DaveinOlyWA
Well-known member
Marktm said:
I have concerns as well. It "appears" they are stacking cell pouches with no air gap whatsoever. Notice the one view where it appears there are a dozen or more pouches stacked up against each other?
Battery Upgrades are very possible
- Thread starter Inventor71
- Start date
Help Support My Nissan Leaf Forum:
I have concerns as well. It "appears" they are stacking cell pouches with no air gap whatsoever. Notice the one view where it appears there are a dozen or more pouches stacked up against each other?
It is actually common practice to pack pouch type cells tightly together. This makes it more difficult for them to swell under unfavorable conditions. Once a pouch swells it is both useless and dangerous - you can't reverse the process. It's counter-intuitive, like using ice to keep something warmer in a really cold environment, but it apparently works. Converting a Vectrix maxi-scooter like mine involves building a horizontal 'stack' of Leaf modules that are held tightly together much like the pack in the video: with long threaded rods, nuts and washers. My conversion (not done by me but by the bike's original owner) is about 5 years old now, with <knocks on wood> no problems so far.
A
Anonymous
Guest
LeftieBiker said:
It's really important, when stacking pouch cells, for there to be even pressure across the whole cell surface and a physical barrier between the cells to contain any out-of-control thermal reactions. Without even pressure, any internal gas pressure from evolved hydrogen within the cell can strain the pouch and cause puffing or short-circuits across the separators. The Vectrix works fine because the Leaf cells already have that addressed at the module level: the steel case contains and constrains the cells with even pressure while air gaps prevent excess heat from flowing through the stack rather than out of it. As long as you b0lt the cells together and compress them with the OEM compression plates, nothing too bad can happen.
EVBatteryRebuilds has assembled a pack with no inter-cell barriers, no air gaps, no emergency vents, no cell-level fusing, no cooling, and no apparatus for applying even pressure to the cells. Not only that, but they compromised the structural integrity of the battery case (by removing the central crossbar) just so they could pack more cells in. It's cool and all that it's 64 kWh, but this an incredibly bad battery design, perhaps the worst ever documented.
http://www.formula-hybrid.org/wp-content/uploads/A123_AMP20_battery_Design_guide.pdf (p. 33-41)
https://www.researchgate.net/publication/323084845_Mechanical_Design_and_Packaging_of_Battery_Packs_for_Electric_Vehicles/link/5b5ed4e80f7e9bc79a6e9b34/download
https://researchbank.swinburne.edu.au/file/b51b5e0a-692f-4bc0-a4a1-1806c8b21642/1/Shashank%20Arora%20Thesis.pdf
Preface: I have a competing product so I'm going to always look like I'm promoting my own company by passing judgement on competitors. The following is the opinion of Emile Nijssen, not of MUXSAN as a whole, and off the cuff/badly edited/probably wrong.
I basically share the opinions of the other previous posts on their battery wholeheartedly. This is a very, VERY badly engineered product and I hope this is just a demonstrator. As a demo pack, just for getting some PR buzz, it's totally fine and I will happily concede that I've done similar rush jobs in the past just to show a proof of concept. But this is not a safe EV battery system - this is a heap of cells.
You're going to get incredibly poor cycle life without compression frames, without interstitials and without an adequate BMS. Additionally, they don't have thermal management. The Leaf's batteries aren't bad chemically, the cells are perfectly competent NMC cells capable of excellent service life. Their life is cut drastically short by poor BMS algorithms and absent thermal management. These LG cells are not great cells to begin with, especially for EV purposes, so they're going to need a much better BMS to get decent performance out of them in the long term.
Additionally, they're simply not charging enough money for this. At MUXSAN, we're aware our prices aren't the lowest, but that is part of our sustainable business model. We're in it for the long haul, charging a bit extra so we can support our products. If I go by BOM costs, RdS/EVBatteryRebuilds is operating on 20%ish margins at best, or maybe even barely breaking even because they rely on selling the old cells for cashflow. This inevitably means that they'll go under in 2 years, leaving a few dozen customers with unsupported, low production volume and thus inherently buggy hardware.
Lastly, we're quite worried that these kinds of stunts are going to eventually cause either the shutdown of this cottage industry or heavy regulation with little opportunity to adapt (effectively regulating us out of business). We're already working hard to essentially become a bona fide car parts manufacturer with all the certifications and compliance work necessary, but that's going to take 2ish years to complete (if you include a complete supply chain audit etc.).
I don't want to end this on too much of a downer. They're doing good work in principle and definitely learning from the experience. EVBatteryRebuilds is not a lost cause or something that can never work. The principle is... fine. Their mission is great. This is not a company that has to be stopped or that deserves hate in any form, but they need a couple of decent engineers and maybe somebody with some business sense to help them.
I basically share the opinions of the other previous posts on their battery wholeheartedly. This is a very, VERY badly engineered product and I hope this is just a demonstrator. As a demo pack, just for getting some PR buzz, it's totally fine and I will happily concede that I've done similar rush jobs in the past just to show a proof of concept. But this is not a safe EV battery system - this is a heap of cells.
You're going to get incredibly poor cycle life without compression frames, without interstitials and without an adequate BMS. Additionally, they don't have thermal management. The Leaf's batteries aren't bad chemically, the cells are perfectly competent NMC cells capable of excellent service life. Their life is cut drastically short by poor BMS algorithms and absent thermal management. These LG cells are not great cells to begin with, especially for EV purposes, so they're going to need a much better BMS to get decent performance out of them in the long term.
Additionally, they're simply not charging enough money for this. At MUXSAN, we're aware our prices aren't the lowest, but that is part of our sustainable business model. We're in it for the long haul, charging a bit extra so we can support our products. If I go by BOM costs, RdS/EVBatteryRebuilds is operating on 20%ish margins at best, or maybe even barely breaking even because they rely on selling the old cells for cashflow. This inevitably means that they'll go under in 2 years, leaving a few dozen customers with unsupported, low production volume and thus inherently buggy hardware.
