made two paper shims and bolted them under the bottom hinge to help angle the top edge of the door closer to the car.

The only surprise was that we found the interior switch assemblies were slightly different (the salvage door was from a '92 model). See picture below. "A" fit both doors while "B" did not due to a difference in tab spacing. That was a head-scratcher.

I've been a firm believer in non-metallic body panels for a long time (fiberglass, carbon fiber, various plastics) and this really confirmed the benefits. Simple to replace, impervious to minor door dings and stone chips, easy to match the paint, and easy repair access. Steel doors can only be accessed from inside, not both sides because they're welded assemblies.

Of course there could be a number of disadvantages as well such as cost, crashworthiness and weight. While the panel itself didn't seem heavier than steel, it's unclear how the overall weight of the door assembly compares to similar vehicles from that era. Still, composites are definitely worthy of consideration.

Once the new door was installed, everything worked as expected: Power window, power lock and of course opening/closing/latching.

This repair required nothing more than basic hand tools. If it was just the body panel all that would be needed is a screwdriver with a Torx bit. Total cost was under $140 for the replacement door and the color matched just fine. Youtube was helpful too (see http://www.youtube.com/user/saturncrzy). The last thing is now to do the pinstriping.

Electric Kart Demo

Recently I had a chance to demonstrate my electric kart at a street festival called Clarendon Day. So I called up a few friends who brought an eclectic group of vehicles out for display. We set up a small course using cones and the response from passersby was very positive. Click here for pictures.

EV Battery Selection

Let's select batteries for a hypothetical car. This car is a conversion of your basic commuter. Let's say we want a minimum range of 60 miles for regular driving.

Key battery information needed from manufacturers is voltage (V), maximum amperage (A), amp hours (Ah), weight or mass (kg), and physical dimensions of each cell (l x w x h).

First what is the peak power target? For example, 50 horsepower (hp), 100 hp or 200 hp? Peak power is a lot more than the nominal power needed. A car may be capable of producing 100 hp but it only needs to use say 15 hp to cruise at 60 mph.

However, if it only had 15 hp it would not be able to go up hills at that speed or accelerate very well at all. So we need to know the maximum power that should be available at some point during its operation. Let's go with 60 hp for now. At this
point I'm switching to metric. One kilowatt (kW) equals 1.34 hp. So 60 hp is roughly 45 kW.

One more thing. We're going to go with a 96V system here. Voltage times current is wattage. And of course a kW is 1,000 W. So a 45 kW system at 96V will require about 470 Amps (45,000 W / 96 V).

By the way, peak power determines top speed while torque determines acceleration. That's a topic for another post but suffice it to say that we need enough torque to get to our top speed and enough power to maintain our desired speed.

Then comes how much energy the batteries need to provide, and how much such batteries weigh?. Energy is power times time.

Will the batteries provide enough power for the time (distance) needed? These are often quoted in terms of kilowatt hours (kWh).

The way to determine how much energy is needed is to estimate or model the expected vehicle drive cycle. This is exactly why it's impossible to give an exact figure for energy or fuel consumption unless very specific conditions are provided, why we
are all well aware of the line "your actual mileage may vary". The way to validate this is with actual vehicle testing.

In the above graph, the two hatched areas are roughly equivalent and illustrate why operating at full power reduces running time (and thus distance traveled) - even though speeds are higher because aero drag is much, much greater. How far a given amount of energy will drive a vehicle simply depends on how it's used.

For this example, let's assume the peak power needed is 60 hp, and our average power output is 30 hp (22.5 kW) for each 60 mile trip, which takes 1.5 hours. Thus the nominal amount of energy needed is 22.5 kw x 1.5 = 33.75 kWh.

But just because the batteries may hold an adequate amount of kWh it does not mean that's all that's needed. Primarily this is because:

1. Battery depth of discharge (DoD) should not approach 100% because that will dramatically reduce battery life (number of cycles it can be charged and discharged)

2. It leaves us with zero margin for error.

3. Does not factor in any inefficiencies (or regeneration for that matter) in the system, whether electrical or mechanical.

If for example we only want to discharge up to a maximum of 70% (30% charge remaining) to extend battery life then we need to divide the kWh by the target percent discharge. So we actually need a battery pack with 48 kWh (33.75 kWh divided by 70%).

One lithium ion battery manufacturer makes cells that have a nominal voltage of 3.2V, provides 90 Ah of energy, and weigh 3.2 kg each. So to spec our pack we need to remember that batteries in parallel add current, while batteries in series add voltage.

To reach close to 48 kWh we need 30 cells in series to get to 96V. Then we need 5 in parallel for each of those 30 cells to reach 450 Ah (we won't make the 48 kWh target). That means 3.2 V x 90 Ah x 30 x 5 = 43.2 kWh.

So 150 of these cells at 3.2 kg each means the battery pack, sans wiring and other hardware will weigh 480 kg (1,058 lbs).

Pretty hefty considering less than 3 gallons of gasoline weighing about 23 lbs would power the IC equivalent (20 mpg).

It's not a favorable comparison but we have to start somewhere on this path to vehicle electrification. And there are a lot of other variables to consider which I'll get into later.