Honda Insight Forum banner

921 - 940 of 942 Posts

·
Registered
Joined
·
237 Posts
I gave some approximate cost numbers back in post #913. It is most likely less than a new Bumblebee.
Which occurred to me as I hit "Submit Post"... I'll lamely hide behind my starting to get a cold I only recently got over as my excuse. The total cost would be higher for the SCiB cells version vs. the Prinsight pack, but considering the greater return on effort investment + lack of needing anti-swelling straps the SCiB pack could be considered overall be a better value (as long as one cares about kWh improvement not just reliability, and doesn't mind manually rebalancing the cells now and then).

I'll discuss it with the two locals whose packs need attention like mine does. If neither's interested in the SCiB conversion, I might just do myself for my Insight even though with the Volt I'm already getting my LIB propulsion battery fix.

Roger
 

·
Registered
Joined
·
135 Posts
Ive watched this thread, your youtube videos, even talked through messages. I really like the plug and play idea with your board but understand life gets in the way. Would there be any chance releasing what you do have even if its not full feature packed or releasing a bms for 72 cells. I did the lto conversion as it was really really easy. Currently have just a home made bcm fooler. Im looking to do more. My main quirk with the other fooling metjods is the soc gauge not working and being faked. I dont know why but thats a big thing to me. Im not a fan of screens and controllers everywhere.

Just a shot to see if you would even to sell pcb's and a bom or a parts kit of components with a programmed chip. Many others lebowski for example has his boards open source and just sells his chip for his bldc controller.
 

·
Linsight Designer
Joined
·
1,841 Posts
Discussion Starter #923 (Edited)
Sorry for the delay... the programming is "all" I have left to do, but that isn't happening anytime soon. At present Linsight supports up to 256 series stacks, but after 36s you'll need a daughterboard per 16s... I'd like to get 48s on the RevB PCB, but that's also not happening anytime soon. It makes sense that I'd eventually sell an additional daughterboard that fits inside the OEM BCM enclosure, which would offer another 48s.

Yes, I will sell you the RevA PCB, but it comes with about a dozen bugs you'll need to be aware of; none of them are showstoppers, but certainly they'll be fixed prior to release. Do note that Linsight is open source, but is also still pending release, so I haven't publicly made anything available; I will though, if people are interested. To be honest, I've been hoping another engineer will take the Linsight project off my hands, and have several times offered this proposal to the community.

Send me a PM and we can discuss whether you're willing to take on this ALPHA, PROTOTYPE hardware; additional firmware development is required. I don't know what I'll charge for this hardware, but I'll certainly send a PCB (with a BOM) for $25 shipped. I'll send you the files for free if you want, too. I'm certainly not guarding any IP on this project; I always make my work product open source (unless the person I'm contracting for prevents that, but I charge less for open source contracts, so that is often enticing to payroll).

Apart from me telling you what presently exists, don't expect me to do any new work on the Linsight project anytime soon... sorry, I'm presently midway through the development cycle on two new CNC machines and they're commandeering my "free time" (along with my neverending house project). Linsight is going to be a fun project to pick up again at some point in the future, but it's always been a pet product, whereas other products I've designed gross seven figures every couple years... Linsight probably won't gross $200,000 ever, so it's certainly a pet project.

To be honest, though, if you're just looking for a BMS, you can buy a complete released system for pretty cheap... Linsight's goal is true drop-in, and I haven't done enough on the firmware to share at this point.
 

·
Registered
Joined
·
135 Posts
Mudder sent pm about what you would release for beta testing. Email adress in pm. Thanks for the contributions to the forum and towards g2 insight. Appreciated.
 

·
Registered
Joined
·
39 Posts
I hate to chime in being completely incompetent with battery fundamentals as the rest of you guys, but I have found very little about, what the title of the thread says "dropping in" some Lithium Ion batteries with no other modifications at all. I am sure for good reason it hasn't been talked about much - but I can't help but think that two 18650s in series sum up to the voltage range we need to see with a factory stick. Theres these flashlight conversion kits that convert 3 d batteries into two 18650s. What if someone manufactured a converter that did 2s2p 18650s (or even 2s4p if it would fit) in the space where the six d batteries are welded in series, what communication problems would arise? If this has been addressed the search option on here is sub par - a link to the info would be great instead of a tounge lashing for not doing my due dilligence and searching first - something you knowledgeable folks love to dwell on and a big reason why amateurs stray away from forums for info. 20 converters, 40 of the 3500 mah 18650s, would certainly total out to less than any alternative - if no other mods were required......
 

