Compare to the Volt

[DDB]

I thought this article about the Volt's design was too interesting to fail to mention. The Tesla is even compared here. I was absoutely shocked at GM's testing of the batteries. Although Tesla is mentioned once in the article, I suppose Tesla didn't see a point in testing its pack 1,000 ways from Tuesday like GM is doing.

And I understand Tesla needed more dense packs for range etc...but wow, anyone ought to be impressed with GM, as ugly as their past reporting quarters have been.

 

[DDB]

A link would be nice:

Chevrolet Volt Update - Volt Setup - Auto News - Motor Trend

 

[TEG]

How do you know Tesla didn't test their pack so thoroughly?

 

[doug]

Might help to have a link to page 1 even:
http://www.teslamotorsclub.com/technical/1070-compare-volt.html#post7225

I thought Tesla did quite a bit of testing with their pack, primarily with regard to safety.

 

[BBHighway]

Tesla put a lot of time and effort into thoroughly testing the batteries.

It was the transmission that they didn't pay enough attention too, at least not until way too late in the development process. I think they are plenty focused on it right now.

 

[vfx]

Yet one more try at the link:

Chevrolet Volt Update - Auto News - Motor Trend

 

[stopcrazypp]

Tesla put a lot of time and effort into thoroughly testing the batteries.

It was the transmission that they didn't pay enough attention too, at least not until way too late in the development process. I think they are plenty focused on it right now.

Yep, I remember quite clearly Tesla did very extensive tests on their battery packs. After all, it's pretty obvious the public is a little nervous with more than 6800 laptop battery cells in the car after hearing about li-ion battery explosions, and I don't think many people would put their deposits down if the batteries went untested.

 

[DDB]

Yep, I remember quite clearly Tesla did very extensive tests on their battery packs. After all, it's pretty obvious the public is a little nervous with more than 6800 laptop battery cells in the car after hearing about li-ion battery explosions, and I don't think many people would put their deposits down if the batteries went untested.

That's my point, we're yet to see how the packs go. Tesla did it all in-house whereas GM outsourced, another obvious difference. And I never said Tesla didn't test the pack. I'm just impressed by GM's proprietary software and how they went about testing the TWO packs.

 

[stopcrazypp]

That's my point, we're yet to see how the packs go. Tesla did it all in-house whereas GM outsourced, another obvious difference. And I never said Tesla didn't test the pack. I'm just impressed by GM's proprietary software and how they went about testing the TWO packs.

Ah, I see what you mean. GM's testing is more of a result in that they have a competition between two suppliers, they are using newer chemistry while Tesla is using very established conventional li-ion, and they actually have to supply a 10 year warranty, which involves probably more extensive testing than the 5 year life of the Tesla.

This pdf mentions:

"Upon completion of our design, we collaborated with an outside firm known for expertise in lithium-ion batteries to perform hundreds of tests to validate the abuse tolerance and effectiveness of our design."
http://www.teslamotors.com/display_data/TeslaRoadsterBatterySystem.pdf
Seems like they did rely on some third party help, but most of the results obviously ended up internal.​

But looking over the tests on the Volt batteries, there were accelerated life-cycle tests, climate (temperature) testing, and shock and vibration testing. I'm fairly certain Tesla did all of those kind of tests in one way or the other. Yes, we don't know much about the results, except they "passed".
This one mentions calendar and cycle life (makes it sound like they relied on the calendar and cycle life of the battery cells themselves, designed the pack to keep them at optimal condition and didn't sound like they did testing):
Tesla Motors - think
This blog post mentions temperature testing for the whole car:
Tesla Motors - touch
This blog post mentions shock and vibration testing for the batteries:
Tesla Motors - touch

The Volt testing doesn't really mention safety testing, but I suppose similar to how Tesla relied on battery cell specs to explain calendar and cycle life, GM relied on the safety specs of the battery cells themselves for safety (or more likely they tested safety, but didn't show it or Motor Trend didn't report, and just like it's more likely Tesla did do some battery life tests to ensure their claims are at least close). Tesla obviously had done safety tests, as mentioned in the same blog that talked about the shock and vibration testing.