Lastly, we're quite worried that these kinds of stunts are going to eventually cause either the shutdown of this cottage industry or heavy regulation with little opportunity to adapt (effectively regulating us out of business). We're already working hard to essentially become a bona fide car parts manufacturer with all the certifications and compliance work necessary, but that's going to take 2ish years to complete (if you include a complete supply chain audit etc.).
I don't want to end this on too much of a downer. They're doing good work in principle and definitely learning from the experience. EVBatteryRebuilds is not a lost cause or something that can never work. The principle is... fine. Their mission is great. This is not a company that has to be stopped or that deserves hate in any form, but they need a couple of decent engineers and maybe somebody with some business sense to help them.
knightmb
Well-known member
mux said:
I think that last statement is important, from what I see, a lot of time is put into the electronics part of the build (battery setup, wiring, electronics, etc.) but many of these don't have someone who is knowledgeable in battery chemistry or physics to help. I think too many people want to treat a EV like a flashlight where you simply change out the batteries and everything else will just work out for the best. :roll:
coleafrado said:
Ahhh, that isn't good. I didn't watch the video closely enough to see they had ditched the modules and were just installing pouches in a more or less exposed fashion. You are correct, of course.
coleafrado said:
Great references - so much to learn about Li battery pack designs. The recommendation that FEA be done on any commercial AMP20 (pouch) battery pack design probably suggests the need for extraordinarily carefully constructed and tested final products. OTOH, I wonder if the cylindrical designs get around the dimensional stability issues?
A
Anonymous
Guest
Now that they've de-unlisted their video, it's reached over 100,000 views and 150 comments. I'm seeing about 100-300 views per hour. :|
A few recent comments from EVBR:
The 51º they mentioned must be centigrade, i.e. 128 degrees Fahrenheit, in which case they are literally playing with fire. Even a first-gen 24 kWh pack doesn't get anywhere near that hot unless it's fast charged 3 or 4 times in a row within a span of a couple hours. Who even knows where they're measuring that temperature? If that's the external shell temperature, the center of the pack could easily have passed 80ºC. Ordinary cells that spend lots of time sitting near 50ºC usually don't last long: https://sci-hub.tw/https://www.cell.com/joule/fulltext/S2542-4351(19)30481-7
There's a reason we haven't seen a thermally managed aftermarket pack from anyone yet. Designing this stuff, iterating on it with FEA, building and blowing up prototypes and then mass-producing it takes years even with "unlimited" resources and dozens of engineers. As far as I know, zero of the companies working on Leaf batteries have more than ten employees let alone ten engineers.
Cylindrical cells are better, yes - their steel cases eliminate the need for constant external compression and protect the actual cells from physical damage - but they come with their own issues. Cooling is more complicated since it usually requires access to the curved sides of the cells rather than flat faces like pouches. Tesla runs pre-formed extruded aluminum microchannel between every other row of cells, then plumbs them all together. The tooling to produce that plumbing is not cheap. Electrical assembly is harder, too, since cylindrical cells require specialized 3D busbars and wirebonding at two points per cell. Not something you can make with a drill press and some copper stock.
It's not impossible, of course. The Model 3 proved that you can do cylindrical cells cheaply if you're willing to sacrifice repairability. But the setup costs for automated assembly are in the ~tens of millions unless you're willing to spend 100-200 hours per car in manual labor costs putting together just the battery. There's a reason Tesla has their module assembly line right next to their cell production line. Even at sub-$80/kWh cell costs, Tesla barely breaks even on their cars and Panasonic is basically making 0%.
A few recent comments from EVBR:
The 51º they mentioned must be centigrade, i.e. 128 degrees Fahrenheit, in which case they are literally playing with fire. Even a first-gen 24 kWh pack doesn't get anywhere near that hot unless it's fast charged 3 or 4 times in a row within a span of a couple hours. Who even knows where they're measuring that temperature? If that's the external shell temperature, the center of the pack could easily have passed 80ºC. Ordinary cells that spend lots of time sitting near 50ºC usually don't last long: https://sci-hub.tw/https://www.cell.com/joule/fulltext/S2542-4351(19)30481-7
Marktm said:
There's a reason we haven't seen a thermally managed aftermarket pack from anyone yet. Designing this stuff, iterating on it with FEA, building and blowing up prototypes and then mass-producing it takes years even with "unlimited" resources and dozens of engineers. As far as I know, zero of the companies working on Leaf batteries have more than ten employees let alone ten engineers.
Cylindrical cells are better, yes - their steel cases eliminate the need for constant external compression and protect the actual cells from physical damage - but they come with their own issues. Cooling is more complicated since it usually requires access to the curved sides of the cells rather than flat faces like pouches. Tesla runs pre-formed extruded aluminum microchannel between every other row of cells, then plumbs them all together. The tooling to produce that plumbing is not cheap. Electrical assembly is harder, too, since cylindrical cells require specialized 3D busbars and wirebonding at two points per cell. Not something you can make with a drill press and some copper stock.
It's not impossible, of course. The Model 3 proved that you can do cylindrical cells cheaply if you're willing to sacrifice repairability. But the setup costs for automated assembly are in the ~tens of millions unless you're willing to spend 100-200 hours per car in manual labor costs putting together just the battery. There's a reason Tesla has their module assembly line right next to their cell production line. Even at sub-$80/kWh cell costs, Tesla barely breaks even on their cars and Panasonic is basically making 0%.
Hello friends!
I am Jesus Toucedo, owner of Renovables del Sur, and also of evbatteryrebuilds.com, but now I write as one more user of the forum that has been mentioned.
