Third Party Roadster Pack?

[ecarfan]

Third Party 1.5 Battery Replacement???

Read this over at Edmunds.com:

Thank you for the link. My 1.5 is out of warranty so I'm not concerned about that issue, but remain concerned about Tesla refusing to service my Roadster if a third party aftermarket battery is installed.
I applaud the OP's intent and wish him the best in his endeavor.

 

[efusco]

Tell me where I'm off base....
1)You've never done this before on a Tesla Roadster
2)You want someone to loan you their car and pay you $30,000 for the priveledge in the hopes that they might get a new battery pack
3)You can offer no assurance that they will get a functional vehicle back at the end of your experiment
4)You have nothing to loose and everything to gain, while the owner has a potential gain and a lot to lose.

IMO you should do one of the following:
1)Get a loan, buy a used/wrecked/salvage Roadster and do your own proof of concept to then take to market; or
2)Offer to do the install for free for someone willing to put their own Roadster at risk.

The concept is interesting, no doubt about that, but I pity the fool that decides you to take you up on this offer unless I'm clearly off base on my original points above.

 

[ElectricLove]

Tell me where I'm off base....
1)You've never done this before on a Tesla Roadster
2)You want someone to loan you their car and pay you $30,000 for the priveledge in the hopes that they might get a new battery pack
3)You can offer no assurance that they will get a functional vehicle back at the end of your experiment
4)You have nothing to loose and everything to gain, while the owner has a potential gain and a lot to lose.

IMO you should do one of the following:
1)Get a loan, buy a used/wrecked/salvage Roadster and do your own proof of concept to then take to market; or
2)Offer to do the install for free for someone willing to put their own Roadster at risk.

The concept is interesting, no doubt about that, but I pity the fool that decides you to take you up on this offer unless I'm clearly off base on my original points above.

From a buyer perspective this is all very intriguing, I'll address your points:

1; correct, it has not been done, it is uncharted territory.
2; I am not asking them to "loan" us their car, I'm not looking for a joy-ride; we have a Tesla Roadster here, but it is a 2.5 and still under warranty. I just sold my Roadster last month, it was a 2.0 though... Our objective with the vehicle would be to rebuild the battery pack and we could have most of that process done prior to receiving the vehicle itself, after we have the vehicle here we would drop the battery pack/box, open it up and replace the sheets with the new ones and put it back together. This process has been done before using Tesla provided battery cells.
3; I can offer the assurance of a fully functional vehicle back at the end, if you know much about batteries you would know they are "dumb" they don't actually communicate or do anything other than hold potential energy, there are few conceivable reasons why this swap would fail... And ultimately if it did we would swap back to the original cells and obviously NOT charge the customer.
4; The customer has the potential to gain and also be part of innovation, something that Roadster owners are notorious for...

The last 2 points; 1-Our company is a legitimate company with plenty of money to buy a Roadster but it may be easier to get this initiative going if we have a "first sale" opportunity already in hand: However, at $30K we aren't actually looking to make much money on this, it would be a proof of concept and experience opportunity. Also, the installation labor would be FREE (#2), we are only charging for the batteries and we would be charging at our cost on this first opportunity, this means the price is likely to be less than $30K in the end but I don't want to build false hope or make false promises, I'm not in marketing, I'm in R&D! The worst case scenario is $30K cost, the likely scenario is something more like $25K...

Anyways, I appreciate your feedback and I definitely agree it would be really easy for any owner to do this with us if we did it for free but that would be a $20-30K investment from us to prove a concept in a market that has incredibly limited opportunities (there are only 500 1.5's out there, less than that are in the USA, less than that will ever hear about us and less than that will actually be driving them enough to care about the battery...) So for $20-30K we prove to a market of about 50 people that we can do it... That isn't financially worth it for us, their is some reciprocity with our offer in that we are passionate about Electric Vehicles and do actually want to see Roadster's with improved range and capabilities!

No one would design a business around supporting just 50 or so vehicles, unless those vehicles were $100M airplanes... That isn't our case, we are looking at building custom battery pack solutions for all, if we just wanted to prove a concept we would do it with a smaller application, and in fact, we have already!