The expectations of GM, a large mainstream automaker, is much more in that after 10 years the Volt is expected to STILL get 40 miles per charge. The Tesla is expected to have only around 80-70% of capacity after 5 years or 100,000 miles, whichever comes first. But testing on both is still fairly extensive, with the Volt showing more of the testing process (pretty much unprecedented amount of exposure to the development of a vehicle by a major automaker, but obviously the Volt is also a very good PR tool, with the hype on it probably even more than for the Roadster).

 

[malcolm]

Think we're comparing depth of press releases rather than range and extent of testing methodologies.

A great deal of Tesla's ESS work is bound up in their "secret sauce" - the various proprietory solutions they have developed over the past six years for the electrical systems. The basic safety work is covered; pack durability in the real world remains an unknown for both manufacturers.

Any sports car will exert greater wear and tear on its component parts than a family saloon. We need to compare the Volt with Whitestar - and by the time that appears, Roadster customers will have conducted thousands of miles of real-world road tests and hundreds of recharges on Tesla's behalf.

Obviously GM only needs to be able to claim modest mpg improvements, so they can afford to drive the generator system harder to reduce the depth of discharge on each battery pack.

Even so, it will be interesting to see if the Volt can hit its 40 mile battery-only claim. With an aluminium and carbon fibre car, Tesla can get 245 miles from 53kWh on the EPA Highway. That's about 4.6 miles/kWh of saloon-like motoring behaviour. So an aluminium and carbon fibre Volt (ha!) using only 8 kWh should get about 37 miles.

Except that the Volt (and Whitestar) will be heavier.

GM are going to have to use more of the Volt's 16 kWh energy storage capacity to hit that 40 mile range. Which will reduce calendar life.

 

[DDB]

The expectations of GM, a large mainstream automaker, is much more in that after 10 years the Volt is expected to STILL get 40 miles per charge. The Tesla is expected to have only around 80-70% of capacity after 5 years or 100,000 miles, whichever comes first. But testing on both is still fairly extensive, with the Volt showing more of the testing process (pretty much unprecedented amount of exposure to the development of a vehicle by a major automaker, but obviously the Volt is also a very good PR tool, with the hype on it probably even more than for the Roadster).

Know what's funny about these models? IMO as much as you can create an environment that mimics wear and tear, you can't do a 10 year test without waiting 10 years. We aren't ever going to know for sure until the clock hits 2018. But by then we'll be in the 3rd, 4th, or even 5t generation won't we, so it's probably a moot point.

 

[TEG]

...you can't do a 10 year test without waiting 10 years...

Companies try all the time to get away with simulations and projections.

I recall once a major hard disk manufacturer put long ( > 5 years) MTBF ("Mean Time Between Failure") stats on their new drive based on the worst case expected lifespans of the individual components. The whole batch was made with the wrong lubricant in the motor bearings, and they had widespread failures within a year. That kind of stuff makes you suspect durability projections for new products.

We had a similar problem with bad capacitors (again a manufacturing goof with the wrong chemistry in the component) in a large number of desktop PC from a major manufacturer. Like clockwork they all started to fail within a short time. Looking inside, the capacitors were obviously "oozing goop" all over. We ask the manufacturer to replace the whole lot. They said "there weren't enough reported failures yet for a recall, so they would send a technician out to repair the faulty units". We tried to get them to swap motherboards on all systems at once but instead we had to go through a ritual of calling the repair depot back every few days as another system would go out.

Big jet planes used to need to log lots of test flight hours before they were certified for use, and rules for trans oceanic flight used to call for more than 2 engines. A recent model of popular jet was granted immediate exceptions to the usual rules based on computer projections. On its' maiden flight with company execs aboard the jet engines experienced a flame-out condition that wasn't predicted by the computer models.

Oops.

 

[stopcrazypp]

Know what's funny about these models? IMO as much as you can create an environment that mimics wear and tear, you can't do a 10 year test without waiting 10 years. We aren't ever going to know for sure until the clock hits 2018. But by then we'll be in the 3rd, 4th, or even 5t generation won't we, so it's probably a moot point.