I have read the last posts about the battery update and it seems that our video has caused quite a stir.
I want to start by saying that I agree on quite a few things discussed here, but there are others that I would like to qualify and answer for allusions.
First of all, I want to thank Emile for his contribution with the can bus bridge PCB, if none of this would have been possible or would have been a little more difficult than it has already been. We started testing with Raspberry and its solution has been by far much more elegant, professional and automotive grade.
Second, I want to confirm to Emile that indeed, this battery has only been a prototype, a proof of concept, and that in all probability it will never make a battery like this again. As a business I assure you that it is not profitable as Emile says, and we have only lost 90 days of our time, but you know that, it has been great, someone had to do it. Everything we have done in this battery has been agreed with its owner, who is aware of its limitations and virtues.
Having said this, I want to mention some commented points that I think are important to clarify:
Regarding the stacking of cells, I have access to the LG Chem technical manual on these cells, and the only mention it makes about the stacking is that a tightening with M6 screws and nuts greater than 10kgf x cm must be maintained. I can assure you that the cells have the correct compression, regardless of the number of them stacked.
Regarding the vibrations, well indeed it is a homemade pack, and it could never be a professional one, but we have done many tests and the car has been driven for 4000 km before being delivered. We have subjected it to all possible dynamic tests, acceleration, braking, inertia, potholes, and heavy off road, and after all the tests we have reopened the pack. As expected, absolutely nothing has moved, and has behaved in a stable way. Only time will tell how it goes. We can open this thread in 5 years and take some photos of the inside of the battery if it has not yet exploded
Regarding the space between cells, who said there should be? We all know that the Nissan leaf pack is a disaster dissipating heat, we also know that it is airtight and there is no air circulation inside it, nor is it possible to install a cooling system due to the design. Seriously, what else do you think I can do?
Regarding the temperature, someone asked where the probes have been placed, the probes are obviously between the LG cells. In no test have we gone from 51º celcius in Chademo fast charge, even with temperature days at 35º.
I don't know where you live, but in the south of Spain, 51º is reached by the battery in the first and only fast charge.
The current pack heats up equal to or less than the original, please give its owner some time to report his experience, I am sure that soon he will have a lot to say.
I suppose you have your opinion and of course I have mine, but I want to comment on these points:
"EVBatteryRebuilds has assembled a pack with no inter-cell barriers, no air gaps, no emergency vents, no cell-level fusing, no cooling, and no apparatus for applying even pressure to the cells."
LG cells already have their own emergency ventilation system, what else can I do? Put them in a can for even less refrigeration? I don't see the point.
Air Gaps? I have not read anything about that in the LG assembly manual .. in fact in its professional assemblies for other vehicles I have never seen air gaps between the cells.
Cell Level fusing? No refrigeration? Does Nissan use it? I didn't do the engineering for this battery pack! It would have been nice to put it .. I'm sure it would.
"no apparatus for applying even pressure to the cells." What do you mean? They have 22 pressure points on each side, calibrated to deliver the necessary pressure to each cell.
As I said before, I hope it is the last battery of this type that we make, but I want you to know that we are proud of it, although of course it is not a professional system, but on the other hand, it is impossible.
We have our way and we are going to follow it. We think that putting a lot of cells in the trunk is not a good idea for the safety of the occupants, or to keep the center of gravity of the car, and other problems related to insurance and authorities, but we are glad that at least their Users who for sure are aware of their limitations can once again enjoy their Leafs. I firmly believe that this is the true reason that unites us, to see all the Leafs of the planet on the road again. In the meantime, we will try to learn and improve with each step we take.
I am Jesus Toucedo, owner of Renovables del Sur, and also of evbatteryrebuilds.com, but now I write as one more user of the forum that has been mentioned.
I have read the last posts about the battery update and it seems that our video has caused quite a stir.
I want to start by saying that I agree on quite a few things discussed here, but there are others that I would like to qualify and answer for allusions.
First of all, I want to thank Emile for his contribution with the can bus bridge PCB, if none of this would have been possible or would have been a little more difficult than it has already been. We started testing with Raspberry and its solution has been by far much more elegant, professional and automotive grade.
Second, I want to confirm to Emile that indeed, this battery has only been a prototype, a proof of concept, and that in all probability it will never make a battery like this again. As a business I assure you that it is not profitable as Emile says, and we have only lost 90 days of our time, but you know that, it has been great, someone had to do it. Everything we have done in this battery has been agreed with its owner, who is aware of its limitations and virtues.
Having said this, I want to mention some commented points that I think are important to clarify:
Regarding the stacking of cells, I have access to the LG Chem technical manual on these cells, and the only mention it makes about the stacking is that a tightening with M6 screws and nuts greater than 10kgf x cm must be maintained. I can assure you that the cells have the correct compression, regardless of the number of them stacked.
Regarding the vibrations, well indeed it is a homemade pack, and it could never be a professional one, but we have done many tests and the car has been driven for 4000 km before being delivered. We have subjected it to all possible dynamic tests, acceleration, braking, inertia, potholes, and heavy off road, and after all the tests we have reopened the pack. As expected, absolutely nothing has moved, and has behaved in a stable way. Only time will tell how it goes. We can open this thread in 5 years and take some photos of the inside of the battery if it has not yet exploded
Regarding the space between cells, who said there should be? We all know that the Nissan leaf pack is a disaster dissipating heat, we also know that it is airtight and there is no air circulation inside it, nor is it possible to install a cooling system due to the design. Seriously, what else do you think I can do?
Regarding the temperature, someone asked where the probes have been placed, the probes are obviously between the LG cells. In no test have we gone from 51º celcius in Chademo fast charge, even with temperature days at 35º.
