Chassis Selection

Given my goals clearly suggest a small multi-purpose vehicle or station wagon,   I researched several chassis options that fit that basic description.   Despite overt similarity there are substantial differences in several key aspects of the chassis that affect performance as an EV (or ICE, for that matter).   Chief among these are curb weight and CdA, or drag coefficient multiplied by frontal surface area.

Curb Weight

Curb weight is defined as the weight of the car, ready to drive, with no passengers or cargo but all fluids and a full tank of gas.   Chassis weight is the biggest factor affecting the efficiency of an EV under around-town driving conditions.   Curb weight directly affects rolling resistance, the major factor in lower (under about 40mph) constant speed efficiency.  Curb weight is also a major factor in determining acceleration and handling capability and hill climbing power requirement.

When considering curb weight for an EV conversion chassis,  you should factor in take-off weight of ICE components.   Based on my experience you can assume about 20% take-off weight for ICE and supporting components. This works out to something like 400-500 pounds of take-off weight for a smaller 4 cylinder car,  600-700 pounds for a V6 car, and 800 pounds or so for a larger V8 powered vehicle.   This is not just the weight of the engine itself,  but also the weight of the full fuel tank, radiator,  fluids, exhaust system, heat shielding, and other peripherial components.

It used to be necessary to think about GVW, or maximum gross vehicle weight when selecting a chassis as lead acid batteries were quite heavy and the chassis would have to withstand that extra weight.   With a modern LiFePO4 battery an EV conversion will often end up within a few hundred pounds of its original curb weight and almost certainly within the original GVW.   It may still be necessary to do Suspension Work if the battery is a large one (like mine) or it is located right over an axle (like mine), but with more common LiFePO4 battery size and if the battery weight is distributed throughout the car as often the case, then it will be likely that no suspension changes are needed at all.

When I built my MR2 conversion I was operating on the premise that CdA was the most important factor affecting the overall efficiency of the EV.   My opinion on this has changed and I now feel that the final curb weight of the converted EV is likely more important,  but this will depend on your driving circumstances and habits.

My completed xB weighs about 3040lbs,  which is nearly 400 pounds less than the completed curb weight of my MR2 conversion, despite having about three times the usable battery capacity.

Aerodynamics

Aerodynamic efficiency is the major factor in an otherwise well designed EV at higher speeds (above about 40mph).   For a given chassis, it is defined by a term call CdA, or Cd (drag coefficient) multiplied by A (frontal surface area, in square meters).   This term can be plugged into a mathematical formula to calculate the force of aerodynamic drag at a given speed.  Refer to Performance Analysis for more detail on that.   The CdA term is a multiplier on the formula so you can make relative comparisons between vehicles by just using this number.  A car with a CdA of 2 will have twice the aerodynamic drag at any speed of a car with CdA of 1 under the same conditions.

My MR2 had a CdA of 0.54m^2 which is about the best you can get for a commonly found production automobile.  About the only car that is relatively common, modern, and better is the first generation Honda insight and the 2nd generation Honda CRX.   Both of these in 2006 were too expensive to consider which is why I ended up using the MR2.    The xB I decided on has a CdA of 0.8 meaning it is about 50% worse in terms of aerodynamic drag versus the MR2.   This is not surprising considering the relative shape of the two cars, and as such the xB does need noticeably more power to maintain freeway speeds.

Researched Models

The  following table summarized my research results for vehicles I was interested in.  I including source data where I wrote down where the numbers came from.

I figured 20% weight loss from engine size to get to glider weight.

I did not consider other, similar models such as the Kia Soul, Nissan Cube, and 2nd generation scion xB because they were either too new (and thus expensive) or I simply do not like them.   While I really like certain older station wagon models like 70’s datsun wagons, they are rare and would likely require extensive restoration.

Years Model Drive Train Curb Weight (lbs) Glider Weight Est. Engine Mileage Cd A(m^2) CdA Cd Source
1985 Toyota MR2 RWD 2300 1800 4cyl ?/30 0.32 1.7 0.54 ecomodder.com
2001-2007 Chrysler PT Cruiser FWD 3100 2520 4cyl 19/26 0.39 2.43 0.94 cars.com, edmunds.com
2008-2010 Chevrolet HHR FWD 3200 2540 4cyl 22/32 0.354 2.32 0.81 carspector.com, ecomodder.com
2002-2006 Pontiac Aztek FWD/AWD 3800 3040 V6 17/24 0.5 2.72 1.36 carspector.com
2005-2010 Mazda 5 FWD 3400 2580 4cyl 22/28 0.29 2.64 0.76 automedia.com, edmunds.com
2003-2010 Honda Element FWD/AWD 3400 2580 4cyl 19/24 bad 3.08 1.2? ?
2004-2007 Scion xb FWD 2400 1920 4cyl 27/32 0.32 2.48 0.8 wikipedia.org, edmunds.com

Observations

  • MR2 included for relative comparison.
  • xB is far and away the lightest.
  • xB close to the best on CdA.
  • Element and Aztek are porky and relatively speaking terrible CdA.
  • Only Advantage to Element and Aztek is possibility of utilizing an AWD chassis to do a rear wheel drive conversion.

I pretty quickly narrowed my choices to either the xB, which I also have always personally liked, and the Aztek which I also personally like, largely because most other people hate it.

The Aztek is rather counter intuitive just looking at the numbers, but the rear wheel drive conversion possibility was a plus, as was a large amount of interior space, and while not popular nationally they are pretty common in the pacific northwest and non-running ones are cheap meaning I could compensate for the efficiency issue with a bigger battery at the same overall cost.    it was also likely I could remove a lot more weight from the Aztek relative to a smaller 4 cylinder car.

It is interesting to see that despite being much boxier and having much more interior space, the stock xB weighs virtually the same as the MR2 did, and despite being much boxier the xB still has a very respectable Cd in comparison with the MR2, especially compared with some of the other similar vehicles to the xB that are much worse.

Conclusion

After being heckled mercilessly by my friends for even considering an Aztek sanity reigned and I decided to go ahead and use a first generation Scion xB.   At this point I do not regret the decision.