·
Super Moderator
Joined
·
7,918 Posts
Well...................

Use of the 18650 has been discussed many times. Since you did not have much success with the "search community" window, I suggest you use the advanced search tool, which is much more powerful. To do that click the 3 dots in the top right and select "advanced search." I just did so and returned a larger number of threads, many related to folks discussion of the 18650. I combined the terms "18650 cells, and specified the "1st Gen Modifications and Technical Issues" sub forum, where about 90 percent of the Lithium discussion has taken place. My results are here:


Anyway, as I understand it from what I remember of prior posts, the ordinary 18630 won't handle the regeneration surges even in a 2P configuration, and it suffers from cycle life. I didn't read in detail all the hits under the above search, but I did see that the 18650 is generously discussed, and discarded by most converters.

Most of the successful conversions have focused on high capacity, high current individual cells, such as the A123 or the assembled packs of Toshiba SCiB cells from the Fit electric, I think because the electrical configuration is simple to implement and reasonable compact.

A quick take of the "drop in" option shows that no such design ever emerged for the gen1 Insight. Every conversion has been a one off design by the individual owner. Given the small number of customers, no commercial drop in conversion is likely to emerge. In a like sense, you are encouraged to make a shot at a conversion using a more robust 18650 cell if you can find an adequate cell at a reasonable price. We all learn by trying:)

edited-cleanup
 

·
Registered
Joined
·
6,626 Posts
....I am sure for good reason it hasn't been talked about much - but I can't help but think that two 18650s in series sum up to the voltage range we need to see with a factory stick.... 40 of the 3500 mah 18650s, would certainly total out to less than any alternative - if no other mods were required.
It's not impossible, but rather just impractical, given other options. I'm not totally steeped in the 18650 lore, but it's not as easy as you'd think it should be. The main obstacles are these:

-regular lithium-ion cells can burn, explode, etc., so you need a good, purpose-built battery management system.

-even so, the voltage range of 18650s isn't quite right compared to the Insight voltage range: the NiMH cells are tasked to work within a range of 1V to 1.6V, or 6V to 9.6V per 6-cell stick; 18650s have a range of something like 3.6V to 4.1 or 4.2V, so a 2-cell 'stick' would have a range of 7.2V to 8.4V at most. Doesn't quite fit, does it?...

-regular ole lithium-ion cells, 18650s, have a really imbalanced charge/discharge profile: they can usually handle quite decently high discharge rates, but charge rates are skimpy in comparison. And, overall, the cells can't handle much power, so you need to parallel more cells. The pack would need to handle 100 amp discharges and 50 amp charges. What 18650 cells out there can do this, and/or how many of what's out there would you need? I've seen something like 20 amp, maybe 30 amp discharge capability - so just based on that you'd need at least 3, maybe 4 paralleled 18650s to handle the discharge. But like I said, the discharge handling is often at least double or triple what these cells can handle for charge - so double or triple those 3 or 4 paralleled cells, and we're up to 6 or 8 cells, or even 9 or 12 - per 'stick'. Multiply that by 10 sticks and we're up to a lot of cells.
 

·
Registered
Joined
·
39 Posts
It's not impossible, but rather just impractical, given other options. I'm not totally steeped in the 18650 lore, but it's not as easy as you'd think it should be. The main obstacles are these:

-regular lithium-ion cells can burn, explode, etc., so you need a good, purpose-built battery management system.

-even so, the voltage range of 18650s isn't quite right compared to the Insight voltage range: the NiMH cells are tasked to work within a range of 1V to 1.6V, or 6V to 9.6V per 6-cell stick; 18650s have a range of something like 3.6V to 4.1 or 4.2V, so a 2-cell 'stick' would have a range of 7.2V to 8.4V at most. Doesn't quite fit, does it?...

