ESS software questions …

[PV4EV]

This is a long time lurker's first post … be gentle! I have a Roadster 2.5 due to arrive in a few weeks time, and I'm counting the days. Below are a few questions that I'm curious about to which I cant find answers from older threads, although I can see the subject area has been discussed before.


One thing offered to me at purchase time was the $12k / £9k payment for whatever future ESS replacement solution Tesla Inc might have, if/when the existing ESS keels over, hopefully at well over 100,000 miles time. They hint that a replacement ESS may be something quite different from the existing unit.

Meanwhile, I'm developing a consumer electronic device subject to patent that uses high capacity 18650's, and energy storage is something I'm very interested in generally. I am aware of rapid developments in energy storage research including things like Lithium Air based cells, Graphene based ultra-caps, and other emerging possibilities, all of which hold out the promise of 500 to 1000+ miles 'range' in 5-10 years time …


So, what is possible with an ESS upgrade right now ? …

The ESS uses #6831 x 18650 cells from Panasonic rated at a nominal 2.1Ah, and was designed / developed / refined between 2005-2006, or thereabouts.

Since 2010 Panasonic have been able to supply 18650s rated at approx 3.0Ah.

According to their datasheets, the two cells have identical thermal / electrical characteristics and current cutouts for the same load conditions with the only difference being that the energy capacity is +40% and the charging profile is different to the original cells, and the car would take longer to charge. If the cars performance / power demands remain the same spec, then surely all the thermal controls and SOC monitoring will be, broadly speaking, the same.

If Tesla offered upgrades to the ESS with newer cells it would allow them to claim a headline grabbing genuine 300+ mile ability for the Roadster … or even 400miles if the forthcoming 4Ah cells are use.

If enhancements like this were announced to support the 'older car', combined with all the Model S PR and its 320mile range, then surely the range anxiety arguments from the ICE industry all start to look a bit weak ? ( And both my last ICE cars could not average more than about 280 miles on a tank full… with only 140miles if I hammered them everywhere!)


So :-

(1) Does anyone know if the existing Roadster software has the hidden-facility to initialize and balance an entire ESS fitted-out with a set of new higher capacity 18650's ?

(2) Do you think that the existing software (or via an upgrade) allows for altering the charge profile for slightly different cells?

(3) Do you think that as Roadsters fall out of warranty in the years ahead, that a 3rd party support network will form offering a reverse engineered ESS cell upgrade ?! (and hopefully not as a result of Tesla going belly up - I've just bought shares too!)


Can anyone offer some insight on the above ?!

[qwk]

I'm very interested in this idea also. Unfortunately, unless you are an engineer or higher up at Tesla, you won't know a definite answer to these questions. My speculation is that they have messed around with higher capacity cells in certain roadsters(Elon's for example).

Fitting higher mah cells in the roadster pack is probably not that difficult. The biggest headache would be the software for the PEM/BMS. I'm quite certain that as soon as enough roadsters come out of warranty, people will tinker and find out just how difficult this proposition is.

[fraccy]

The only insight I can offer is, I've had my roadster about 6 weeks, and spent the 2 months waiting for it fussing over every aspect of the possibilities of battery improvements and what-if upgrade scenarios.

Since getting the car, I've not given any of it a second thought, and the only area I feel let down in is the lack of 3-phase charging. The car might just be fine as it is. Just my two cents.

[TEG]

Fitting higher mah cells in the roadster pack is probably not that difficult...

Keep in mind that with 6831 of them, even a "small job" for an individual cell becomes a large job for a whole pack.
Also, how are they packed in the sheets? Can an "amateur" put cells in there by hand, or would you need some sort of custom machine / welding / glue / ?
Basically, I wouldn't assume that it is an easy job. It might not be.

[qwk]

Keep in mind that with 6831 of them, even a "small job" for an individual cell becomes a large job for a whole pack.
Also, how are they packed in the sheets? Can an "amateur" put cells in there by hand, or would you need some sort of custom machine / welding / glue / ?
Basically, I wouldn't assume that it is an easy job. It might not be.Yes, certainly very time consuming, but not that difficult. The cells are glued/soldered in. I had a link to diagrams/pics of the sheets saved on my old computer.

[Lloyd]

Yes, certainly very time consuming, but not that difficult. The cells are glued/soldered in. I had a link to diagrams/pics of the sheets saved on my old computer.


http://www.youtube.com/watch?v=zScUKZwTPHw

Other manufacturer, but interesting nonetheless.

[PV4EV]

Yes, certainly very time consuming, but not that difficult. The cells are glued/soldered in. I had a link to diagrams/pics of the sheets saved on my old computer.


