Overview
To date GM has built 80 pre-production IVers and 300 packs. Those vehicles are in testing 24 hours/7 days per week. Production and process validation builds will be made at Hamtramck from March through August 2010. From August 2010 through November 2010 manufacturing validation builds will be made.

The current prototypes have been tested cumulatively over 1/4 million miles. So far they’ve gone through hot weather trials, mountain trials and 65% long distance calibration buy-off rides. In December they will undergo a 300,000 mile three-lifetime simulator test. “We’re very happy where we are right now,” says Farah.

Battery Update
In December 2008, GM built its first battery pack based on their own in-house design and so far over 250 of these 4th and final generation packs have been made.  No further refinements will occur, others at GM are working on lighter next generation packs.

GM will bring it own battery pack assembly plant online in January and it is currently being built-out.   The first pack to be made there will happen in February 2010, and the first saleable packs will be made in August 2010.  The battery lab is working to develop and refine the processes by which mass production of large numbers of packs is flawless.

Over 50,000 cells have been tested on over 300,000 simulated miles in the lab and to date not a singe cell failure has occurred.

Stress Testing
Cells are exposed to crush, penetration, thermal stress and overcharging.  Modules and packs have been exposed to crush, pressure, shirt circuit,corrosion, thermal stress, and seal integrity.  All is well.

Vehicle Stress Testing
Cars have been exposed to the twist ditch, water exposure, potholes, and vibration.  Crash test have shown battery remains safe at 30 MPH head on, as seen in the picture below.

Learnings
Farah admits NVH (noise vibration and harshness) has been “particularly tricky” especially in charge sustaining mode as the car was tweaked to be extremely quiet in electric mode.  Through changes such as increased insulation, specially designed instead of off-the-shelf bushings and new side airflow vents a pleasing experience has been obtained.

Battery chemistry had to be tweaked slightly to achieve longest lifetime, which will be effectively 10 years, though it was noted in temperate climates much longer lifetimes are possible.  Extreme cold and even more so extreme heat degrade the battery life expectancy.

“10 years is the target life, but depending on where you live, you could see significantly more than that,” said Farah. “In more benign conditions — if you do more city driving — and if you are in a more temperate area, the battery would last significantly longer.”

“But even if you live in Phoenix, as long as you charge at night, and you run during the day, your battery will remain happy,” said Wallace.

It has been found that the battery packs are well protected by the car’s structure and cells handle all abuse situations remarkably well.

Though I asked Farah, he still would not release the size of the gas tank claiming we won’t know until just before launch because its still being tweaked.  The gas engine will likely be programmed to circulate once per month even if the driver never uses it.

Farah also wouldn’t disclose the Chevy Volt’s curb weight though states its was “heavier than I would like.”  Efforts are being made to reduce weight of both the car and 400 pound battery pack.

GM has signed off on the final design freeze of the charging equipment, both 120V, 220V and the in-vehicle on-board charger.

Though not confirming dates, Farah implied he is looking forward to beginning work on the higher performance Cadillac Converj, which has unofficially been greenlighted for production.

Below is a video of the Chevy Volt battery pack undergoing serious crush testing:

Volt in Tennessee

Do you feel the charge-sustaining (CS mode) experience is now pretty solid and are you pleased with it?
We are definitely pleased with the level of progression we’re at. We are right on track to where we need to be in terms of the next stages of development we need to do to meet our target. Is it commercial right now, no, but that’s why were taking the time to get it to be commercial. Does it have the capability of being commercial, absolutely. We’ve proven that to ourselves.

We just took our leadership through a test drive in Warren (Michigan) and comments are that they cant tell when the engine is coming on or off. Those are the kind of things we like to hear.

You drove the IVs around the mountains of Tennessee. What about those big hills and the generator. It sounded like it drove very briskly powerfully and effortless is that true?
Yes, absolutely. Some of our control capability to utilize the battery, the engine, and the motors and to be able to optimize for high load and low load situations we’ve been working on developing that stuff for the last year and half an I couldn’t tell you how happy I was when I was in Knoxville because that’s when I had the opportunity to see it all come together. A lot of those bits of software all came together on one trip and it was a joy to drive.

So you took it up some steep hills and mountains?
Absolutely. We were in the Cherokee area taking it up through those hills and a lot of situations following it. It was a very touristy area. Following the speed of traffic, absolutely no problem, and where I had opportunities I certainly like to push the limits of the vehicle, and we did on those mountains, and I could not get it to degrade in performance.

