Lithium cell swap with stock BMS? - Insight Central: Honda Insight Forum
 
Go Back   Insight Central: Honda Insight Forum > 1st Generation Honda Insight Forum > Honda Insight Forum 1st-Gen Discussion

Insightcentral.net is the premier Honda Insight Forum on the internet. Registered Users do not see the above ads.

» Auto Insurance
» Featured Product
Wheel & Tire Center

Reply
 
LinkBack Thread Tools
Old 03-11-2017, 02:38 PM   #1 (permalink)
Member
 
Join Date: Mar 2017
Location: Wisconsin
Posts: 39
Default Lithium cell swap with stock BMS?

As of yesterday I am the proud owner of a Manual Transmission 2001 silver insight! I am an electrical engineering student so I find hybrids/EVs fascinating and have been a long time lurker and love the hacks/mods done to this car. The long term goal is to eventually convert it to full EV via something like a leaf motor swap, but that won't happen for a while (When I am no longer a student and have a garage to work in).

The car has 102k miles on it and is in pretty descent shape, but the IMA battery is nonfunctional. Its currently disconnected and the car drives fine that way, but I would like to fix it if possible. I have researched the grid charger route, but that is not a long term solution, and the info from the previous owner was that the IMA system hasn't functioned in well over 2 years, so I don't know if it will even recover. Even if it does, I will probably have to replace a couple cells and it won't be long until more of them go bad so I don't want to go down that route. I could also continue to run it as it is. This wouldn't be a bad option as my current commute is mostly highway driving, but the lack of acceleration at higher speeds isn't ideal for merging, and there are some bigger hills where I would like to have the extra power so I don't have to slow down below the speed limit. I could also buy a Bumblebee battery, but I don't think I can justify spending the $2000 and am looking for a cheaper alternative.

Therefore I am currently looking at doing a lithium conversion. Due to the number of hybrids on the road today I think that using a battery from a junked car could be a cheap way to replace a failed battery. On car-part.com I can find used 2012+ Hyundai Sonata or Kia Optima (which use the same batteries) hybrid batteries for $300 and 2013 C-max hybrid (I'm looking at the hybrid, not the energi) batteries for $350, all less than 3 hours away. I am not sure if the optima/sonata batteries use the same cells as the C-max batteries, but they seem very similar. They are Lipo chemistry running at ~270V total pack Voltage.

The Hyundai batteries are discussed here: insightcentral.net/forums/modifications-technical-issues/27114-sonata-hybrid-lithium-polymer-battery-2.html

On the second page IamIan lists the specs:

Quote:
Highlights:
Brand New 5.3Ah Full Cycle ( not just usable window ) (pg1)
Down to 4.54Ah after ~160,000 mile tests (pg1)
Max Regen pulse ~10 Seconds ~30kw (pg3) or up to ~100 Amps (Pg9).. would be around ~14kw max for Insight size.
Max Discharge pulse ~10 Seconds ~46kw (pg3) or up to ~160 Amp (Pg9).. would be around ~20kw max for Insight Size.
Nominal Cell Voltage 3.75 (pgA-1)
Min Cell Voltage 2.4V (pgA-1)
Max Cell Voltage 4.3v (pgA-1)

Which means:
#1>They should be able to take the Gen1 Amp Current Rates ... for Charge or Discharge.

#2> Full Cycle they actually have less total Ah than a OEM 6.5Ah cell... a 39 cell pack with a ~146.25 nominal voltage similar to stock 144V would actually contain less total energy ~775wh full cycle vs NiMH ~936wh full cycle ... then both of them take a hit for usable capacity smaller window ... a 20-80 window for both brings this Li pack down to ~465wh usable compared to the ~576wh usable for a brand new OEM NIMH.

#3> Including a similar enclosure ... 95 lbs for 1.4kwh or ~53 Lbs for the Gen-1 Insight pack.
This indicates that they would be able to handle the load and regen currents fine. My question is: would there be a problem with swapping 40 of these in for the stock sticks and maintaining the original BMS and voltage taps? Most of the threads I found on doing this focus on increasing the battery capacity to make it into some sort of PHEV which uses a custom BMS for the lithium cells along with some sort of fooler to control the stock BMS. Although an upgraded IMA battery would be nice, this add lots of complications and what I really need is just a replacement.

I have read that the stock pack is 6.5 Ah, but also seen people claim that that only 4.5 Ah of that is usable (any clarification would be appreciated). As the replacement cells are 5.3 Ah, this may be a slight upgrade, but it also may be a slight downgrade. (Even though the replacement pack is 1.3 kWh, I would only use 40 cells - 2s to match each of the 20 stock sticks so about 0.76 kWh total) Either way though, it's comparable to a stock pack and would give people who are willing to put in the work a cheap way to restore stock IMA performance to their Insight.