There's virtually nothing I can find in the public domain about the actual internals of the bricks and sheets within the ESS. The original 2008 Tesla patent on the ESS provides some pictorial detail on how they deal with thermal management and the individual cell mounting :-

Mitigation of propagation of thermal runaway in a multi-cell battery pack - Tesla Motors, Inc. (http://www.freepatentsonline.com/7433794.html)

From a purely engineering prospective I would really like to dismantle an ESS and see how they solved the numerous problems getting the whole thing to work as well as it does in a production roadster.

For anyone who hasn’t seen this, here's a 20min presentation by JB Straubel that gets pretty interesting from about 12/13 mins in, as well as showing some interesting stats when comparing various efficiency and performance increases of the Roadster to the Model S :-

2011 Summit Video Presentation: JB Straubel | Institute for Energy Efficiency (http://iee.ucsb.edu/content/2011-summit-video-presentation-jb-straubel)

[PV4EV]

... the only area I feel let down in is the lack of 3-phase charging. The car might just be fine as it is. Just my two cents.

I agree, especially since I have 3-phase at home and its a tad annoying not being able to use it :mad:

I'm hoping Tesla listen to requests for 3 phase support from prospective European Model S owners, myself included.

The USA Model S looks set for Q2 2012 release, so the Euro versions might be 12+ months behind (Q3 2013 ?) and hopefully they will come up with a workable engineering solution for Euro cars.

[Mycroft]

Even the first Roadster owner isn't getting close to the point where they would get a battery refresh so anything would be pure speculation for at least four more years.

From a short video I saw, it looks like lithium air cells are at least 15 years out. Hopefully ultra-caps and better Li-Ion batteries will be viable much sooner.

Like you, I'm also hoping we'll be able to retro-fit our cars with the new battery tech when prices come down. Wow, a 1,000 mile range on my Model S Signature Sport. Now THAT would be cool!

[richkae]

Projecting from my last 2 years of usage, I dont anticipate *needing* a new battery before mine is 10 years old.
However if Tesla offered a new battery in 2-3 years that increased both range and performance I would be very tempted to upgrade from my completely adequate battery.

[Mycroft]

Yeah, I think that most people who pre-bought the battery refresh will postpone it as long as possible.

[qwk]

Yeah, I think that most people who pre-bought the battery refresh will postpone it as long as possible.This would only make sense as you would get cash back if you wait over 7 years...

Why get a battery replacement if the battery you have is adequate for your needs?

[PV4EV]

Yes, certainly very time consuming, but not that difficult. The cells are glued/soldered in. I had a link to diagrams/pics of the sheets saved on my old computer.


There's virtually nothing I can find in the public domain about the actual internals of the bricks and sheets within the ESS. The original 2008 Tesla patent on the ESS provides some pictorial details on how they deal with thermal management and the individual cell mounting :-

PATENT LINK Mitigation of propagation of thermal runaway in a multi-cell battery pack - Tesla Motors, Inc. (http://www.freepatentsonline.com/7433794.html)

From a purely engineering prospective I would really like to dismantle an ESS and see how they solved the numerous problems getting the whole thing to work as well as it does in a production roadster.

For anyone who hasn’t seen this, here's a 20min presentation by JB Straubel that gets pretty interesting from about 12/13 mins in, as well as showing some interesting stats when comparing various efficiency and performance increases of the Roadster to the Model S :-

2011 Summit Video Presentation: JB Straubel | Institute for Energy Efficiency (http://iee.ucsb.edu/content/2011-summit-video-presentation-jb-straubel)

[richkae]

I did not pre buy the Roadster battery replacement.

I *think* I will not be motivated to replace my Roadster battery until the range is below 150 miles - unless I can replace it with something that is superior to the original battery. Even though I wont need it, I will want it.
I know they can make a better battery, and I want the better one.

[qwk]

I did not pre buy the Roadster battery replacement.

I *think* I will not be motivated to replace my Roadster battery until the range is below 150 miles - unless I can replace it with something that is superior to the original battery. Even though I wont need it, I will want it.
I know they can make a better battery, and I want the better one.150 in Range mode, or standard mode?

[doug]

There's virtually nothing I can find in the public domain about the actual internals of the bricks and sheets within the ESS. The original 2008 Tesla patent on the ESS provides some pictorial detail on how they deal with thermal management and the individual cell mounting :-

Nothing new, but there are some drawings here: Roadster battery (ESS) (http://www.teslamotorsclub.com/showthread.php/3810-Roadster-battery-(ESS)

[Picasso]

Depending the price Tesla offers referb packs and or new packs I'm sure an after market option will come along. A few people no longer work for Tesla who know about the pack software.