 

[ElectricLove]

I would be willing to bet some bucks that Tesla also do "coulomb counting" to protect them against charging a damaged battery which may appear to continue charging at the peak voltage but in fact have significant internal leakage. So I would expect the car to cut off not much further than 10% over the rated kWh (or Ah capacity in this sense), perhaps even logging an error. Hence why Tesla need to update PEM firmware with the 3.0 upgrade.

I sort of doubt this is actually the case, its possible, but unlikely, an error in the BMS would be triggered by out-of-balance current absorption but I wouldn't imagine it happening across the board if the cells are all balanced. It won't be able to identify the starting point, so if it coulomb counts it is going to do it from "full" down to empty not from empty to full. Think about a Roadster that ran out of juice and sat for a few weeks and just before becoming a bricked pack gets charged, this pack would take much more Amps in before reaching the "full-point" and the BMS likely wouldn't cut it off short and throw an error (I've witnessed this process, it took a lot more energy than typical to get full but it still got there and started balancing the cells at the end...

Come to think about it, assuming it was doing coulomb counting strategy; I wonder if it would save the excess energy in the TOP or BOTTOM of the SOC range? OR distribute it evenly between both? If it were evenly distributing that would mean you could do full-range-mode charges at all times and never really fully-charge the pack and also never fully-discharge the pack running it down to ZERO, this would result in a pack that lasts a much longer time and maintains its "usable" capacity for a much longer time... It may not give you the predicted range adjustments the 3.0 from Tesla would but I still tend to think it would be a significant improvement over what would otherwise be available...

To your point also, if we ran into that wall the worst case would be do decrease the number of cells, this would shave about 200-300 pounds off the weight of the pack but maintain the same capacity; this process would also result in much lower pricing (maybe $20K would pay for the whole new thing if we take on that strategy?)... The cool thing here is that it is "custom" so that means we can make it exactly what the customer wants it to be! (and feels comfortable with)

 

[efusco]

From a buyer perspective this is all very intriguing, I'll address your points:

1; correct, it has not been done, it is uncharted territory.
2; I am not asking them to "loan" us their car, I'm not looking for a joy-ride; we have a Tesla Roadster here, but it is a 2.5 and still under warranty. I just sold my Roadster last month, it was a 2.0 though... Our objective with the vehicle would be to rebuild the battery pack and we could have most of that process done prior to receiving the vehicle itself, after we have the vehicle here we would drop the battery pack/box, open it up and replace the sheets with the new ones and put it back together. This process has been done before using Tesla provided battery cells.
3; I can offer the assurance of a fully functional vehicle back at the end, if you know much about batteries you would know they are "dumb" they don't actually communicate or do anything other than hold potential energy, there are few conceivable reasons why this swap would fail... And ultimately if it did we would swap back to the original cells and obviously NOT charge the customer.
4; The customer has the potential to gain and also be part of innovation, something that Roadster owners are notorious for...

The last 2 points; 1-Our company is a legitimate company with plenty of money to buy a Roadster but it may be easier to get this initiative going if we have a "first sale" opportunity already in hand: However, at $30K we aren't actually looking to make much money on this, it would be a proof of concept and experience opportunity. Also, the installation labor would be FREE (#2), we are only charging for the batteries and we would be charging at our cost on this first opportunity, this means the price is likely to be less than $30K in the end but I don't want to build false hope or make false promises, I'm not in marketing, I'm in R&D! The worst case scenario is $30K cost, the likely scenario is something more like $25K...

Anyways, I appreciate your feedback and I definitely agree it would be really easy for any owner to do this with us if we did it for free but that would be a $20-30K investment from us to prove a concept in a market that has incredibly limited opportunities (there are only 500 1.5's out there, less than that are in the USA, less than that will ever hear about us and less than that will actually be driving them enough to care about the battery...) So for $20-30K we prove to a market of about 50 people that we can do it... That isn't financially worth it for us, their is some reciprocity with our offer in that we are passionate about Electric Vehicles and do actually want to see Roadster's with improved range and capabilities!

No one would design a business around supporting just 50 or so vehicles, unless those vehicles were $100M airplanes... That isn't our case, we are looking at building custom battery pack solutions for all, if we just wanted to prove a concept we would do it with a smaller application, and in fact, we have already!