That’s in charge sustaining mode?
Certainly in charge depleting, we have no issues because we have battery power, I’ve got it all at my fingertips. Now in charge sustaining we know that the engine power is slightly less than the peak vehicle power but we have controls ways to manage that and to utilize the engine in conjunction with the battery to get a little bit of extra power when we need it, and replenish it back when we don’t.

I know the battery runs down to roughly the 30% level before for the engine comes on. Is that 30% itself the whole potential buffer band?
We certainly don’t utilize the full 30% but there’s a portion of that that we utilize as a buffer.

You wouldn’t go to zero ever?
No. When the engine cannot meet peak load requirements we’ll suck a little bit out of that buffer and as soon as we have a situation when we can, we’ll put it right back in. All the controls that we work do that to optimize not only the driveability but the efficiency as we’re doing it, NVH (noise vibration harshness) as we’re doing it, the total driving experience as well as to protect the battery from a life experience. This is what allows us to give really good warranty life on our batteries as we’re doing I all in a controlled fashion.

NEW: Discuss this story in the GM-Volt Forum

Lotus EREV Engine

Moving to production brought about many technical and engineering challenges, and few have followed in GM’s footsteps with most companies opting instead to go with simpler pure EVs.

Tony Posawatz is the Volt’s vehilce line engineer who has been involved in the Volt program since day one. Here’s how he answered the following question.

Does GM plan to offer a portfolio of electrical vehicles or just Voltec and its derivatives?

GM’s general position is going forward we will have a broad choice for customers. We see the option of a Voltec system continuing forward. How many alternatives off of that is interesting. Theoretically a pure battery electric vehicle would still be a Voltec. That’s one of the beauties of why we like the option we’ve taken, because it has that level of optionality.

This issue is now, for example, if you start with a pure EV there no place to put a range extender, or if you start with a hybrid that has at its roots an internal combustion engine, you take the engine out and it does nothing.

We will still do hybrid systems, further enhance them, get costs out because there are certain applications they work better in until we get further development on pure electric systems.

Many people ask us why there aren’t others following us in droves in developing EREVs. It’s a very hard configuration to make work. Once an engine is burning it changes the game, and we have non-intentionally thrown some agencies for a loop, like EPA and CARB, because their existing rules don’t apply. A lot of regulatory compliance stuff goes along with it not to mention the pleasability, the noise, the efficiency of the operation and the maintenance of it.

Therein lies the challenges associated with it and why maybe some companies never made the leap, because its hard.

In an absolute technical sense its hard because of its overall complexity, and the balance and interface and integration of all these things together add to the challenge.

It doesn’t mean the Volt can’t be a second or third car, but (if it is) you can never get to the volume we want, the building of interest to drive the cost down to get suppliers engaged and involved, and to get competitive bids for our components. This is the big picture. Everyone is so focused on the price of the car, but remember the price of the first cell phone.

We are in this for the long haul. There was a lot of deep thought in how to play the endgame here. Even there’s still debates such as shouldn’t it have been a Cadillac. Maybe in the near term it would have been great, because maybe we could have lost less money or even made money because we could have charged more.

/Special thanks to our Veterans today.

Chevy Volt Powertrain

The design of the electric motor, is there a separate generator or does the motor itself just turn the other way and act as a generator?

Very interesting question. There are two motors. One is considered the traction motor and the other one is the generator. However, and they are two motors, the traction motor is  higher-powered and designed specifically to meet the traction requirements.  The generator is designed to efficiently couple to the engine to generate what we need and match the efficiency band of the engine as much as possible in all the operating modes. So we look at that motor as coupled with the engine in system and then we also have a traction motor.

Some of the interesting pieces though of this are, for example, in EV operation I have two motors on board and I typically use the traction motor only to drive the vehicle. However, I do have some mechanisms to couple those motors and in some points of operation these two motors can be coupled and have a more efficient state.

Does that produce more power if they’re coupled?

It’s actually not additive for power, it’s actually the way it’s architected, and a lot of this is proprietary so I can’t get into the full architecture, but what it does is optimize the rotating speed and the losses of the motors so in certain states its better to operate both to propel the vehicle and in some states its better to utilize more of the generator and less of the traction motor. In some states its more efficient to use more of the generator and have more of the traction motor actually be a generator. That would be for example in coast down situation often we use our traction motor as a generator on regen.

We do have the ability to utilize both motors in propulsion mode.

The issue is and the direct answer to your question is we do have a primary traction motor and a primary generator motor and they are designed specifically for those levels of operation. However, we have a little bit of flexibility in exactly how we use them.