I don't know where you live, but in the south of Spain, 51º is reached by the battery in the first and only fast charge.
The current pack heats up equal to or less than the original, please give its owner some time to report his experience, I am sure that soon he will have a lot to say.
I suppose you have your opinion and of course I have mine, but I want to comment on these points:
"EVBatteryRebuilds has assembled a pack with no inter-cell barriers, no air gaps, no emergency vents, no cell-level fusing, no cooling, and no apparatus for applying even pressure to the cells."
LG cells already have their own emergency ventilation system, what else can I do? Put them in a can for even less refrigeration? I don't see the point.
Air Gaps? I have not read anything about that in the LG assembly manual .. in fact in its professional assemblies for other vehicles I have never seen air gaps between the cells.
Cell Level fusing? No refrigeration? Does Nissan use it? I didn't do the engineering for this battery pack! It would have been nice to put it .. I'm sure it would.
"no apparatus for applying even pressure to the cells." What do you mean? They have 22 pressure points on each side, calibrated to deliver the necessary pressure to each cell.
As I said before, I hope it is the last battery of this type that we make, but I want you to know that we are proud of it, although of course it is not a professional system, but on the other hand, it is impossible.
We have our way and we are going to follow it. We think that putting a lot of cells in the trunk is not a good idea for the safety of the occupants, or to keep the center of gravity of the car, and other problems related to insurance and authorities, but we are glad that at least their Users who for sure are aware of their limitations can once again enjoy their Leafs. I firmly believe that this is the true reason that unites us, to see all the Leafs of the planet on the road again. In the meantime, we will try to learn and improve with each step we take.
A
Anonymous
Guest
Jesus, thank you for posting to the forum!
This is great to hear, thank you for the clarity.
The concern with the cells is that those specifications are for a single cell stack - i.e. five or ten cells on top of each other, with bolts all around. It's pretty easy to create an even gradient of pressure when the bolts are only 3-5 inches apart, the stack height does not matter too much. But with the cells next to each other, and no space for bolts between the stacks, the only way you can apply pressure is from the edges of the modules (30cm minimum separation). It is quite visible in the video that the plates bend upwards towards the center of the stacks. Unless you can find an infinitely rigid top plate, or create an asymmetric spacer pad, the possibility of pouch puffing or other mechanical failure in the stacks is higher than it should be. I hope this is clear enough.
It is explained elsewhere, but the amount of movement necessary to cause mechanical damage to a pack this large is so small that it's not really noticeable by eye. I have seen pouch cells suddenly explode from relatively minor (and ultimately invisible) asymmetry in their compression.
The emergency vents on the cells are intended to be used in conjunction with pack vents - i.e. burst discs or plastic breakable vents that can release any built-up gases in a safe direction. As you said, the battery case is air-tight; if one cell activates its emergency vent, the rest may follow, and it would be bad news if the battery case itself were to burst.
The Chevy Bolt uses either the same or very similar cells to what you're using, but they're implemented very differently. Each Bolt stack is made up of several modules, each module being several cells welded together and enclosed in a steel or aluminum case with vents and liquid cooling plates. Not only does the steel shell on each module protect the cells, it slows any possible spread of fire or heat.
It's the difference between a seawall made of sandbags and a seawall made of sand. Compartmentalization is very important for large batteries; the criticism of your prototype is that there are only three compartments (left, right, rear) in the whole thing. The lack of cooling and mechanical rigidity just compounds that.
see here: https://www.youtube.com/watch?v=54SshSAH_k0
Nissan doesn't use cell-level fusing or refrigeration because they're idiots. Why repeat their mistakes? Essentially everyone else uses cell-level fusing and cooling.
I can assure you that Nissan and GM made, at one point, prototype batteries that were no less dangerous! The crucial part is that they didn't sell them until they were certain they wouldn't injure anyone. Thermal runaway on a cell phone battery (Note 7, anyone?) is bad enough. Multiply that by a factor of a few hundred and it should make sense why we would discourage this kind of pack design.
ramdoor said:
This is great to hear, thank you for the clarity.
The concern with the cells is that those specifications are for a single cell stack - i.e. five or ten cells on top of each other, with bolts all around. It's pretty easy to create an even gradient of pressure when the bolts are only 3-5 inches apart, the stack height does not matter too much. But with the cells next to each other, and no space for bolts between the stacks, the only way you can apply pressure is from the edges of the modules (30cm minimum separation). It is quite visible in the video that the plates bend upwards towards the center of the stacks. Unless you can find an infinitely rigid top plate, or create an asymmetric spacer pad, the possibility of pouch puffing or other mechanical failure in the stacks is higher than it should be. I hope this is clear enough.
Regarding the vibrations, well indeed it is a homemade pack, and it could never be a professional one, but we have done many tests and the car has been driven for 4000 km before being delivered. We have subjected it to all possible dynamic tests, acceleration, braking, inertia, potholes, and heavy off road, and after all the tests we have reopened the pack. As expected, absolutely nothing has moved, and has behaved in a stable way. Only time will tell how it goes. We can open this thread in 5 years and take some photos of the inside of the battery if it has not yet exploded
It is explained elsewhere, but the amount of movement necessary to cause mechanical damage to a pack this large is so small that it's not really noticeable by eye. I have seen pouch cells suddenly explode from relatively minor (and ultimately invisible) asymmetry in their compression.
The emergency vents on the cells are intended to be used in conjunction with pack vents - i.e. burst discs or plastic breakable vents that can release any built-up gases in a safe direction. As you said, the battery case is air-tight; if one cell activates its emergency vent, the rest may follow, and it would be bad news if the battery case itself were to burst.
The Chevy Bolt uses either the same or very similar cells to what you're using, but they're implemented very differently. Each Bolt stack is made up of several modules, each module being several cells welded together and enclosed in a steel or aluminum case with vents and liquid cooling plates. Not only does the steel shell on each module protect the cells, it slows any possible spread of fire or heat.