Of the threads I have read, the closest to doing this seems to be Peter Perkins on this lithium conversion that failed since the batteries he was using could not handle the regen. He kept the stock BMS and simply fooled it so that the voltage taps seemed balanced. He then installed his own BMS to measure the voltage balancing. It seems that he did this since the cell voltages of the ones he was using were quite different than the cell voltages of the stock cells. This also seems to be the case for the a123 cell conversions people have done.

However, it looks like the cells from the packs I am referencing would fit into the voltage range that the stock BMS keeps the stock cells at. From the threads about the grid charger/balancer, it seem that the total pack voltage of the stock pack ranges from 120-170 volts. Based on the info above, the Lipo cells range from 2.4 to 4.3 volts so a pack 40 of these Lipo cells would range from 96 to 172 volts, which is a larger range than the stock pack, so there wouldn't be any danger of over-charging or draining them too low. This larger voltage range will decrease the capacity that the stock bms would use, but I don't think it would be too much depending on the cells linearity (correct me if I'm wrong). Another option would be to buy 2 packs and run 2 cells in parallel which should give plenty of capacity. It should also fit in the space where the stock pack goes, carving up the stock pack so that the new cells can be wired in shouldn't be too hard.

Is there something I am missing? Any suggestions? Unless you all think I am going to blow up my nice new car I am thinking about starting this pretty soon.
midnightsnacker41 is offline   Reply With Quote Quick reply to this message
Sponsored Links
Advertisement
 
Old 03-11-2017, 03:12 PM   #2 (permalink)
Senior Member
 
Join Date: Dec 2016
Location: El Paso, Tx
Posts: 543
Default

Look up the forum member called mudder
Sam12345 is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 03:34 PM   #3 (permalink)
Member
 
Join Date: Mar 2017
Location: Wisconsin
Posts: 39
Default

I've seen his work on the LInsight drop in conversion, and while its cool, he is going for a PHEV, which is much more than I am looking for. If I am adding that large of a battery I might as well go all electric (which I plan on doing at some point, but not for a couple years at least)

Last edited by midnightsnacker41; 03-11-2017 at 03:44 PM.
midnightsnacker41 is offline   Reply With Quote Quick reply to this message
 
Old 03-11-2017, 05:33 PM   #4 (permalink)
Banned
 
Join Date: May 2014
Location: near Canberra, Australia
Posts: 1,283
Default

1) I encourage you in your pursuit - it's something that should have been done years ago, and it sounds like you have the background and skills to do it.

2) AFAIK, only Peter Perkins has done anything like you're considering, and IMHO only he will have directly relevant advice backed up by experience (as opposed to reams of theory).

3) Again IMHO, you will need to be very much on guard to sort the wheat from the chaff in the advice that you will receive here. Be especially careful of those who are locked into paradigms of what they have done is best (because that's what they have done). For example, I quite agree with you regarding your comments on grid charging, Bumblebee batteries and replacing of individual bad cells - but in making those comments, you are already going heavily against the 'prevailing wisdom' of this group.

4) Good luck! A project like you describe will always be about 5-10 times the work that you expect, but that goes with territory of breaking new ground.
JulianEdgar is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 05:39 PM   #5 (permalink)
eq1
Senior Member
 
eq1's Avatar
 
Join Date: Jan 2012
Location: PNW
Posts: 4,598
Default

I like your goals, but it's a lot harder than it would seem, after you dig in a bit. Also, Linsight will be able to accommodate any sized pack, though it will make the most sense to get a big one...

First you should pick up an inexpensive power supply to use as a grid charger (~$30) and try to charge your pack that's been sitting for 2+ years. You might be surprised that it comes back to life just fine, it happens, and not atypically, either... Cheap, easy, and potentially effective enough.

On the Hyundai pack and specs, the main problem is the voltage range the car uses and the way it's divided and managed by voltage taps. For example, the max voltage the car allows on charge/regen is 192V. If the Hyundai cells have a max voltage of 4.3V each, then a 40 cell pack could be pushed over the top during regen (40 X 4.3=172V). It'd be fine on the low end, about 120V or 12V per tap. OK, so maybe 46 cells will work?: 46 X 4.3V=197.8V. The problem here is, how do you divide 46 cells evenly among the 10 voltage taps? Also, with a nominal voltage now at 46 X 3.7V=170.2V, you're undoubtedly gonna run into problems with the car's charging routine/s...

In any event, you should dig around a little more and keep the dream alive. There might be a few of Peter P.'s products, BCM Fooler, BCM Interceptor, 'Boost Box', etc. that might allow you to work around some of the mismatches between OEM and one of these lithium packs... For example, the 'boost box' allows you to adjust the OEM commanded assist and regen, within the OEM envelope - so you could dial down the amount of regen if you find that it's pushing your experimental pack voltage too high. And the fooler and/or the interceptor I think might allow you to avoid the BCM's OEM-prying eyes into your pack. You'd have to read-up on those devices. Long term, one problem is you won't have the ability to easily monitor cell voltages - and with such a volatile chemistry, you're sort of asking for trouble...