So split the battery cost 50:50 so you both assume some risk and benefit. I think there is risk for the owner here when dealing with the unknown when it comes to HV batteries. Maybe you guys really do know your stuff. I don't have a dog in this fight so it will be my last post. But I think you're asking an awful lot of an owner. Now, while the 500 1.5 Roadster numbers is likely accurate, there are another 2000 or so roadsters out there that once you've proven this concept on the 1.5s you could expand to the other versions. Also, if you're that confident in your process, then it seems reasonable to do it on the car you have, whether under warranty or not.

 

[TOBASH]

Tell me where I'm off base....
1)You've never done this before on a Tesla Roadster
2)You want someone to loan you their car and pay you $30,000 for the priveledge in the hopes that they might get a new battery pack
3)You can offer no assurance that they will get a functional vehicle back at the end of your experiment
4)You have nothing to loose and everything to gain, while the owner has a potential gain and a lot to lose.

IMO you should do one of the following:
1)Get a loan, buy a used/wrecked/salvage Roadster and do your own proof of concept to then take to market; or
2)Offer to do the install for free for someone willing to put their own Roadster at risk.

The concept is interesting, no doubt about that, but I pity the fool that decides you to take you up on this offer unless I'm clearly off base on my original points above.

These are exactly my concerns, and why I posted earlier.

I applaud the effort BUT this needs to be done on a car with a bricked battery that is on a salvage title that Tesla has already refused to service.

No way anyone should have to fork over 30K for the privilege of allowing this person to experiment on his/her car. No way they should have to spend anything. It should be FREE. You also need to insure the value of the car for damage, because there is a very real risk that you will damage things beyond repair. There are mechanical issues as well as proprietary computer issues that will need to be hacked. Without everything being dealt with you will damage the vehicle and it's value beyond repair.

If someone wants to start a business they should assume all costs of research or INCORPORATE to allow others who share the risks also to share the future potential financial benefits.

ElectricLove, I appreciate what you are trying to do BUT I think you are going about it the wrong way.

As it sits right now, your offer is lacking.

 

[ecarfan]

Third Party 1.5 Battery Replacement???

1; correct, it has not been done, it is uncharted territory.
2; ...Our objective with the vehicle would be to rebuild the battery pack and we could have most of that process done prior to receiving the vehicle itself, after we have the vehicle here we would drop the battery pack/box, open it up and replace the sheets with the new ones and put it back together. This process has been done before using Tesla provided battery cells.
3; I can offer the assurance of a fully functional vehicle back at the end, if you know much about batteries you would know they are "dumb" they don't actually communicate or do anything other than hold potential energy, there are few conceivable reasons why this swap would fail...
The worst case scenario is $30K cost...

I really appreciate your intent and what you are trying to do, however I would like to point out that if modifying a 1.5 battery to use modern cells was simply a matter of replacing the original cells with some type of modern cells then it seems to me that Tesla would have done just that. But Tesla has not done that so there must be other significant issues involved that we are not aware of, don't you think?
You wrote: "This process has been done before using Tesla provided battery cells." Are you saying that you have done that using cells you bought from Tesla? And those were the same cell type as what you were replacing, correct?

 

[supersnoop]

It should work in a 1.5 Roadster which Tesla said they will not build.

When?

Tesla has not said that they will not build one. All they have said is that the current offering is incompatible with the 1.5 melts chill on the hyperbole.

 

[PokerBroker]

This may be necessary at some point and I hope someone with a bricked pick is willing to give this a shot. Unfortunately I can't afford to take that risk at this time. Good luck! I'll be watching!

 

[spaceballs]

... we have a Tesla Roadster here, but it is a 2.5 and still under warranty. ...

If I was ElectricLove I would Just offer the 2.5 in exchange if the car being upgraded somehow gets destroyed. So no risk for the buyer, unless it was an 2.5 Sport.

 

[bart513]

Just the thought of a potential option (even as a 2.5 owner) is relieving. I agree that the price point would have to be lower (new 2.5 battery is now at $29K) and hopefully get Tesla's blessing as an approved 3rd party provider.

 

[TOBASH]

ElectricLove:

I just want to point out that you are the same guy who sold his salvage Roadster because Tesla would not support it.

To many of us that implies that you were unable to keep up with the repairs, also implying that you are unable to properly keep up with repairs on PEM's and battery packs.