It's the difference between a seawall made of sandbags and a seawall made of sand. Compartmentalization is very important for large batteries; the criticism of your prototype is that there are only three compartments (left, right, rear) in the whole thing. The lack of cooling and mechanical rigidity just compounds that.
see here: https://www.youtube.com/watch?v=54SshSAH_k0
Nissan doesn't use cell-level fusing or refrigeration because they're idiots. Why repeat their mistakes? Essentially everyone else uses cell-level fusing and cooling.
I can assure you that Nissan and GM made, at one point, prototype batteries that were no less dangerous! The crucial part is that they didn't sell them until they were certain they wouldn't injure anyone. Thermal runaway on a cell phone battery (Note 7, anyone?) is bad enough. Multiply that by a factor of a few hundred and it should make sense why we would discourage this kind of pack design.
Completely agree with the idea of an even pressure to the cells. In fact we mounted an "apparatus for applying even pressure to the cells" like you can see in this picture. It may not be the most suitable material, but it does its job. It is a 15mm wooden board, with a 1.5mm steel plate on top, I can ensure they distribute the pressure very well and evenly throughout the pack as you can see in the attached photo.
https://photos.app.goo.gl/8r1snSN3wgQHjzND9
On top of this comes the cover of the pack, so I think those cells cannot be expanded anywhere. On the sides, there is no room for the table, nor for intermediate screws, but the number of cells stacked is much less, so we opted for 22 points of lateral pressure. I hope they endure.
Regarding refrigeration, of course I would love to have been able to include it, but to date I believe that no one has succeeded.
Again, I accept and appreciate your comments and note for future improvements.
Regards
https://photos.app.goo.gl/8r1snSN3wgQHjzND9
On top of this comes the cover of the pack, so I think those cells cannot be expanded anywhere. On the sides, there is no room for the table, nor for intermediate screws, but the number of cells stacked is much less, so we opted for 22 points of lateral pressure. I hope they endure.
Regarding refrigeration, of course I would love to have been able to include it, but to date I believe that no one has succeeded.
Again, I accept and appreciate your comments and note for future improvements.
Regards
Hi:
I think it's a good thing that options like this appear, since Nissan doesn't give a **** to customers and only wants to sell cars. Yes, there are some safety issues, but i think Jesus has answered to all questions and since we have a "beta" tester, we will get more information over time.
I to am considering change the battery of my 30kWh Leaf when the time comes (still some time to reach the end of warranty - currently with 3.5 years and 107000km) and since i'm from Portugal, the company from Jesus is my best option.
Besides, i think that when the upgrade is made to the 42kWh battery, they still use the old Leaf cells aluminium pack (can you confirm this, Jesus?).
So by the time i will made the upgrade, there will be probably cells with more energy density that can be used inside the cells pack and give a bigger energy pack
I think it's a good thing that options like this appear, since Nissan doesn't give a **** to customers and only wants to sell cars. Yes, there are some safety issues, but i think Jesus has answered to all questions and since we have a "beta" tester, we will get more information over time.
I to am considering change the battery of my 30kWh Leaf when the time comes (still some time to reach the end of warranty - currently with 3.5 years and 107000km) and since i'm from Portugal, the company from Jesus is my best option.
Besides, i think that when the upgrade is made to the 42kWh battery, they still use the old Leaf cells aluminium pack (can you confirm this, Jesus?).
So by the time i will made the upgrade, there will be probably cells with more energy density that can be used inside the cells pack and give a bigger energy pack
Jesus, great to have you here. Great to hear you're using the open source branch of our CAN bridge, make sure you keep an eye on the repo as we're going to port our stable driver/setup updates there.
Now, you're coming off a as a little bit defensive. I completely sympathize, I'm the same way. When I've made something, I'm going to defend the design choices even if I'm deep down not sure they were actually the best. However, please take these criticisms constructively and learn from them. Yes, we're essentially saying your design choices betray that you don't know enough about battery design to really be doing and selling this. That does't mean we're saying you are dumb or stupid, at all. Look at my first videos back in early 2018, with bare batteries in the back of my car, no connected BMS, etc.. - it's all quite clearly unsafe and unfit for anything but a proof of concept. We are just worried that you're going to cause problems - initially for yourself, but consequently also for other companies like mine. Or to say it differently: your problems are our problems as well. So it's in all of our interests to fix the issues with your design before commercial deployment.
coleafrado already made some comments, mine partially overlap but I'll try to expand on them a bit as well.
OK, this is not a good start. I know it is an unsatisfying way to answer this but: the experience of every systems engineer and battery expert says there should be a thermal management system of some kind. You KNOW, you even say so, that there is a big heat dissipation and thermal management issue with Nissan batteries, it is probably the #1 concern that even laypeople have about the battery. You're rebuilding the pack, almost from scratch. You have the opportunity to fix this! This should in fact be the #1 issue to fix in a rebuilt pack. And we've done the math on this quite a while ago - it is possible even with fairly limited space usage to improve the thermals of this pack so much. Yes, you will have to sacrifice a few kWh of batteries to do this - but keep in mind: you're going to lose those kWh very quickly due to the rapid degradation that this pack will certainly undergo.
If you're looking for a way to do this: buy an e-NV200 battery pack and see how they did it there. That's the way to go. It shouldn't surprise anyone that those packs degrade at about 1/3rd the rate of Leaf packs.