As the car is technically a mule or prototype, it is not production-ready and has had some issues.  A month or two ago it began popping loudly into neutral whenever the accelerator was floored.  The power electronics control unit was replaced and after that it  almost never happened (it happened one more time). So I’m gentle with the accelerator.

The other day I was driving to work and went over an unexpected construction zone pothole.  The car was jostled and suddenly it went into neutral.  After that it could no longer be put into drive.  Despite turning it on and off and moving the shifter in and out of drive neutral and park several times, that was it, dead.  A tow truck was called and off it went to the dealership for a MINI “flying doctor” to come and repair it. After a few days I found out it was the power electronics control unit again which was again replaced.

This made me immediately realize the importance of extensive testing of new electric cars over rough road conditions, potholes and the like.  With new technology electric cars there may be cables that can simply come out stopping the car dead in its tracks.  Fickle electrical connections andsensitive electronics may be more vulnerable to these effects than mechanical combustion powertrains which have been road-tested for a hundred years.

I reached out to Volt vehicle line director Tony Posawatz to see how carefully GM was evaluating pothole effects on the Volt prototypes.

He responded graciously:

We do more tests to our cars and especially the Chevy VOLT than anyone could imagine including some pretty severe potholes on our Milford Provings Grounds and other very difficult road surfaces. As you know, the car quite easily navigated up and down Pikes Peak, through the hills of West Virginia as well as Death Valley during the hottest part of the summer (it was 118 degrees when I called once to check on the team).

So although there may be a lot of new electric car startups on the horizon, the Volt may well have as another advantage GM’s long heritage of specialized quality control testing facilities and expertise.  As has been said before, they must get this one perfect.

It is this range-extender that make the car so unique. Although the Volt always operates as an EV, once the battery’s state of charge reaches roughly 30% this generator powers on, providing electricity to the motor. Energy is also obtained from regeneration and the battery buffer when power demands are high.

GM has yet to publicly demonstrate the car operating in this charge-sustaining mode, but most accounts indicate it is unnoticeable.

In spite of many months of rampant speculation here, GM’s lead Volt engineer Andrew Farah has finally disclosed some details about the engine’s operation. He said the generator would “operate from 1200 to 4000 RPMs and from a 30% to 100% load.”

Further explanation comes from Volt Powertrain Engineer Alex Cattelan:

“We don’t keep it at a fixed RPM, we have a window of operation that is optimized. We have been able to optimize the engine for a window of efficiency but it is still best to change your power and torque levels within that window as the customer torque request varies.”

“We don’t want to always be operating at one state because really you may be putting too much energy into the battery or drawing too much energy out of the battery. It is still good to vary that engine power and torque. Not to follow exactly what the accelerator pedal does, but to optimize efficiency.”

Finally Volt vehicle line director Tony Posawatz explains it this way:

“In charge sustaining or range extender mode, the Volt will not follow the throttle position. The challenge is to select the right operating points (RPMs) that are 1.) efficient, 2.) pleasing to the driver, and 3.) meet regulatory requirements.”

And he teases us, “we are about ready to expose people to this experience.”

Now that these cars are finished what happens next?
Interestingly enough this is probably where the intense work begins. From a build perspective, we won’t build again until early next year in Hamtramck. What we do now is we take the production intent designs that were built up in the integration vehicles and we test the living daylights out of them and continue to do iterations and iterations. A lot of it is software. A lot of it is taking some of the crudeness in fit in the body fits and tighten them up. Although we do it on math, variations occur and flushness and fit come out a little differently then they do on paper.

So we’re now tuning it. And between that time from October of this year and March of next year, we test the vehicle to confirm the production design works, tweak and fix things, and as we tweak and fix things those changes have to be incorporated in the production tooling and those things typically have a longer lead time.

Then the production tooling is in place. Those are the big heavy expensive dyes to stamp the sheet metal. Those are more significant molds and cavities and processes to manufacture plastic parts in high volume. That’s kind of what happens right now. Right now is the testing, problem-solving, refining and putting all that learning into the production tools. Come the March timeframe and we’re running Volts with production tools. And that is still an iteration process or learning process to refine it for what’s going to happen later in the year.

Are you beginning to put hardware into the Hamtramck plant?
Oh yeah, the Hamtramck plant and the battery plant are on a project plan to upgrade themselves to be ready to build in the early 2010 timeframe. So right now there are tools in toolshops being produced and as we find out, we need this or that, we’ll change it on a production tool. That’s the process we’re in.

