Honda Insight Forum banner

1 - 20 of 146 Posts

·
Registered
Joined
·
118 Posts
Discussion Starter #1 (Edited)
Hello everyone,

I am reconditioning my IMA battery pack. Or at least that is what I hope I am doing.

For those who are interested, I am sharing my google drive folder with the following information:

- Pictures from my IMA pack dismantling and setup for reconditioning.
- A document with a RAW collection of information from my research. There you will find all relevant information IMO about the subject. Many pieces of posts from this forum as well.
- A report of my progress which is a living document for now.
- An excel where I collect my data of my cells and modules (pairs of sticks)

Link: https://drive.google.com/drive/folders/0B4RNujDTR9UCYlJZVlRIMm16UG8?usp=sharing
 

·
Premium Member
Joined
·
3,415 Posts
Good start. Pretty sure that 27 or whatever pages could be boiled down into 2 or 3... :)

Consistency is the most important part.
 

·
Administrator
Joined
·
12,424 Posts
Not sure if you just plan to put stuff in the google folder?

That's a bit much in one lump for me to absorb.. Time factor.

Can you drip feed us in this thread as well, then it's not such a shock to the system :)
 

·
Premium Member
Joined
·
3,415 Posts
Peter,

The google sheet is where all the action/data is happening, and it's pretty much self-explanatory.

The rest is just a bunch of various pictures and a compilation document of a crap ton of things plucked from the forum and likely other sources.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #5 (Edited)
OK,

The RAW collection of information is a document where I copy pasted anything I found that could be useful. Many different opinions and crap in there indeed. But it helped me to make my choices based on my objectives.

The Report is my own journey, course of action and changes of course of action, remarks, thoughts, findings, etc. Steve is right, the most useful is the google sheet.

In short:

My initial plan was to measure IR of all individual cells. Diagnose the cells wit high IR and do as deep discharge as I can with my equipment to try to reverse the high IR. The other cells with normal IR, can be just cycled through the whole module (pair of sticks) to restore capacity. Simple, right? NOT.

I realised I can not measure the IR of cells accurately. So the new plan is:

To avoid reversals, I discharge each cell separately to 0,6V with a rate of 2A, which is right under the 0,68V mark. If I find a funny acting cell, I bring it down to 0,4V @ 0,2A, charge 2000 mAh @ 6A and back down to 0,6V.

Steve, on hindsight, I now would like to have the possibility to discharge at 6A, but my "scientific gut feeling" tells me not more. I want to be gentle on my sweet cells, they have suffered a lot. 2A will do for now. The only thing is I now practically live in my workshop with short naps around the clock to get this done in a reasonable amount of time. I am lucky I have an understanding woman.

After the individual cell's "balancing discharge" to 0,6V, I bring the whole module to 0,2V/cell @ 0,2A. I assume that there will be no reversals this way. (And my thinking is that the slow deep discharge, will keep the cells for some time in low voltages and I trust sssr2's theory about reversal of crystallization and oxidation is correct and this will provide some improvement in IR in combination with the subsequent charge at 6A after.)
Let them rest "empty".

Start again with module 1 in the same sequence up to 11:

CHG 6A to DV cut off
CHG 0,2A for x hours - to full charge
(Target input 8800 mAh total - Thanks Steve! - Calculate “trickle” charge duration based on this number, after the 6A charge)
DSC 2A to 1,1V/cell
DSC 0,2A to 0,8V/cell
CHG 6A to DV cut off

Based on what I read and on my "scientific gut feeling", this regimen will do most good. I am not sorting sticks, I don't have access to replacements. I gently bring them down low and balance while avoiding reversals and do 1 thorough cycle.

Later on I can get more advance equipment, IR measuring, Reaktor 1000W, etc. Maybe a second battery and more time, and do a more in depth work. Who knows. I am definitely enjoying this. I have good hopes.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #6
I am noticing that the sticks located in the centre of the battery pack show a poor health. The sticks located on the outside show good health. And this is also the case with the cells in the length of the sticks. The central cells show less good health than the cells on the edges.

My conclusion is that the cooling in the inside of the pack is inadequate.
The pack is relatively compact. Not much space for free air flow. I need more forced convection.