Some roadsters now do not use the same cells. So yes software can be changed depending on cell.

[TEG]

...Some roadsters now do not use the same cells. So yes software can be changed depending on cell.

Intriguing!

[qwk]

Some roadsters now do not use the same cells. So yes software can be changed depending on cell.Can you be more specific? Is this heresay?

[Picasso]

Sure I can after Xmas when my NDA runs out.

[hcsharp]

You are asking the 64 dollar questions that many of us have wondered for some time. I've tried to answer below based on my own speculation.



So :-

(1) Does anyone know if the existing Roadster software has the hidden-facility to initialize and balance an entire ESS fitted-out with a set of new higher capacity 18650's ?

It's likely that it does. But it depends on what software you are talking about. There are circuit boards mounted within the ESS itself with firmware in them. I don't think all of these get updated when you get a so-called software upgrade. I don't think all of them are even connected to the bus. That is based on what I was told by a Tesla tech who said he had disassembled the ESS. I think they could easily update this firmware when rebuilding the ESS with new cells. I suspect they will offer an upgrade to the $12k owners to get more than 53 kWh.

I have also heard of Roadsters with particularly good batteries that after being cycled a few times show more than 245 ideal miles remaining on the vds. That would indicate that there is at least some capacity in the hardware/firmware for bigger cells already. Just how much is anybody's guess.


(2) Do you think that the existing software (or via an upgrade) allows for altering the charge profile for slightly different cells?
Yes. They added the ability to balance the cells when doing a standard mode charge in a recent software update. They tweaked the charging profile quite a bit on the original prototypes so it's reasonable to assume they have the capacity to do this without too much trouble.


(3) Do you think that as Roadsters fall out of warranty in the years ahead, that a 3rd party support network will form offering a reverse engineered ESS cell upgrade ?! (and hopefully not as a result of Tesla going belly up - I've just bought shares too!)


Not sure about this but I doubt it. Tesla has automated equipment to solder and glue the cells that might be too costly to setup for such a limited market. In addition, Tesla is building recycling facilities to re-use most of the material in the cells. They claim to recoup some of the cost that way. It also really depends on whats in there! Some of it might be impossible to reverse engineer (electronics, firmware), or too expensive given the small market.

A service tech told me it takes about a whole day (8 hrs) to swap a ESS out of a Roadster if he periodically gets help from another mechanic. That's a lot of shop time! When I heard that I decided $12k was a good deal for the battery replacement program.

I also think Tesla will offer better cells for an upgrade price to those of us who bought the program.

[richkae]

150 in Range mode, or standard mode?

Its hard to predict how I will feel about it in the future, especially when I have the Model S because it will do any of my long distance driving.
If right now magically today my Roadster battery only had 150 mile range - in range mode ( so about 115-120 standard ), I would be pestering Tesla and asking them for a new battery made out of the same cells in the top of the line Model S.
I would wait for that battery to be available and would not buy a replacement battery made of the old cells. If they told me I had to wait until next year, I would wait.

Its purely an academic exercise. I have 99.9999% confidence that they will offer an upgraded battery before my capacity drops below 200 miles.

[PV4EV]

Nothing new, but there are some drawings here: Roadster battery (ESS) (http://www.teslamotorsclub.com/showthread.php/3810-Roadster-battery-(ESS)


Thanks for the link to the images !!

I know there's supposed to be 9 bricks x 69 cells = 621 per sheet, 6831 total.

However, its hard to know how precise the drawings were supposed to be, or if the artist couldn’t count … because influenced by a bit to much wine and 10 minutes to spare, I counted 627 cells in each sheet, giving 6 more than stated, and 6,897 total !

Maybe I should sober up and count again ...


What the illustrations show nicely is how densely the cells are packed and how the cooling system is plumbed in - details I've not seen before.

[hcsharp]

It looks like whats going to cause your ESS to lose capacity is when you burn out a few cells here and there. If there are currently 69 cells in parallel in each brick, the current is divided between them. But if 6 burn out in one brick, the other 63 have to carry the load which means they will burn out faster. At some point it will reach a tipping point where the cells in that brick will start going dead faster and faster until the entire brick dies. I don't know if the ESS is useless at that point or the hardware has the capacity to bypass it. In the mean time, your total capacity is limited by that one brick or the sheet it's in. You can't ever discharge it below 3v while the rest of the bricks will still be at, say 3.4v.

You are more likely to "kill" cells when they are low. The car passes more current through them to get the same power so you are more likely to heat them up. This theory would support keeping the car charged in std mode every day as opposed to keeping it closer to 50% SOC with the idea that a little lower SOC is better for the hours you're not driving it.