The issue with this statement is that you have no knowledge of the cell internals. There are only 3 or 4 sensors in Leaf battery packs, so you can only test... 3 or 4 cell surface temperatures. How do you know that these temperatures properly reflect extremes in the pack? Do you know the amount of heat generated in the cells under all use cases and at all temperatures, also in the degraded state of the cells? For instance, a cell at 90% SOH will have roughly 1.5x the internal resistance of a cell at 100% SOH, so it will generate 1.5x the heat - do you reprogram the BMS to properly adjust the charging speed? Do you know the thermal properties of the cell and thus do you know how hot the hot spots inside a cell get?
Similarly, temperature isn't everything; temperature differential also plays a big role. In fact, that's most of what thermal management does - evening out the temperature of a pack, especially during QC sessions.
And don't forget, even if you DON'T include any active thermal management, you still need to do thermal modeling to make sure you don't develop hot spots and to properly limit charging. Even if you're not overheating the battery, you're still going to have to consider expansion rate, electrolyte gas pressure and current density anisotropy within the cells. This is one of the big reasons why the Leaf NCM cells charge so slowly at high SOC - they're high capacity NCM chemistry, so the charge tends to bunch up in the center of the cell instead of diffusing evenly. This sets up an internal chemical potential that can wreck your cell in no time. And likewise, that's why we use very high performance cells - they use much thicker electrodes and much more permeable electrolytes to even out this charge distribution inside the cell, allowing for much higher charging currents.
This is just the beginning, just the basic theory. Actually implementing it, testing and getting feedback from the field takes months. If you need to do cell testing yourself: another month extra. There's a lot to this!
51C is too high for a fast charging session, that is dangeously close to peak electrolyte temperature (typically around 55C). At those temperatures, your electrolyte starts to build up significant gas pressure, both in response to being charged and the high temperature. In an improperly compressed stack, this may lead to delamination of your cells and fire.
The emergency venting does not prevent fire or thermal runaway. All it does is prevent an explosion or deflagration. And once a cell vents, it is nonregenerative - you lose the cell and in this case the entire pack.
Like a fuse, cell venting is a very last resort measure and not a replacement for other passive and active safety systems.
There's two things that may be confused here: large air gaps for compartmentalization and small air gaps for compression handling.
Large air gaps help to establish cooling and physically isolate one module/cell from another in case of failure. This is exceedingly effective at limiting a battery fire to just one part of the pack, especially if you remove oxygen from the enclosure quickly.
Small air gaps are used to manage compression properly. A fully discharged cell is slightly smaller than a fully charged cell. Pouch cells are generally sealed at the edges, so they're mechanically constrained there - but they can bulge up in the middle. That bulging, especially at higher temperatures and charge rates (as discussed earlier) has to be limited, and you do that by applying equal pressure to the entire cell, from edge to center. Cells are - generally - designed to be compressed in this way. If you have a small number of cells, the total amount of strain on the compression stack is small and it's possible to use just a cheap stamped/bent compression plate to hold everything in place. But with larger numbers of cells, if the cells are compressed tightly when discharged, they will expand so much that the compression on all the cells is unequal at high SOC. There are multiple solutions for this, but e.g. Nissan uses thin steel shells around each cell that are bent inwards towards the center of the cell. Then on the ends of each stack of cells, they use a compression frame that only compresses the OUTSIDE RIM of the cell. Together, this compression strategy enables even compression across a wide range of SOC, even with cells of different health.
Another way to do this, is by leaving air gaps strategically at low SOC, eg. by putting a thin shim of metal between all cells.
You're only compressing the stack with a thin sheet of metal and threaded rods around it. That provides pressure in the wrong place - it concentrates pressure on the outside rim of cells (which are already constrained) and leaves the centers free to expand.
Here's a schematic idea of what's happening with Nissan's compression strategy vs. yours:
Don't say this in public about your own product! You're not making this (just) for fun, you're a company selling products. You have a responsibility towards your customers to deliver a reasonably well-made product. You absolutely, 100% for sure, need to be able to guarantee some level of professional responsibility in the design, marketing and support of your products. People are going to expect this from you. Again, I've been in the same boat as you, I made a prototype all the way back in March 2018 and customers were streaming in by July. It took until the end of 2019 before we actually had something we could call a well-designed, finished product. I've made it clear all along the way that we're not done yet, this is just a prototype, this is just a beta test, etc. etc.
Our first customers have been absolute angels about the stuff we put them through. It's so great to have people willing to help, but this well of easy-going, fault-tolerant people is going to dry up quickly. At some point, even small, cosmetic issues are going to cause you trouble. You WILL have to be working towards a product that is professional, compliant and dependable.
Please hire a lawyer and insurance specialist! What you're saying here is quite the reverse of reality. In order to sell a remanufactured battery, Nissan needs to homologate your battery. That means you have to convince Nissan to accept your battery as a legal part of the car. Nissan has been quite clear over the past years that they're not homologating anything for the Leaf - not even simple things like tow bars or wheel sizes, let alone remanufactured batteries. Even then, if they do accept to go this way, you will need to pay them quite a significant amount of engineering fees and you will likely need to become a member of SAE to become a battery manufacturer for the car. It is quite a complicated and winding road towards properly selling remanufactured batteries for the Leaf. Ask Blue Cars in New Zealand about their experiences in this regard.
And this is stuff you have to figure out and have really clear before you start trying to sell a product. Especially across country lines.