Well, as TEG mentioned, this is just the standard way of doing testing in most industry. The Volt really doesn't have 10 years to wait around if they want to release their car in 2011 or 2010. So budget and time constraints really cut into testing time and effectiveness, esp. for smaller companies. Which is why was so hard to make a mainstream car these days, there is so much testing. But I think we are glad that Tesla went through at least most of the tests, including the federal ones. In some of the smaller companies, like the ones who sell "quadricycles", there will probably be close to NO tests; saves them a lot of time and money.

Actually like malcolm mentioned, this exposure of testing is more like a PR blitz, designed to assure the public and media that the cars will do well. In the end, we may not get the full results (besides from pass/not pass) or most of the details of the testing. That's likely to stay internal, unless there is another PR round at the release of the car (since GM obviously wants this to be a crown of technical achievement for them and it's also one of their responses to the Prius and criticism of the killing of the EV1).

 

[malcolm]

Ooops!

I got the right answer, but used the wrong numbers.

The Roadster gets 220 miles on the EPA Highway (combined cycle) not 245. However, not all of the Roadster's 53kWh is available to the motor. 10/11th of the ESS energy can go to the motor and 1/11th powers the car's computer systems etc.

Which gives 220/48.2 = 4.6 miles/kWh.

Anyway, the heavier Volt will still need 5 miles/kWh to get 40 miles from only 8kWh. GM are going to have to increase the depth of discharge - maybe as much as 90% to 20% instead of 80 to 30.

Or drop the commitment to the 40 miles battery-only range.

Or use a gentler driving test cycle than the EPA Highway.

On a completely different line of thought, what would happen if the ESS could change the "sheet" of batteries used to power the car's electronics systems? Every few hundred miles (or every 5th recharge - whatever) the system switches over and a different sheet gets a rest from powering the motor and is required to take its turn powering the other subsystems.

Or perhaps the changeover rate should be controlled by average pack temperature - in hotter seasons this sheet-cycling behaviour is faster than in winter.

 

[malcolm]

The Volt's crash diet has started:-

GM-VOLT : Chevy Volt Concept Site » Blog Archive » Volt Gas Tank Shrunk - Maximum Range Reduced

Hope they can bring the weight down.

 

[Albern]

An interesting post on the Volt website - from 2 12 gallons tanks to a single tank with a yet to be determined volume. What really caught my attention was the 50 MPG fuel rating in range extended mode. Can ICEs be that "efficient" when generating electricity for a vehicular application?

 

[malcolm]

Can ICEs be that "efficient" when generating electricity for a vehicular application?

Absolutely. The big advantage is that the ICE gets to run at optimum revs the whole time. Bit noisy in quiet traffic of course.

 

[SByer]

Heck yeah. So, assume that a company the size of GM can get close, if not match, the efficiency of the Roadster drive train, say, at 90%. Now, take your ICE, size it perfectly for the matched generator, and run the pair at the optimal efficiency RPM point, tuning the generator for the ICE, not the other way around.

From another view, take an ICE torque curve. Flatness comes from oversizing the ICE. Right size the ICE to provide a good torque peak - probably in the 3-4k rage - tune the matched generator to that. Much smaller engine than you are used to.

Overall, you're taking an engine that has pretty narrow efficiency bands, running it in that band, and driving a system that has a very wide efficiency band that's also quite high. Average goes up as long as the transfer of power isn't too inefficient.

 

[mt2]

Overall, you're taking an engine that has pretty narrow efficiency bands, running it in that band, and driving a system that has a very wide efficiency band that's also quite high. Average goes up as long as the transfer of power isn't too inefficient.

And add to the equation that all energy being created by the ICE is being stored (within the efficiency limits of the generator and battery). No wasted energy by sitting idle at red lights or in grid locked traffic.

 

[malcolm]

Positive noises continue:

GM-VOLT : Chevy Volt Concept Site » Blog Archive » Even Chevy Volt Battery Suppliers Surprised at GM’s Volt Aggressiveness

Love the quote from Ric Fulop,"Somebody lit a fire under [GM’s] butt in 2006........I’ve never seen a large company move so fast and put so many resources behind something."

Bob Lutz is on record citing Tesla as the Firestarter:

Q&A: GM's Bob Lutz on the Volt and More - US News and World Report

When Tesla announced they were building a car, that kind of tore it for me. I thought, "If some little West Coast outfit can do this, we can no longer stand by."