Has anybody tried installing extra fans in the battery pack compartment?
I am also thinking to change the location of the thermocouples of the pack more inside the centre of the pack, so the cooling fan of the car will be engaged earlier and more often. Only one thermocouple is relatively in a central position, but even that one is on the second cell. Cells number 3 and 4 will be surely hotter than no 2.

What do you think?
 

·
Premium Member
Joined
·
3,415 Posts
Your findings are typical.

The thermistors are located strategically in the pack to monitor different areas.

The HCH2 IMA fan is a leaf blower. I'm not kidding you. At 100% duty that thing moves a tremendous amount of air. Others have installed 120mm fans at the exit. IMHO, it's likely not effective. The key would be to increase utilization of the IMA fan. If you wanted, you could run it full blast all the time by shorting the PWM wire to ground.

A good seal between the IPU lid and the base is very important.

A good solution would be to install a temperature sensor at the IPU exhaust in the trunk and an automatic PWM controller that varied fan speed accordingly. The car has this, but in many climates, it's inadequate.

The Netherlands are fairly mild compared to AZ where we have 3-4 months worth of 38-46°C weather. In your situation, you probably don't see much heat effects.

I can no longer access your shared folder in your first post.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #8 (Edited)
Thanks Steve,

It is not the climate. It is the temperature of the cells during operation that I am worried about.
Good to know there is a good fan in there. I will check how to implement your advice on how to make it work some more. Having it on all the time is a bit overkill I think, but a better regulation and a better temperature monitoring is in order I think.

Yes, I turned off link sharing on my google folder. I will post updates in this thread for people.

Although I mean to share my experience to help others, the report and amount of data I am collecting is gradually turning into a research paper. A friend of mine who is an electrical engineer will work together with me to make this a real publication maybe.

I gave you access temporarily. What do you think about my work?

EDIT: Since I will not be writing a research paper, I restored the link.
 

·
Premium Member
Joined
·
3,415 Posts
Climate is a massive influence. AZ eats batteries. It's why I have over 100 dead HCH2 subpacks.

There is also a fundamental issue with the cooling of the "D" cell configuration. The surface area to volume ratio and high section thickness are pretty much the worst combination possible.

So far, so good. I'm digesting it. I started going through your trimmed down document concerning the process, and there's a lot I don't agree with in there concerning conclusions, theories and processes, but it looks like you're NOT doing the things I don't agree with when I look at the data you're collecting so far... :)

Consistency is key. With the exception of the data holes, you're definitely on the right track so far. I'm looking forward to the results.

Steve
 

·
Registered
Joined
·
118 Posts
Discussion Starter #10
Define "dead subpack".
What are the symptoms exactly?

Regarding cooling, When I discharge at 2A without cooling, the cell temp is at 39C. (Measurement of the internal temperature with the IMAX).
With my ventilator on (which is always on) at a distance of 2 feet, the temperature drops very fast (within a few minutes) to 31C.
With the ventilator at 3 feet, the temperature is 34C. The air flow has a huge and immediate influence.

The outside temperature of the cells is then at 20C. When the internal temperature reaches 50C, which is the limit according to the manufacturer, the outside temp (thermocouple reading) is much much lower.

I bet in the Arizona sun, it reaches much more than 50C...
 

·
Premium Member
Joined
·
3,415 Posts
"Dead" means completely unsuitable for use in a car. The vast majority of them have 1-2 cells in 1-2 sticks with excessive SD. The vast majority of them have 5000mAh+ capacity, but they won't hold a charge.

I would not expect to see 39°C at 2A assuming 25°C ambient.

The airflow has a huge and immediate influence on the surface, but it has little to no influence on the core. It merely changes the temperature distribution between the surface and the center. Additionally, flow around a cylinder is very inefficient.

That condition can be addressed by knowing the temperature distribution and regulating cooling at lower temps. There is evidence this occurs because prius packs tend to run 5-10°C hotter than IMA packs based on my experience. Prius section thickness is 1/2 that of a "D" cell with far more efficient cooling flow. It's uncommon to see a Prius pack UNDER 40°C even in moderate weather.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #12
I would not expect to see 39°C at 2A assuming 25°C ambient.
It is 18-20C ambient. Would you expect more or less than 39C internal temp?