Now, you're coming off a as a little bit defensive. I completely sympathize, I'm the same way. When I've made something, I'm going to defend the design choices even if I'm deep down not sure they were actually the best. However, please take these criticisms constructively and learn from them. Yes, we're essentially saying your design choices betray that you don't know enough about battery design to really be doing and selling this. That does't mean we're saying you are dumb or stupid, at all. Look at my first videos back in early 2018, with bare batteries in the back of my car, no connected BMS, etc.. - it's all quite clearly unsafe and unfit for anything but a proof of concept. We are just worried that you're going to cause problems - initially for yourself, but consequently also for other companies like mine. Or to say it differently: your problems are our problems as well. So it's in all of our interests to fix the issues with your design before commercial deployment.
coleafrado already made some comments, mine partially overlap but I'll try to expand on them a bit as well.
ramdoor said:
OK, this is not a good start. I know it is an unsatisfying way to answer this but: the experience of every systems engineer and battery expert says there should be a thermal management system of some kind. You KNOW, you even say so, that there is a big heat dissipation and thermal management issue with Nissan batteries, it is probably the #1 concern that even laypeople have about the battery. You're rebuilding the pack, almost from scratch. You have the opportunity to fix this! This should in fact be the #1 issue to fix in a rebuilt pack. And we've done the math on this quite a while ago - it is possible even with fairly limited space usage to improve the thermals of this pack so much. Yes, you will have to sacrifice a few kWh of batteries to do this - but keep in mind: you're going to lose those kWh very quickly due to the rapid degradation that this pack will certainly undergo.
If you're looking for a way to do this: buy an e-NV200 battery pack and see how they did it there. That's the way to go. It shouldn't surprise anyone that those packs degrade at about 1/3rd the rate of Leaf packs.
The issue with this statement is that you have no knowledge of the cell internals. There are only 3 or 4 sensors in Leaf battery packs, so you can only test... 3 or 4 cell surface temperatures. How do you know that these temperatures properly reflect extremes in the pack? Do you know the amount of heat generated in the cells under all use cases and at all temperatures, also in the degraded state of the cells? For instance, a cell at 90% SOH will have roughly 1.5x the internal resistance of a cell at 100% SOH, so it will generate 1.5x the heat - do you reprogram the BMS to properly adjust the charging speed? Do you know the thermal properties of the cell and thus do you know how hot the hot spots inside a cell get?
Similarly, temperature isn't everything; temperature differential also plays a big role. In fact, that's most of what thermal management does - evening out the temperature of a pack, especially during QC sessions.
And don't forget, even if you DON'T include any active thermal management, you still need to do thermal modeling to make sure you don't develop hot spots and to properly limit charging. Even if you're not overheating the battery, you're still going to have to consider expansion rate, electrolyte gas pressure and current density anisotropy within the cells. This is one of the big reasons why the Leaf NCM cells charge so slowly at high SOC - they're high capacity NCM chemistry, so the charge tends to bunch up in the center of the cell instead of diffusing evenly. This sets up an internal chemical potential that can wreck your cell in no time. And likewise, that's why we use very high performance cells - they use much thicker electrodes and much more permeable electrolytes to even out this charge distribution inside the cell, allowing for much higher charging currents.
This is just the beginning, just the basic theory. Actually implementing it, testing and getting feedback from the field takes months. If you need to do cell testing yourself: another month extra. There's a lot to this!
51C is too high for a fast charging session, that is dangeously close to peak electrolyte temperature (typically around 55C). At those temperatures, your electrolyte starts to build up significant gas pressure, both in response to being charged and the high temperature. In an improperly compressed stack, this may lead to delamination of your cells and fire.
The emergency venting does not prevent fire or thermal runaway. All it does is prevent an explosion or deflagration. And once a cell vents, it is nonregenerative - you lose the cell and in this case the entire pack.
Like a fuse, cell venting is a very last resort measure and not a replacement for other passive and active safety systems.
There's two things that may be confused here: large air gaps for compartmentalization and small air gaps for compression handling.
Large air gaps help to establish cooling and physically isolate one module/cell from another in case of failure. This is exceedingly effective at limiting a battery fire to just one part of the pack, especially if you remove oxygen from the enclosure quickly.
Small air gaps are used to manage compression properly. A fully discharged cell is slightly smaller than a fully charged cell. Pouch cells are generally sealed at the edges, so they're mechanically constrained there - but they can bulge up in the middle. That bulging, especially at higher temperatures and charge rates (as discussed earlier) has to be limited, and you do that by applying equal pressure to the entire cell, from edge to center. Cells are - generally - designed to be compressed in this way. If you have a small number of cells, the total amount of strain on the compression stack is small and it's possible to use just a cheap stamped/bent compression plate to hold everything in place. But with larger numbers of cells, if the cells are compressed tightly when discharged, they will expand so much that the compression on all the cells is unequal at high SOC. There are multiple solutions for this, but e.g. Nissan uses thin steel shells around each cell that are bent inwards towards the center of the cell. Then on the ends of each stack of cells, they use a compression frame that only compresses the OUTSIDE RIM of the cell. Together, this compression strategy enables even compression across a wide range of SOC, even with cells of different health.
Another way to do this, is by leaving air gaps strategically at low SOC, eg. by putting a thin shim of metal between all cells.
You're only compressing the stack with a thin sheet of metal and threaded rods around it. That provides pressure in the wrong place - it concentrates pressure on the outside rim of cells (which are already constrained) and leaves the centers free to expand.
Here's a schematic idea of what's happening with Nissan's compression strategy vs. yours:
Don't say this in public about your own product! You're not making this (just) for fun, you're a company selling products. You have a responsibility towards your customers to deliver a reasonably well-made product. You absolutely, 100% for sure, need to be able to guarantee some level of professional responsibility in the design, marketing and support of your products. People are going to expect this from you. Again, I've been in the same boat as you, I made a prototype all the way back in March 2018 and customers were streaming in by July. It took until the end of 2019 before we actually had something we could call a well-designed, finished product. I've made it clear all along the way that we're not done yet, this is just a prototype, this is just a beta test, etc. etc.
Our first customers have been absolute angels about the stuff we put them through. It's so great to have people willing to help, but this well of easy-going, fault-tolerant people is going to dry up quickly. At some point, even small, cosmetic issues are going to cause you trouble. You WILL have to be working towards a product that is professional, compliant and dependable.