When you first start building in March what are those vehicles called?
We call them PPVs, pre-production vehicles or validations. We have a couple more “flavors” but those are all flavors of production vehicle that we work through. This is fairly state of the industry, our terminology is a little different, but every manufacturer does these iterations or flavors of vehicles that have the next phase of software.

At some point in time for example the interior parts have a certain graining to them that you do as one of the last things. You put in a little grain into the tool, but once the tool is grained and they have this nice little texture to the plastic parts it’s very hard to change the tool if you have any functional dimensional changes. Later in the Spring we’ll actually grain the parts so that they really look like production and piece the cars together.

Right now things are actually going pretty well.   We have certainly a lot of issues but that’s standard for this type of project and where we are at in the program. Issues are actually a good thing. We know no one is that good to get it spot-on or if the issues don’t properly surface they somehow surface later. If you have a lot of variation in some of the parts, and you don’t see it early enough to try to control it, to get thefir finsihs and execution elements perfect.

PPVs are not saleable?
No, they’re not saleable yet.

So you will build a few hundred of those?
No we build in the two to three thousand range. We’re finalizing that number right now. A lot of that is practicing the processing of that. Ultimately their will be a line rate of many jobs per hour when we do this so you want to get the guys actually practicing in the production environment.

What is your role in the Volt program?
I head up the team that does all of the propulsion systems for the Volt. That includes motors, power electronics, we don’t do the battery but we work extensively with the battery team to do the integration of the battery, and the EV propulsion system

Are you only working on the Volt?
I am dedicated to the Volt and I heave a team of people that are dedicated to the Volt and I’m extremely enthusiastic about the Volt. We have segregated a team of people that are working on the Volt and the Volt only.

What are you doing now?
The specific stage of the program that we are in, for powertrain, is building on our third level of hardware which is integrated in the the vehicles. We’re doing all of our validation testing on that level of hardware; we’ve got past development where the architecture needs to be, through two generations of hardware we’ve confirmed all of our performance requirements with that architecture.

There are three major areas I’m focused on right now. Making sure that all of our hardware that has been built to date, that third level of hardware is in testing. We are validating all of our parameters for it; durability, that’s the hugest piece, the reliability, making minor fixed to areas that we’d like to improve.

The other major area that we’re focused on is the calibration of the system which is huge, because it’s a very complex system from the perspective of the torque generators we have, the engine, the motors, the power electronics. So we’re tuning all of those systems to make the vehicle fun to drive, to meet all of our efficiency requirements, our drivability requirements, noise vibration and harshness requirements, and we are now entering the phase of program where we will do all of the development for certification. So that’s fuel economy label certification, and that will continue for some months now. We’re going to test per the procedures that are being developed for this program and verify that all of our calibration is tuned in for optimization of the driving experience along with the efficiency of the vehicles. And that’s really the stage that were tuning into now.

In addition the third element that we’re working on is tuning in the manufacturing systems. So we’ve been building our production in our manufacturing plant and we’re right now preparing for what we call pre-production level hardware. That before we build saleable parts we’re going to build pre-production parts in the assembly plant so that its ready for volume. Make sure that our suppliers are prepared for that, make sure that our supplier’s tools are prepared for that and tweaked for quality, our manufacturing plant is tweaked for quality. I was just there last week and walked through all of their systems and systems development to make sure there ready for the production phase.

So those are kind of my three major areas of focus right now.

When you mention the production plant are you talking about Detroit-Hamtramck (DHAM)?
No that’s vehicle assembly plant but ahead of that we’re going to our suppliers and their manufacturing plants for individual parts and systems. And in addition, our drive unit and engine have their own manufacturing facilities that we are walking through and making sure they are ready.

Is GM building its own electric motor for the Volt?
The motor is actually supplied to us but we will be integrating that motor in our drive unit so its encased in our drive unit, we provide tooling, rotating components and all of that which is built in a manufacturing plant owned by GM. We are getting ready to do all of that manufacturing , so the housings, for example we take the castings from a supplier we do all the machining of the housings, the bearing the gearing, all of that kind of stuff.

So you get some of the parts from outside but you put them all together?
Exactly, so it’s a manufacturing step that happens and we send that drive unit and that engine to the assembly plant in DHAM for installation into the vehicle.