Your decision to sell your services immediately after selling your salvage car is strange and seems suspicious, as you essentially sold the perfect experimentation platform.

What you are proposing is not a simple battery swap. This is not a simple AAA battery swap on a toy car.

The batteries consist of sheets which are sealed units with epoxy everywhere. The batteries are controlled by IC boards to regulate charge and discharge rates. The batteries have integrated cooling pipes epoxied within their structure. The integrated circuit boards have numbered and coded chips that are "married" to the Roadsters' other computers via software and firmware.

You essentially need to re-engineer the battery charging and discharging software and hardware while re-engineering the cooling system and cooling software while bypassing the handshake protocols that are in the software and firmware required by these "married" pieces.

Unless you are an engineer or electronics wiz, I think you need to step back and look at the way you are presenting yourself.

There are many behind the scenes PM's questioning your motives and questioning your common sense.

I don't think any of us will let you experiment with our valuable vehicles, especially after you sold the perfect testing platform in a manner that makes you look like you are unwilling and unable to support such a project on your own personal vehicle.

I agree that reverse engineering a battery pack and the related hardware and software is a great idea, but I'm not sure you are going about this in the correct manner nor are any of us convinced that you have the facilities or the backround to undertake such a huge job when others who are very qualified are still struggling with the basics.

This is not meant as an attack. It is just observation and repeating what others are speaking behind the scenes.

Sorry,

T

 

[jeremyz]

ElectricLove, I applaud your trying to offer a third party battery upgrade. Assuming the Roadster pack isn't a brick, I wonder what the residual value of the cells/sheets would be. Maybe you could sell the removed cells to vaping Juggalos for $2 or $3 a cell? Or sell them into some type of storage application?

Since my roomba isn't very good at spot welding batteries, I think I would prefer to see a Roadster pack made out of large format cells. A battery pack that involves taking out and sticking back in 6831 cells by hand is never going to be cheap. In 20 years, is anybody really going to be making 18650 cells anymore? Tesla won't be making them from day one in the gigafactory.

I would rather see an upgrade to the Roadster like what driveeo did with their Pikes Peak Roadster (Electric eO PP02 at Pikes Peak International Hill Climb 2014 Drive eO). I like how their battery and drive electronics are so much smaller then the original Tesla versions (Tesla Roadster 360 by Drive eO | Flickr - Photo Sharing!). A 400 pound weight loss with 482hp and 600 ft-lbs of torque sounds delightful. Even better, put in 2/3 of the drivetrain from their last Pikes Peak car; 912hp in a Roadster sounds like Maximum Plaid to me.

 

[ElectricLove]

ElectricLove:

I just want to point out that you are the same guy who sold his salvage Roadster because Tesla would not support it.

To many of us that implies that you were unable to keep up with the repairs, also implying that you are unable to properly keep up with repairs on PEM's and battery packs.

Your decision to sell your services immediately after selling your salvage car is strange and seems suspicious, as you essentially sold the perfect experimentation platform.

What you are proposing is not a simple battery swap. This is not a simple AAA battery swap on a toy car.

The batteries consist of sheets which are sealed units with epoxy everywhere. The batteries are controlled by IC boards to regulate charge and discharge rates. The batteries have integrated cooling pipes epoxied within their structure. The integrated circuit boards have numbered and coded chips that are "married" to the Roadsters' other computers via software and firmware.

You essentially need to re-engineer the battery charging and discharging software and hardware while re-engineering the cooling system and cooling software while bypassing the handshake protocols that are in the software and firmware required by these "married" pieces.

Unless you are an engineer or electronics wiz, I think you need to step back and look at the way you are presenting yourself.

There are many behind the scenes PM's questioning your motives and questioning your common sense.

I don't think any of us will let you experiment with our valuable vehicles, especially after you sold the perfect testing platform in a manner that makes you look like you are unwilling and unable to support such a project on your own personal vehicle.

I agree that reverse engineering a battery pack and the related hardware and software is a great idea, but I'm not sure you are going about this in the correct manner nor are any of us convinced that you have the facilities or the backround to undertake such a huge job when others who are very qualified are still struggling with the basics.

This is not meant as an attack. It is just observation and repeating what others are speaking behind the scenes.