-regular ole lithium-ion cells, 18650s, have a really imbalanced charge/discharge profile: they can usually handle quite decently high discharge rates, but charge rates are skimpy in comparison. And, overall, the cells can't handle much power, so you need to parallel more cells. The pack would need to handle 100 amp discharges and 50 amp charges. What 18650 cells out there can do this, and/or how many of what's out there would you need? I've seen something like 20 amp, maybe 30 amp discharge capability - so just based on that you'd need at least 3, maybe 4 paralleled 18650s to handle the discharge. But like I said, the discharge handling is often at least double or triple what these cells can handle for charge - so double or triple those 3 or 4 paralleled cells, and we're up to 6 or 8 cells, or even 9 or 12 - per 'stick'. Multiply that by 10 sticks and we're up to a lot of cells.
got it. wasn't quite sure of the chemistry in comparison - i certainly didn't realize the charge / discharge differential. yeah i would say that makes it quite a bit impractical. The core hch1 battery sticks i went through did top out a charge somewhere above 9v thats for sure - I didn't even think about that. I just wasn't around when it would switch from charge to discharge so I only really got to see AH numbers after the cycles. I'm just putting this thing back together and trying to figure ways to make this setup more reliable. I knew there was a reason there hasn't been any 18650 packs for the gen 1's.... this pretty much sums it up. Those LTO packs you all have built are amazing - I just need to become more financially stable with free time to commit to pulling that off. For now - reconditioning the old crap design will have to do. But man that manual control looks so cool!!!!
 

·
Registered
Joined
·
39 Posts
Well...................

Use of the 18650 has been discussed many times. Since you did not have much success with the "search community" window, I suggest you use the advanced search tool, which is much more powerful. To do that click the 3 dots in the top right and select "advanced search." I just did so and returned a larger number of threads, many related to folks discussion of the 18650. I combined the terms "18650 cells, and specified the "1st Gen Modifications and Technical Issues" sub forum, where about 90 percent of the Lithium discussion has taken place. My results are here:


Anyway, as I understand it from what I remember of prior posts, the ordinary 18630 won't handle the regeneration surges even in a 2P configuration, and it suffers from cycle life. I didn't read in detail all the hits under the above search, but I did see that the 18650 is generously discussed, and discarded by most converters.

Most of the successful conversions have focused on high capacity, high current individual cells, such as the A123 or the assembled packs of Toshiba SCiB cells from the Fit electric, I think because the electrical configuration is simple to implement and reasonable compact.

A quick take of the "drop in" option shows that no such design ever emerged for the gen1 Insight. Every conversion has been a one off design by the individual owner. Given the small number of customers, no commercial drop in conversion is likely to emerge. In a like sense, you are encouraged to make a shot at a conversion using a more robust 18650 cell if you can find an adequate cell at a reasonable price. We all learn by trying:)

edited-cleanup
yeah i did read back a bunch and figure out that nobody had done it yet but couldn't find anything about actually trying to drop in replacement cells of a different chemistry right into the existing battery and was wondering the reason. someone at some point i think it was peter thought of making a bracket to fit a battery cell of a different size into but I couldnt find it when i looked again. its alot of info to read through some of these threads are 50 pages long. I did read through that very exciting time when Peter pulled off the first hacks.... then the complete conversion happened pretty fast from there with all of you guys working together. Pretty amazing achievements that sort of clown on these factory engineers that made this design in the first place. I wish I was current with this stuff getting figured out behind the scenes because I always rebelled against factory recommendations when in the dealership world. I guess I could have searched more - been reading all of these archived threads for almost two weeks now and got frustrated I guess.
 

·
Super Moderator
Joined
·
7,918 Posts
There is certainly a lot of stuff to read. The Insight community went through about 2-4 years of trying to figure out some sort of practical conversion(with a ton of posts), with several false starts. THEN - one guy figure it out with the Fit LTO cells (insightbuyer) and the whole thing grew rapidly along the Toshiba SCiB lines. Something else will emerge later, but this is the current game. No standardization - no drop ins.
 

·
Premium Member
Southern California
Joined
·
784 Posts
Another thing with trying to drop in lithium cells is that you NEED to have a balancing system. Unlike NiMH where we can use a grid charger to "overcharge" cells until all cells are topped up and balanced (the extra energy is just dissipated as heat), overcharging li-ion batteries will cause them to catch fire and explode. You'd need a balance wire going to every single one of those cells so you can monitor the voltage and balance them as needed. Not a simple system to engineer!

