Motor Trend's Update on Volt Development

energy balance finally makes sense

For the longest I didn't understand how the ICE could provide all the power the electric motor needs since it is much less than 160 hp. Now I understand...

From the chart, what happens is not that the batteries do all the work until the batteries reach "customer empty" at 30% SOC then the ICE does all the work until you get home with the batteries more or less off line.

The problem with doing it that way is that the ICE WOULD have to be as powerful as the 160 hp electric motor.

Instead, the batteries and the ICE seem to work together in a narrow band, like maybe 30% to 40% where during peak load, (acceleration, hills, etc) the batteries help out, and then during the lower loads, the batteries recharge up to a certain point.

I think that it's possible that if you were to try and accelerate up an extremely long hill, carrying 4 passengers into a headwind, it's possible you might see a decrease in performance as the batteries approach 30%, ( or would the system sense the extreme load and authorize a lower SOC?)

Either way, the battery management strategy does allow them to use an ICE that is sized for the average rather than the peak loads. Very clever.

Perhaps people should just try to imagine a big barrel of water. there is a spicket at the bottom that feeds the motor. At the top are a few smaller hoses filling up the barrel. One is from the garage plug, one from the brake regeneration, one is from the ICE generator and one is from the solar panels on the roof (sorry, I just had to).

If you conceptualize this simple model almost all answers can be visualized very easily. For example. How fast can the car go on a flat road for an extended amount of time if the governors were disabled (yeah, some hacker will figure this out). Answer? Just think of the barrel. The bottom spicket is open up full and it’s a BIG spicket. What is feeding the barrel? Garage plug? Nope. Regen? Nope. Solar? Of course because it’s a beautiful day in Texas - for this example however assume it’s nighttime. ICE generator? Yep! It’s the only hose filling the barrel. Unfortunately, it’s only 55 kW - much smaller than the size of the spicket feeding the motors. Thus, when you have the pedal to the medal the motors will attempt to suck as much from the bottom spicket as possible. Thus the barrel will start to empty, even when the ICE is running flat out. When the barrel empties to the place they stick the spicket in (say 30% from the bottom) no matter how hard you press the pedal it will NEVER be able to produce more than what is being dumped in - 55 kW. Thus, you can calculate the long term running speed with little difficulty. Easy huh?

Let's take another one: I want to be able pass cars when I want and have a much more lively feel than the tiny ICE alone can give me. How should I set the ICE start and stop points to achieve this? Simple! First you ask how much power do you require and for how long. Next you have to ask at what interval this maneuver will be needed. For example. To pass a truck I may need 70 kW for 1 minute. I figure I will not have to do that again for another 5 minutes. Ok, now that you have your barrel model you can easily calculate the set points and ICE generator loads. You will only have the gen-set pouring into the barrel in this example (you are on the highway again and not using your brakes) so it is easy to calculate. You will need an additional 15 kW for that one minute so all you have to do is determine what point above the spicket the water level needs to be. Maybe that comes out to 40%. Done. Thus, you better make sure the barrel is at 40% before the truck passing maneuver is attempted or you will drop down to 55 kW too soon. Since you have 5 minutes in-between maneuvers and normal highway driving only requires a small fraction of the ICE's generator capacity you can actually determine the minimum amount of power the generator needs to produce for that 5 minutes. Cool huh?

From these simple examples you can begin to see how difficult it will be to determine these settings for different styles of driving and for different drivers and loads (like the poster above mentioned - 4 people driving up a hill - with that example you can easily calculate the maximum speed the Volt will be able to climb that hill - it will soon only be able to give you 55 kW after the spicket level is reached).

I'm thinking that version 1.0 software will have different set points and gen-set loading for different conditions (highway, city, hilly terrain, heavy loads, light loads, etc.) Next generation software will have AI that can learn your driving style and run the gen-set accordingly. For me the gen-set could be run mildly. For a high school hell raiser you might have to keep the ICE running flat out at all times until 90% charge is reached and then start up again when the charge drops to 80%. That way the barrel will be able to supply maximum power to the motors as frequently as possible. I hope they put parental controls on this thing!

This is why I never understood why people kept arguing that the ICE will directly run the motors and follow the load of the pedal exactly. You can see from the barrel model that by doing that you will not be able to re-fill the barrel in time for the next extreme maneuver. While the ICE engine may be made to seemingly follow the pedal movement so the driver is more comfortable in reality it would be more efficient to get the barrel to the proper level within the predetermined time frame. If that time frame allows varying the ICE with the pedal movement for a more natural sound then fine, whatever the customer wants. However, I don’t care about that. I want the ICE to run as efficiently as possible to get the barrel to the level the software has determined is best for my driving style and situation. Of course, I would like the ability to set it to hell-raiser once in a while.

Long Cruise Mode

For long distances, it might be advantageous to let the ICE charge the batteries back to 80%, and then shut down the ICE for 40 miles. With day-to-day driving, and going 50 miles instead of 40, it makes sense to come home with your battery at 30% SOC with minimal ICE operation.

However, for long trips of several hundred miles or more, especially where you may encounter long hills where battery boost would be needed, it might make sense to let the ICE run at its lowest BSFC (highest efficiency) until the battery pack is back to 80%. Then the ICE can shutdown for the next 40 miles.

This could eliminate cycling the ICE on and off, and keep it at its best operating point, versus having it go to max power for passing, climbing hills, etc.

I would think that a control override could be provided so that the driver could switch to this "long trip" mode. Once within 50 miles of your final destination, the "long trip" mode could be disabled, and you would arrive at you destination at 30% SOC so that the battery pack could get the maximum benefit from an overnight charge.

I like Texas's bucket example but I think it needs a slight modification to accurate reflect what the graph is showing as GM's current battery management plan. With high loads when the battery's SOC is at or close to the generator starting point (currently proposed at 30%), the SOC can dip below this starting point. This will happen if the total vehicle electric load exceeds the generator's max power. So, the out spicket can theoretically be lowered below the generator starting point if necessary. This should only happen for relatively short duration periods so the SOC should not drop much below the starting point.

The graph also does show that the battery will charge from the generator when appropriate and most likely to take advantage of set optimized output levels for the generator. The system will operate in this way until the battery is charged to a set generator turn off point that appears to be about 10% above the starting point. I originaly interpreted the graph to be showing that the generator will stay off until a secondary starting point is reached. I think this may be a mistake. I don't think it makes sense to lower the subsequent starting point, but I believe the graph is showing that the vehicle load could be such that the battery can continue to discharge beyond the generator starting point so that the Volt does not become performance limited.

This leads me to believe the genset will be optimized around a particular KW, in the 15-30KW range. I hope GM is also testing some user selected modes that have been mentioned several times over the course of this blog. One being high performance mode if the motor is capable of higher power than the battery can efficiently supply (probably not for me but may be a strong selling point for others). This would allow the generator to supplement the battery for maximum acceleration even before charge sustaining mode is reached. More importantly, I hope they investigate a max range mode. If the driver is going to be driving a long distance 250+ miles, it may be more efficient to start the generator earlier in the discharge cycle and run the generator at a minimal battery depletion level (8-10 KW) until the normal generator starting point (30%) is reached. Obviously this will only make sense if the generator efficiency (KWh/gallon) is higher at this load.