That is what the IMAX B6 reading is. And it is consistent on all the cells. Without cooling.
With the ventilator on them, it drops to 30C. (I can even measure the air speed around the cell and make a few measurements with different air velocity. Maybe tomorrow, I will sleep at home tonight, not at the office.)

Of course the air flow must have an influence in the core temperature of the battery. It is metal. The battery is a very thermally conductive object. Isn't it?
 

·
Premium Member
Joined
·
3,415 Posts
Lower, but I didn't think it through. A 2A discharge is going to have a pretty uniform cell temp given the 2+ hours of discharging. My discharges are at 20A, and they end in 15 minutes. I never see temps over 35°C in 26°C ambient, BUT I don't wait around to see what the final temp is if I leave them sitting... plus there's a radiative/natural convective flow after the discharge terminates, which would lower the surface temp even as the distribution normalized.

Back to... consistency is key. If all your cells are responding the same, then it's probably normal... :)

The cell construction is spiral strips separated by a mesh, so I have to assume heat transfer within the cell occurs slower and is less uniform than with a solid metallic.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #14
Preliminary Conclusions

Over half way of the project.

The Imax B6 mini is a good charger, but definitely not powerful enough for this endeavour.

Capacity wise I have seen no problems. The IMA pack was definitely unbalanced.

The cells IR range from 2 to 10 mOhm. Many modules are just ok with IR's of 4-7.

The pack will probabably function well if the current demand of the car stays under 50A. The high IR cells will drop too low with higher current and risk of reversal is almost certain.
Especially if the SOC is less than 80%.

What remains to be seen is if high IR can be improved one way or another.
 

·
Registered
Joined
·
118 Posts
Success is sweet.

And... HE 'S A [email protected]%*ING ROCK STAR!

IMA pack back in the car. Drove for a while.

No recals, no IMA light, the battery charge level stays up, the pack behaves like NEW!!!!
So I took her to the car wash and pampered her!

I am sooooooooo happy!

First I will go drink 3000 euro of beer, and then I will post a more detailed summery of what I did to help others as you guys helped me!
Massive thanks to the battery veterans in here. Thank you!!
 

·
Registered
Joined
·
118 Posts
Discussion Starter #17
No, I did not do SD testing. For two reasons:

1. I don't have replacement modules (subpacks). And, SD testing would take me another week.
2. My car has been used all the time I have it. It has never stood still for long periods. I read that most packs that suffer from high SD, are packs that have been sitting a lot.

Although it would be interesting to see if reconditioning has improved the SD rate.

I took the car to the Honda dealer this morning and asked him to hook up his computer and read the IMA pack.
It was beautiful. The voltage was 184V, all the modules perfectly balanced at 16,7V, and the capacity restored to 75%. It was 35% before reconditioning.

I guess my IMA pack was gradually deteriorating and I was not noticing much difference. It was just charging more and more often while driving. Now, after reconditioning it is an astonishing difference. The pack is charged, and it can give a few minutes of assist (gentle) before one bar/block goes away on the gauge. And during regen, it just charges back that top block and is full again. I was used to see it go empty every 5 minutes and start forced charging with the engine.

The car also feels more powerful. During assist, 2-3 bars feel more powerful than 2-3 bars did before.

I plan to do another reconditioning (not as meticulous as this time) in about 3 months. After that, maybe again 6 months later. We 'll see. I will keep my IMA pack in shape now.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #18 (Edited)
Introduction

In the next few posts I will describe the steps I took.

Honda Civic Hybrid 2009 model. Great car. Never had any problem with it (knock on wood!).
The IMA battery dealership warranty is 160.000 km in Holland.
The IMA light started showing up at 180.000 km. The dealer said the pack should be changed and it's 3000 euro. A guy at the dealership told me it is not much of a problem, the car will still run fine, only the battery will charge more often. I thought.. ok, it can wait then.

Now, at 250.000 km, the IMA light was on always. It would recalibrate often and would go to empty and full every 5 minutes. I had no idea this is really bad for the battery.