Oh, BTW, no one replaces bad cells - they're welded together and hard to get apart. Also, "Lipo" is not a chemistry...
__________________
2000MT, CAN, ~168K miles

Last edited by eq1; 03-11-2017 at 05:44 PM.
eq1 is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 05:47 PM   #6 (permalink)
Lifetime Member
 
S Keith's Avatar
 
Join Date: Oct 2014
Location: Phoenix, AZ area
Posts: 2,172
Default

Quote:
Originally Posted by midnightsnacker41 View Post
My question is: would there be a problem with swapping 40 of these in for the stock sticks and maintaining the original BMS and voltage taps?

Is there something I am missing? Any suggestions? Unless you all think I am going to blow up my nice new car I am thinking about starting this pretty soon.
Yes.

Yes.

You will need a layer of management to spoof the current BCM, and you would need a layer of management to ensure that the Lithium pack stays balanced and within range on a per cell basis.

NiMH is extremely forgiving in terms of battery management. It's over charge tolerant and fairly over discharge tolerant. It's filled with highly caustic (strong basis), non-explosive, non-flammable KOH. In extreme cases, it can generate both Hydrogen and Oxygen and create a very small explosion, but it's more loud and scary than dangerous. Risk of fire from NiMH chemistry is extremely low.

Lithium is outright explosive in comparison. It is not at all over charge or over discharge tolerant. Take it outside the normal operating voltage range, and shit can go south pretty quick.

You pretty much need battery management at the cell level with Lithium, OR you handle it externally with some form of balance charging.

The Insight also counts current, so you would need to spoof that as well; otherwise, it's going to factor that into the SoC calculation and fail to use all available capacity.

It has clearly been done before, but it is very much NOT "drop-in".
__________________
'05 G1 CVT w/192K mi as of 7/2016; '02 G1 CVT w/182K mi as of 7/2016 (P1449 HR pack); '06 HCH2 #1 w/145K mi as of 7/2016 (SOLD!), '06 HCH2 #2 w/223k mi as of 2/2016 (SOLD!); '03 Prius w/166K mi - Pack rebuilt with Gen2 modules; '08 Prius w/135k mi as of 7/2016; '07 Prius w/185k mi as of 7/2016

To view links or images in signatures your post count must be 2 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 2 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 2 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 2 or greater. You currently have 0 posts.


To view links or images in signatures your post count must be 2 or greater. You currently have 0 posts.
S Keith is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 06:21 PM   #7 (permalink)
Member
 
Join Date: Mar 2017
Location: Wisconsin
Posts: 39
Default

Thank you all for your Insight and encouragement

I guess the 192V regen is what I was missing. I was unable to find a good source for this so I made a guess based on what I saw in the grid charger threads. That means that I need to either find some way to limit the regen or design a custom BMS with BMS fooler.
midnightsnacker41 is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 06:27 PM   #8 (permalink)
Senior Member
 
Join Date: Dec 2016
Location: Bay Area, CA
Posts: 730
Default

I was also once an EE student and appreciate the enthusiasm of new discovery. I must throw in a word of caution here, using your own words about the potential to "blow up my nice new car." The dangers of fire with lithium chemistries is very real. Watch this documentary
if you haven't already to see how much energy these batteries contain and several ways to set them on fire.

Your assumptions about the ease of swap and electrical requirements ring a little naive to me, especially regarding the assumption you simply need to wire some lithium cells in the right voltage range, drop them in, wipe your hands, and be on your way. It's entirely possible you're beyond most mortals, and what take others years to learn will take you weeks to ramp up to speed and implement successfully.

Since you've asked the question though, I will assume you are still a mere mortal and suggest you start with smaller steps, reading up on lithium chemistries and caveats, then building some smaller projects first so you experience some hands on mistakes with minor misadventures.

In the meantime, I think your assumptions about the hopelessness of your pack may have jumped the gun. Compared to the project you're proposing to tackle, there's some outright trivial work you can do to suss out the condition of your pack. I've recently had experience with two packs that sound very similar to your's which are operating surprisingly strong, probably at 85% of original capacity. I had to do very minimal work for those two packs, one just a 12V reset to have the vehicle online the IMA battery, and the other a 3.5 hour grid charge to bring the pack voltage high enough to online the IMA battery. After initial revival the packs were probably around 70% level of health, but over subsequent weeks of driving become stronger to their current 85% level of health. If you're up for trying this trivial revival, here's the simplest way you can do it.