Have they done anything at DHAM yet to get ready?
There’s a ton of work going on at DHAM to get ready, and we’ve built our what we call our integration vehicle, and we’re towards the end of that phase. That is being built in pre production operations, however the DHAM team has been very integral in those builds. They have been overseeing the builds they’ve been doing slow builds, they’ve been identifying all issues for their production processes. They’re currently working on tooling for the body. There’s a number of things that DHAM’s doing to get engaged and ready. Because the next phase of vehicle, which is not a saleable vehicle yet, but its our next phase of product, will be built in DHAM so they are getting all of their systems tuned, the personnel trained and ready to go.

When will the first vehicle be built at DHAM?
We’re tuned towards first quarter of next year for that date.

I had a chance to discuss this extremely important and unique mode of Volt operation with Volt executive Tony Posawatz who explains what functions GM still has to work on.

When the generator goes on will it come on gradually?
We’re still playing with it. The initial transition we like very much, it is almost imperceptible. We want to tune it and exercise it more for production readiness, but we have an algorithm of software that feathers it in so that its almost imperceptible.

To many people it is imperceptible, to those that are a little bit more tuned in…Frank (Weber) says its not quite perfect, he wants it so that no one can hear it.

That transition isn’t handled in most hybrids that well, particularly when you stop at a light an the engine shuts off and you get that shudder. We have this benchmark we set, that we’re going to blow them away.

By the way, our motor is so much bigger than theirs, the power output of the electric motor, that effectively is the starter. We’re in the point of tuning it and it’s a very very complicated tuning exercise if you want to make it perfect.

It’s a balance equation. Here lies the challenge. You will get some feedback from the car about its environment, or the road load. You’re demanding so much power from the car, and you will get some feedback. That’s one input that you have to be able to respond to to deliver the right amount of power. The other aspect that we have to play with is this area of OK when I respond to it, the engine generator will run in different modes or RPMs. How do I transition from mode to mode, a timing issue, do I smooth it, do I try to react immediately to that response? If I act immediately to that response how will customers receive that? such as a wild of swing of RPMs?

There’s the responding to the car’s demands based on the loads number one, then there’s the affect on the customer, we want that to be really pleasant. The transition to charge sustaining, we’ve got that down, it’s the operating in the very different modes we’ve got to figure out. The next leg is making sure the engine is running the most efficiently that it can.

Generators are typically run at a sweet spot that’s very efficient. We could do that. The question is, does the amount of output of power that results from that manage or balance with what the car needs?…hmm how do you do that? That’s another reason we have this battery buffer reserve.

How often do you turn the engine off and on is another question. These are the factors that have to be balanced.

People ask why can’t you show it to me? Because we want it just right.

This requires a lot software. The other interesting challenge is the regulatory issues. Theoretically you could run the engine enough to build a surplus of energy. The regulatory guys are not going to let us do that, it’s a five legged stool.

Is the EPA sitting on a committee while you engineer the car?
This is more CARB than it is EPA. EPA is interested from the perspective that they need to understand how to label it. CARB really doesn’t want you to burn fuel if you don’t have to.

The message on charge sustaining mode is we’ve got a good beat on it, its the four or five things that have to be balanced together and the team needs time to work.

Will the engine rev higher when you step on the accelerator?
Yes, but the question is, if I have a little battery reserve at that time, do I turn the engine on right away to follow that or is it a smoothing function?

Also lets say you just went up a monster hill and the engine feathered up a little to support you and you get to the top of the hill and hit a light. Should the engine keep on running to allow you to recover or do you turn off the engine because the customer expects it too because they’re stopped at a light?

Technically are these difficult challenges? No. Its hard development work balancing the calibration of a lot of software.

He recently shed some light on how the battery pack detects and reacts if there are any problems.

“The reliability of the Chevy Volt and its battery are essential to the success of this technology,” he said. “Our battery design includes multiple computers that run hundreds of tests to monitor the cells and the overall battery to confirm everything is working correctly.”

He explains that all of this testing is continuous and automatic and that the system is designed to respond in case of trouble.

“If any of these tests identify an issue a diagnostic trouble code (DTC) is set to identify the specific issue and an indicator light on the instrument panel will light informing the driver that service is needed,” said Ceisel.

Ceisel also notes GM’s satellite system OnStar can help out both the driver and the mechanic in this situation.

“If the driver wants immediate feedback, OnStar can access these codes to provide more information,” he said. “These codes will also provide the dealer technician with specific information on the repair required.”

He explains how the battery cells are configured both in parallel and series and how the failure of some cells wont take down the car. If need be, the generator can also be used to help out.

“If a diagnostic issue is detected the control system will adapt to use available battery power and, if needed, the gas engine/generator to maintain propulsion until the vehicle can be serviced,” he said.

Source (GM)