I started studying. I read about grid chargers. Found a US company that can make them for about $300. But the guy was one month abroad. That didn't work. (I am now happy it didn't)
I studied some more and found information about reconditioning. I ordered 3 IMAX B6 mini's and kept studying. I read everything I could find on the internet for 3 weeks. Most useful information came from this forum.

And I started experimenting with two small AA NiMH batteries.

The RAW collection of information from my research is here:
https://drive.google.com/open?id=1q1iucYvPwNAsKXKjVziCBCLDun5IHoM0XatGeB0elck

A lot of pieces of copy/pasted text is in there. Good and bad stuff and different stuff for different objectives.

Attached a few informative pdf's

I read that I might need to replace bad modules and started looking for used old battery packs to buy one. No way. I called about 15 car junk yards, no battery pack. One of them told me that they have deals with dealers and battery packs are always sent to them. Mafia! On the other hand, those packs are hazardous waste and they are sent back to the factory. Individuals are not allowed to buy loose IMA packs in EU.
Then I found out that all hybrid cars are sold in Eastern Europe after they are used here. So, no IMA pack to be found in Holland.

My objective was to restore as much as possible my own pack. I was not sorting sticks to rebuild a new one. So I based my strategy on that.
 

Attachments

·
Registered
Joined
·
118 Posts
Discussion Starter #19 (Edited)
Step 1: Balancing discharge

I took the pack out of the car and tried to measure the internal resistance of each cell to diagnose / identify problem cells. That did't work. The IR measurements of the IMAX B6 are not accurate.

I made small cuts in the shrink wrap of the cells and inserted electrodes.

I discharged each cell at 2A to 0,6V (zenner voltage?) and wrote down the capacity.
This gave me a good overview of how unbalanced the pack was and it balanced all the cells very accurately.
On hindsight, now I would discharge at 2A until 0,9V and then slowly at 0,6A to 0,6V. Deeper than 0,9V should be slow.
Then I discharged the whole modules at 0,2A to 0,2V/cell.

During this module discharge, the cells started discharging very equally. But the last 10 minutes 1 cell reversed and the last 5 minutes about half of the cells were reversed. But not more than -0,12V.
My observation was that some of the cells were at -0,12V and others at 0,7-0,8V. So, deep discharging a module does not mean you are deep discharging all the cells.
 

·
Registered
Joined
·
118 Posts
Discussion Starter #20
Step 2: CHG and IR calculations

I read about the benefits of deep discharge for prolonged periods of time. That it is beneficial to the cells to sustain long periods of being empty.
So, between each step of the process I let the modules rest after discharges, while I treat the next module. And I hoped this would be good.

After the initial balancing discharge and module deep discharge, I charged each module using the IMAX B6 mini's Auto Charge mode.
The Auto charge mode tests the battery and regulates the current. It was starting with low current to get them out of the deep discharge state and would gradually increase current. The maximum power of the IMAX was 3,4A for a whole module (60W).
Some modules never reached that. Some eventually were charged with 3,4A and some only got 0,7A (which is a forming charge of 0,1C). That was interesting. Did those modules need that and others didn't? Was it beneficial? I don't know.

After full charge of 8800 mAh, I let them rest for about an hour and then:
1. I put a load and measure the V drop of each cell.
2. I discharged the module for 2000 mAh.
3. Put a load and measure V drop again.
4. Discharge the module further to 0,9V/cell.

I used those V drop measurements to calculate the IR of each cell at full charge and at -2000 mAh state.
That gave me a very good overview of the cells with high IR.

The discharge of a whole 12-cell module with the IMAΧ goes at a rate of 0,3A. IMAX B6 has 5W of discharge power. It was painstakingly slow. But again, I do wonder if this slow rate of discharge was beneficial to the health of the cells.

One thing was clear: the cells discharged very equally at this slow rate to 0,9V. I was occasionally checking cell voltages.

One theory is that the Charge Transfer (electrons) is a very fast process. But the Mass Transfer in the chemical reaction is a slow process. According to this theory, slow charges and discharges should be more rejuvenating to the cells. Again, I don't know.

Each module was treated and left to rest "empty".

Result at the end of this step:
1. The cells are balanced
2. All modules have had 1 cycle
3. I have a full map of the cell's IR
 
1 - 20 of 146 Posts
Top