-Measure pack voltage. If it's below ~100V, grid charge is almost guaranteed to be required.
-If it's above 100V, try reconnecting the harnesses and do a 12V reset. Then start the car and see if it starts charging the pack.

That may be all you need to discover you have a surprisingly strong pack. If that still doesn't cut it, then you need to go build a grid charger, charge it for 3-4 hours, then start the car to let it charge and gauge performance from there. It's up to you what level of maintenance cycles to perform or not perform on this sack of cells you're apparently willing to throw away already without even a minimal test. I have found no maintenance necessary whatsoever in the 2-3 months of ownership for those 2 packs that turned out to be good after extended self discharge.

With a decently working NiMH pack, you can now take your time to research/create your lithium pack and see how much more performance you can manage to push out of this marvelous little car.
__________________
2004 HCH1 180k, 2001 I1 208k, tooting around bay area, CA
esprout is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 07:56 PM   #9 (permalink)
Member
 
Join Date: Mar 2017
Location: Wisconsin
Posts: 39
Default

esprout, thanks for your reply, that was a very interesting documentary - definitely hope solid state batteries prove scalable soon. Of course anything that stores energy has the potential to release it in an uncontrolled and dangerous manner. however, pretty much every manufacturer of hybrids/Ev's is switching to lithium ion, and it hasn't turned into a PR nightmare, so it is possible to use them safely and its just a matter of figuring out what parameters need to be controlled and the limits of those parameters.

The cells I am looking at can handle the load/regen and using the stock thermal management will take care of temperature. I will need my own BMS to handle voltage, and I guess that is my biggest question. Is simply controlling the voltage enough, or is there more that I need to look at? The stock BMS uses coulomb counting to determine state of charge, but is there something more complicated going on that I am not seeing?

Its great that your two packs have been running for 2-3 months, but I wonder how long they will keep that up. As the time from manufacture date increases, so does the likelihood of failure, and with Honda no longer making the G1 insight (or any vehicle where the sticks can be directly swapped), I don't see this as a long term solution. I think I will try and revive them as you suggest, but with only minimal effort and money.

I am currently looking into Peter Perkins BCM interceptor and BMS designs
midnightsnacker41 is offline   Reply With Quote Quick reply to this message
Old 03-11-2017, 10:32 PM   #10 (permalink)
Senior Member
 
Join Date: Dec 2016
Location: Bay Area, CA
Posts: 730
Default

Are you from a climate that gets really hot like AZ or TX? The cells have a higher propensity to suffer irreversible damage when exposed to high heat. Otherwise these NiMH cells seem to run for quite a while with some basic maintenance, and I'd wager that most packs getting replaced by dealers can last many more years with the type of maintenance documented to death on this site. I'd handwavingly guesstimate the two packs I've worked with will continue to provide useful assist for 3-5 more years with occasional maintenance. Unlike lithium chemistries, NiMH actually doesn't degrade appreciably with long term storage, and quite the contrary, they seem to revive themselves through self deep discharge.

New cells continue to be manufactured, and by the time they are no longer profitable to manufacture, I'm sure someone like you or mudder will have figured out a superior solution using lithium or even better storage mediums. I hope you appreciate the recursive forward looking statement in the previous sentence...hehe.

I haven't really looked into the nitty gritty of what it takes to create a lithium pack for the Insight, and have just browsed through a couple of threads out of curiosity. I'd agree with your astute hypothesis you'll inherit much of the built-in protections designed into the donor vehicle's batteries. My recollection is that state of charge can be reliably determined from operating voltage with lithium batteries, and the complication actually comes from translating a relatively well understood state of charge in real time to a form that the Insight's proprietary firmware will understand. People have come up with creative ways to do this translation, and you'll likely need to invent your own, involving a lot of reverse engineering and creative license. The only ways to accelerate this learning curve is if Peter and/or mudder are willing to share their designs and give you a brain dump of what they know. But they are both extremely busy people simultaneously juggling multiple projects, so the leg work to get up to speed and create your MVP lithium pack will most likely fall on your own shoulders. Ergo my recommendation to get it running first on NiMH, which will take you dozens of hours, rather than at least hundreds of hours, and more likely thousands of hours to get to a working lithium pack.
__________________
2004 HCH1 180k, 2001 I1 208k, tooting around bay area, CA
esprout is offline   Reply With Quote Quick reply to this message
Sponsored Links
Advertisement
 
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Insight Central: Honda Insight Forum forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:
Insurance
Please select your insurance company (Optional)

Log-in

Random Question

Thread Tools

Posting Rules
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On


Powered by vBadvanced CMPS v3.2.2

All times are GMT -4. The time now is 07:49 AM.



Powered by vBulletin® Copyright ©2000 - 2017, vBulletin Solutions, Inc.
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
 

Content Relevant URLs by vBSEO 3.3.2