A drop-in battery, in the context of this thread, doesn't mean making no modifications or keeping the stock circuitry. The idea is to make a conversion kit that can be dropped in without modification to the rest of the car. So take out your IMA battery and BCM/MCM, and bolt in this new system, plug it in, and it will all work like stock. That's the idea, at least. Actually making it is of course difficult, expensive, and time-consuming.
 

·
Registered
Joined
·
95 Posts
Has anyone looked at using these li-ion bags to build a pack? My quick calc is that ~45 pouches would crank up to right voltage range. 8 amp hour rating looks workable. Of course, there is the challenge of a BMS charging system and possibility of explosion and fire inside the small cabin space, but you know... progress requires some sacrifices.

 

·
Linsight Designer
Joined
·
1,841 Posts
Discussion Starter #935 (Edited)
Yeah, the state of lithium in G1 insights circa 2019 is certainly bespoke. When I started this thread SEVERAL years back, my goal was to sell a product that replaced both OEM hybrid computers (BCM+ MCM). My replacement product - Linsight - used the same connectors, so no wire harness cutting was required. The customer would then use whatever lithium chemistry they wanted, knowing that Linsight would take care of the battery changing/discharging maths. It was a super promising project...

...but then I. got. busy. Like really busy. Like "my startup raised two million dollars in a week" busy. Nine months later I sold/gave away nearly everything I owned, moved into a van, and spent the next 18 months decompressing... during this time I began working on Linsight again in earnest, but also on other more lucrative work. I also decided to build my own house from the ground up, which took way longer than I ever anticipated (I'm still not technically finished with it). All the while I kept designing other projects, and I kept finding myself busy. Then, back in May I was thinking "wow, my schedule looks pretty open, maybe now's the time..." but just a few days later I picked up another project, which we just announced a few weeks back... which raised another couple million dollars, so now I'll likely spend another six months fulfilling all those orders. <I get that this paragraph is far from humble; this will surprise nobody that knows me>.

And even though I work on all these super fun projects, the one I yearn to come back to is Linsight. There's a reason I drive this particular $3000 car (on a good day). I've always known Linsight would never yield any substantial profit, and yet it's the one project I want to finish. But I'm just not sure when that will happen. Worst case: I'll be Jime's age before Linsight ships. But ship it will, at some point.

Following Eric Raymond's sage advice ("when you're done with a project find a new owner"): If anyone wants to take over the project, please be my guest. It will take proficiency in low-level microcontroller programming... Linsight uses an ATmega64M1 for the BLDC sine drive, and a much larger Arm processor for everything else. I have hardware... I've verified each subsystem works... but the firmware just isn't far enough along for the general "plug-and-play" user. If you can prove to me that you know how to program in C, I will set aside as much time as you need to get up to speed and take over this project. Fair warning: the scope of this project is far more advanced than "I know how to program an Arduino." Super respect to all those that have taken that baby step, but you'll need to know how to manage a realtime operating system (I'm using FreeRTOS).

And that's the state of Linsight circa 2019Q4: forever on hold, at least for the foreseeable future. I believe Mario and I have a $0 wager on who ships their product first; my money is on Mario ;). However, he, too, is wrapped up in the tech world.
 

·
Registered
Joined
·
95 Posts
I love the insights. I will have to wait on someone else to program my first linsight, I don't have the programming skills described above.
 

·
Registered
Joined
·
10 Posts
... but the firmware just isn't far enough along for the general "plug-and-play" user. If you can prove to me that you know how to program in C, I will set aside as much time as you need to get up to speed and take over this project. Fair warning: the scope of this project is far more advanced than "I know how to program an Arduino." Super respect to all those that have taken that baby step, but you'll need to know how to manage a realtime operating system (I'm using FreeRTOS).
Any consideration on open sourcing it?
 

·
Premium Member
Southern California
Joined
·
784 Posts
I believe Mario and I have a $0 wager on who ships their product first; my money is on Mario ;). However, he, too, is wrapped up in the tech world.
I didn't realize we had such a high-stakes bet going on! I'd better make sure I win, then. :)
Yeah, I've been busy. I've also been distracted with side projects. Like Linsight, Pegasus will be finished eventually, but when that is exactly is uncertain.
 
921 - 940 of 942 Posts
Top