Sorry,

T

To address some of your points:

I didn't sell my Roadster due solely on the lack of support from Tesla; it played a role in my decision-making but the reality is I have never really owned/driven a car for more than a year and even the Roadster didn't keep me satisfied, I currently own 4 Electric Vehicles and it made sense to get rid of the Roadster b/c it was the most "needy" of the bunch. It shouldn't imply anything about my ability to conceive/design and deliver on a custom battery pack solution, I think the fact that I took the vehicle from inoperable to operable and did all of my own service/work on it should speak to the contrary. I have been inside the PEM on my Roadster, I have done some minor repairs of the electronics in the Tesla domain.

I totally agree with you at this point that I should have kept my Roadster and perhaps used it as a test-mule, however I had already decided to sell and sold it prior to conceiving the potential for this new business opportunity... Hind-sight is 20/20 after all!

I am aware of some of the complexities in doing this swap, unfortunately, until it is actually done it will be difficult to put a time/value to working through the process, which is why for the first "customer" I intend to do it at "parts-cost" alone. rewiring boards and modules is not a difficult thing to do, all of the "hardware" will remain intact as it is in the box, the "software" however tends to be a different animal altogether, that would be a challenge if it NEEDS to be changed, however my entrance to this arena would be under the assumption that the software can be left alone, we are using batteries with the same characteristics in every way EXCEPT capacity, the "computer/brains" won't be able to identify an LG branded cell vs a Panasonic branded cell; this I am confident of.

I believe there is less "re-engineering" than you may be assuming there to be, of course I could be wrong but I've worked on many EV battery packs and have worked with many BMS/etc... One thing for certain is that if you can figure out a way to leave everything in place and ONLY swap the cells for higher capacity/newer cells the BMS's tend to have no issues with that process, if they base capacity on "in/out" Ah counting that poses an issue of what range of the SOC will be utilized BUT, it wouldn't cause a BMS to become inoperable...

Anyways, I sincerely appreciate your concerns and hope to navigate through some of these waters, if you know anyone with a totally dead/bricked Roadster (salvaged would be GREAT!) I'd be a buyer, I'd even buy it from them with a promise/guarantee to sell it back to them afterwards at a $30K mark-up if they want it with the new battery pack...

 

[ElectricLove]

I haven't seen it posted in this forum yet, but an article in the NASDAQ news site saying that Tesla is in negotition with LG Chem for the 400 mile replacement Roadster battery. Sorry I didn't get the link.

While I no longer feel compelled to offer a 3rd party battery option for the Roadster I will say the cells I landed on are the LG Chem cells, they have some that just went into production this spring with excellent energy density as well as high discharge/charge ratings! I spent a lot of time investigating and testing cells and LG Chem (especially the HG2 model) came out on top! You could certainly take the MJ1 which is rated at 12.95Wh per cell (3.7V 3.5Ah) and get an 88.5kwh pack in the Roadster! This would be 400 miles capable... The cells cost right around $3 per unit when purchased en masse; so Tesla's cost for the batteries only would be around $21K, this likely fits into the $29K at cost model they are running...

I was getting excited about offering that exact pack as a 3rd party option, especially to 1.5 owners, but since Tesla is now demonstrating they will support the 1.5's, I'm not so concerned...

 

[tvuolo]

While I no longer feel compelled to offer a 3rd party battery option for the Roadster I will say the cells I landed on are the LG Chem cells, they have some that just went into production this spring with excellent energy density as well as high discharge/charge ratings! I spent a lot of time investigating and testing cells and LG Chem (especially the HG2 model) came out on top! You could certainly take the MJ1 which is rated at 12.95Wh per cell (3.7V 3.5Ah) and get an 88.5kwh pack in the Roadster! This would be 400 miles capable... The cells cost right around $3 per unit when purchased en masse; so Tesla's cost for the batteries only would be around $21K, this likely fits into the $29K at cost model they are running...

I was getting excited about offering that exact pack as a 3rd party option, especially to 1.5 owners, but since Tesla is now demonstrating they will support the 1.5's, I'm not so concerned...

What about offering other options? I know I will probably never road trip my Roadster or need 400 miles. I would be interested in a 250 mile, lighter pack, and associated suspension changes (if necessary).