Please hire a lawyer and insurance specialist! What you're saying here is quite the reverse of reality. In order to sell a remanufactured battery, Nissan needs to homologate your battery. That means you have to convince Nissan to accept your battery as a legal part of the car. Nissan has been quite clear over the past years that they're not homologating anything for the Leaf - not even simple things like tow bars or wheel sizes, let alone remanufactured batteries. Even then, if they do accept to go this way, you will need to pay them quite a significant amount of engineering fees and you will likely need to become a member of SAE to become a battery manufacturer for the car. It is quite a complicated and winding road towards properly selling remanufactured batteries for the Leaf. Ask Blue Cars in New Zealand about their experiences in this regard.
And this is stuff you have to figure out and have really clear before you start trying to sell a product. Especially across country lines.
DougWantsALeaf
Well-known member
Mux
Very excited you are a part o this board. As you have seen the inside of our Leaf's much more than we have, I have a quick question. For those of us with the newer battery, we have seen that the BMS runs a kind of quarterly (its pretty much 90 days to the day) re-calibration. It seems to run first 6 months post build date, and then every 90 days after that. Most of us have seen that in the first year to 18 months the Battery is brought down 1-2.5% each quarter until it gets to the low 90's (% SoH) then flattened out. Between the quarterly cycles, you may see a .01 or .02 change on any given day or in a week, but just tiny moves.
Do you have any insight as to what this is? Our speculation is that Nissan is cleverly (or not so cleverly) bringing down usable capacity after the first 6 months to create a bigger buffer on the battery for longer battery longevity. This also allows them to show a longer EPA range new, than would be achievable other wise. I don't know what you can see when interfacing the leaf at a code level, but appreciate any comments.
Thank you! Danke!
Very excited you are a part o this board. As you have seen the inside of our Leaf's much more than we have, I have a quick question. For those of us with the newer battery, we have seen that the BMS runs a kind of quarterly (its pretty much 90 days to the day) re-calibration. It seems to run first 6 months post build date, and then every 90 days after that. Most of us have seen that in the first year to 18 months the Battery is brought down 1-2.5% each quarter until it gets to the low 90's (% SoH) then flattened out. Between the quarterly cycles, you may see a .01 or .02 change on any given day or in a week, but just tiny moves.
Do you have any insight as to what this is? Our speculation is that Nissan is cleverly (or not so cleverly) bringing down usable capacity after the first 6 months to create a bigger buffer on the battery for longer battery longevity. This also allows them to show a longer EPA range new, than would be achievable other wise. I don't know what you can see when interfacing the leaf at a code level, but appreciate any comments.
Thank you! Danke!
jlv
Well-known member
Nissan is a car company. Thus, they want to sell cars.jfr2006 said:
OTOH, a company that sells a replacement battery with a safety issues seems to be the one that doesn't care about it's customers.
jlv said:
Bad publicity is very hard to fight and an unhappy customer brings bad publicity. This brings down sales, and Nissan should think about that.
Just look at all the bad publicity they are getting from the recent news about them asking for a new battery the same value as a new car!
DougWantsALeaf
Well-known member
I believe we have now seen at least 2 Kona fires, and a few Tesla ones.
I have yet to year about a Leaf Pack blowing. Not saying the 0 thermal mgmt is a good idea. I wish it had it, just saying Nissan Leaf's have been extremely safe vehicles. On par with SUVs.
https://www.torquenews.com/1083/volkswagen-golf-and-nissan-leaf-defy-safety-trends-much-safer-other-small-cars
I have yet to year about a Leaf Pack blowing. Not saying the 0 thermal mgmt is a good idea. I wish it had it, just saying Nissan Leaf's have been extremely safe vehicles. On par with SUVs.
https://www.torquenews.com/1083/volkswagen-golf-and-nissan-leaf-defy-safety-trends-much-safer-other-small-cars
A
Anonymous
Guest
DougWantsALeaf said:
Correct: Nissan neglected everything but safety. The OEM Leaf pack doesn't have cooling or enough capacity for most users, nor does it protect itself from degradation, nor does it have particularly performant or efficient charging algorithms, but it is safe because it addresses the relevant design risks mentioned earlier in the thread. It's also been crash tested. A modified pack that doesn't address those factors is an entirely different animal.
DougWantsALeaf
Well-known member
Agreed. The 62 Pack I would argue meets most people's needs that don't include those who do frequent cross country trips. My personal experience is that up to 500 miles is very easily achievable if you time your charging stops to meals, and long breaks.
DougWantsALeaf said:
As far as I know nobody has completely reverse engineered this BMS code, and neither have we. We are working off of (secret, but shared with us through friends) engineering datasheets and specifications, we are specifically NOT trying to reverse the BMS code itself. This too has some insurance and liability implications.
We have been capacity testing and just capacity estimating, and honestly the degradation measures match up pretty well with actual degradation as far as we can tell. But then again, we only have experience with 60 or so batteries, and for only a handful we have in-depth data. We're not the source to ask for this kind of info.
By the way, SOH is not necessarily a measurement of capacity. In battery engineering, SOH is generally measured from a combination of SOC, temperature and internal resistance. As a very general rule of thumb: SOH = 80% when internal resistance at 25C is about 2x as high as a new battery. SOH scales with roughly IR^2. So you don't generally see exactly 20% capacity loss when the battery internally reports 80% SOH. And we use this to estimate SOH during driving.
Similar threads
Latest posts
-
-
Central console start-up view - is it configurable?
- Latest: nealchampion
-
-
-
-
-
-
The Definitive "CLICKING NOISE FROM FRONT OR REAR AXLE DURING TAKE-OFF/ACCELERATION" list
- Latest: KeepTheCoalInTheHole