 

[ElectricLove]

What about offering other options? I know I will probably never road trip my Roadster or need 400 miles. I would be interested in a 250 mile, lighter pack, and associated suspension changes (if necessary).

We could certainly figure something like this out... Something with same capacity but less weight or less capacity and even less weight, using cells more than capable of the discharge it would be subjected to...

Something "around" the same capacity as the original pack could weigh in about 200 lbs lighter using the modern cells (~55kwh or 150Ah), something that is capable of the discharge but smaller capacity overall could come in around 300 lbs lighter (~44kwh or 120 Ah)... If it is to go in a non-sport model (ie peak discharge is about 540A) could even go down to a 400 lbs lighter pack! (~33kwh or 90Ah) I'd think any of those options could be built and installed for less than the Tesla 3.0 price. But, if you are going to drive with deeper DOD (Depth of Discharge) cycles you would also have to expect a diminished life of the pack... Maybe its possible to get the 33kwh pack down to $15K though, and in that case it only needs to last 1/2 as long as the 3.0 pack to represent value parity... And I'd imagine 400lbs lighter would make a HUGE performance impact, that would ~15% weight reduction, suddenly the brakes would feel more than adequate and assuming enough weight is still over the rear wheels (I'd think it'll be about 50:50 distribution at that point) acceleration could improve dramatically!

Perhaps I'll look more into this potential, especially if someone is serious about doing this to their Roadster, a sort of High Performance upgrade... Additionally it would recharge in only 2hrs (of course range would be much less, but less weight would help a little with that, maybe 120-150 miles range)

 

[jeremyz]

A 400 pound lighter pack would also benefit handling by leaving the top part of the battery empty. A lower center of gravity is always nice.

 

[ElectricLove]

The LG HG2 cells (if that is what Tesla is using) are great cells. I have been doing some R&D with different cells and identified these HG2 cells as a candidate for our purposes. One of our intentions was to produce a 3rd party option for a Tesla Roadster battery pack and the cell that I chose after R&D for a few months is in fact the LG HG2 cell!

We will continue our progression but have shifted gears a bit, rather than offer a long-range pack using the LG HG2 cells we intend to offer a lighter-weight "high performance" version of the Roadster battery pack, it will have "roughly" the same range as the Roadster originally had but will recharge at a higher miles per hour rate (due to increased efficiency, same actual kw rate of energy into the pack but each kwh would yield a higher number of miles) and also provide some performance advantages. A slight shift in weight distribution (therefore stronger "steering" power), better braking capabilities (due to less weight) and increased acceleration performance!

We actually just acquired a Roadster to begin the final "stages" of actually building out this pack in a prototype and hitting the streets with it! If all goes well we should be offering an option early next year, price target is $25K and depending on what Tesla provides for warranty we will attempt to be at parity with them!

In short, the LG HG2 is an excellent choice, I am proud to say I chose it far ahead of Tesla's announcement and it is the cell I would have chosen for Tesla to use.

 

[ibcs]

The LG HG2 cells (if that is what Tesla is using) are great cells. I have been doing some R&D with different cells and identified these HG2 cells as a candidate for our purposes. One of our intentions was to produce a 3rd party option for a Tesla Roadster battery pack and the cell that I chose after R&D for a few months is in fact the LG HG2 cell!

We will continue our progression but have shifted gears a bit, rather than offer a long-range pack using the LG HG2 cells we intend to offer a lighter-weight "high performance" version of the Roadster battery pack, it will have "roughly" the same range as the Roadster originally had but will recharge at a higher miles per hour rate (due to increased efficiency, same actual kw rate of energy into the pack but each kwh would yield a higher number of miles) and also provide some performance advantages. A slight shift in weight distribution (therefore stronger "steering" power), better braking capabilities (due to less weight) and increased acceleration performance!

We actually just acquired a Roadster to begin the final "stages" of actually building out this pack in a prototype and hitting the streets with it! If all goes well we should be offering an option early next year, price target is $25K and depending on what Tesla provides for warranty we will attempt to be at parity with them!

In short, the LG HG2 is an excellent choice, I am proud to say I chose it far ahead of Tesla's announcement and it is the cell I would have chosen for Tesla to use.

Outstanding information. Thanks for sharing.