Honda Insight Forum banner

1 - 20 of 151 Posts

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #1 (Edited)
The visit to Ron's battery repair shop got me thinking that here is a perfect oppertunity to see how subpacks are failing and if there is something that can be done to further restore them.
Ron sent me home with a couple of dozen bad subpacks to play with.
I needed a way to make non destructive high current connections to each cell, so after looking through my options, I cut a piece of 1.5" hard copper tube in half, then into 1" lengths.
A 1" steel rod,a vice, and a hammer, and I soon had some really strong copper spring clips. I soldered some heavy copper braid to each, and started by fully discharging each cell.
Since I can work a cell at a time, this full discharge has no danger of reversing a cell as the other series cells are not part of the discharge circuit.
I found that the subpack voltage rose from zero to 1.2V/cell in less than a minute, and that on the first subpack one cell was behaving much differently that the other 5. In 2 minutes, I recorded the voltages. You can see that one cell was at 1.22V and the others were at 1.39V.
I suspect that the one cell is ok and the rest are toast?
I have it on a CC charge and will see where it settles out.
Another project for my ever expanding list.
Photos:http://www.99mpg.com/Projectcars/mimapackwhack/whatiswrongwiththe/subpack_tester.jpg,0
Low capacity cells, shorted cells, high internal resistance cells, other issues? It will be interesting to finally see what is failing.
 

·
Premium Member
Joined
·
2,399 Posts
I've played with bad subpacks too and found that manipulating the individual cells such as cycling them seperately didn't help out with the limited amount that I've played with them. The cells with high internal resistance and low capacity will both make more heat on both a charge and discharge and I don't know of any way to reverse high internal resistance in any dry cell cylindrical battery, I think they are dead to the world at high current rates forever at that point. The low capacity cells will reverse on a discharge if you don't stop it soon enough since it is the first empty and that same low capacity cell will get hotter than the rest of the cells in the stick when at the end of charging and is likely getting a fair bit of overcharge, possibly damaging levels of it in the process. Cycling them seperately didn't help when I tried it as they seemed like they were probably the most cycled in the pack in the first place and probably got killed in the process by being driven down and charged more, essentially getting worked harder than the rest of the pack. It would be really nice if some of yours come back and since you have more sticks than I do and probably better equipment, I can only charge at 150 watts(or 7 amps if voltage is too low for 150 watts) with a rated max of 27 cells, enough to series charge 4 sticks at a rate that gets them quite warm and I blew up a computer power supply in the process, now at one that seems to be ok with 80 watts or less though and it only discharges at about 25 watts (3-4 amps during discharging for a stick depending on its state of charge). So it takes some time and I can't test too extensively, especially with the discharging. Charging sticks in series is out of the question unless its well known what their capacity is and they match fairly well, so I haven't had much time to play with them but it is quite fun to see how they perform. I've got a stick where I drove the stick to 4 volts and not one cell dropped below .6 volts but that was only at a 2 amp draw, trying it with a 5 amp draw and I found cells that would drop to .6 volts when there was another above 1 volt.

A handy gadget to use is an infrared thermometer, I found the temperature sensor on my charger doesn't help unless I can put it on the hottest cell and the temperature varies widely at anything over 0.5C on most of these, even if everything else seems pretty equal with them. I had a similar idea to yours, mine was to use automotive hose clamps and slip a wire in before tightening them down, but before that idea I've sat there holding the two wires when charging cells seperate of the sticks, I just did it at my max 7 amps for a few hundred mAh and then discharged the sticks trying to make it so they had similar voltage at discharge but that same cell was the one to end up at 50C and making quiet short durations of a high pitched whistling sound, yet I couldn't see it venting and I figured it would need to be hotter to do so in the first place.

Keep us filled in with what you find. These are fun to play with. :)
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #3
I am starting off gently with a 350MA CC charge, so I can leave it on indefinitely.
So far this morning, I am up to 8.45V for the subpack, and not even warm to the touch. The room is at 43F, so that is taking care of any heat generated.
The ability to discharge at the cell level eliminates any possibility of cell reversing.
Time will tell if any of these are salvageable.
 

·
Premium Member
Joined
·
2,399 Posts
Just an FYI, At 3 amps I've seem them terminate in the high 9.3's but they stayed below the 45 celcius temperature that I will interupt a charge, while at a 19c room temp. I wouldn't do this with anything but once knowing the first discharged capacity I've pumped 7 amps in and 1Ah before it reaches the total discharged capacity or if any cell gets to 35c, I drop the amperage down to 3 amps and they will still finish below 45 celcius unless one cell is seems to be over 500mAh or so less capacity but that is a very rough number. I looked at the specsheet for these Panasonic cells and they say charging ambient temperature up to 40 degrees C but they don't specify at what charge rate or what their maximum safe temperature is. I figure that 45 degrees C is safe and approaching 50C is damaging based on information that I've come across. For a well balanced stick I've been able to charge at 3 amps with it almost hitting the 40 degree mark with a dT of roughly .5 degrees C at end of charge. At 1C they are supposed to be finished charging if they raise 1 degree per minute. The time between the cells being cool and getting hot is a short window. I took a stick that was charged up to 1Ah short of its normal capacity that it accepted before terminating and tested this at 7 amps, the temperature can be steady and feel cold until about 5 minutes from a full charge where the temperature rockets. It that they generate higher temperatures at 1amp charging than they do at 5 amps at the end of a discharge. Honda has good reasons to avoid regeneration at high rates into the high end of their capacity, even with the active cooling, I see it brutally destroying them in no time. I wonder how the car prevents a stick that has a low capacity that is hanging high in state of charge compared to the others from destroying itself because the delta V comes along and prevents a high voltage safety from tripping a positive recal and you can't measure negative delta V very easily with regen rates varying all over the place.
 

·
Premium Member
Joined
·
9,977 Posts
Thanks for making this public. I adopted recharagable batteries early on, my parents did too in life and it was pretty common for 1 cell in a device that takes multiple cells to go bad to make the device not work well or at all. I would put that cell aside and use another one. I tried to group those that died faster than others and vice versa so I do not over charge good cells as the charger them to be in groups.

Some cells would reverse, others just read dead. I found leaving them on the charger for a few days would help to bring them back some, but they were only good for powering led lights. These were nicads.

When white leds came out I use to use multiple alkline and other dead cells in series to power a few leds as a night light for the bathroom. Those went on for ever and as it got dim, I just added another cell. I had people who see it ask if its a fired hazzard to have a bunch of mismatched dead batteries wired together.

When I was disabled and involved in electric wheelchairs that used deep cycle batteries, I found the 2nd battery always had a lower voltage than the first battery. Never knew why, but would reverse their position and on a 3 day weekend hook them up directly one at a time to a 12 volt charger on trickle charge to balance them out.

Ive assumed nicards, nickle metal and lithium were dead once the membraine was damaged, the cells vented out the magic smoke or smell that makes theo work. I know lead acids its when the plates become so sulfured coated they touch and cause a short, thefluid boils out, the fluid is not the right ph or some jerk of a mechanic/battery sales person put an antiacid in the.

Any chance of seeing a Mike Dabrowski 2000 replacement prius 12 volt battery pack for your starter battery? :)
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #6 (Edited)
I got a little time on the subpack tester, and added some 1/4" terminals.
In years past I have used a capacitor to zap shorted NI CAD batteries back to life.

I ran some test on the first stick from Ron's "shorted cell" pile of subpacks.

Cell #1 (most positive) was lower voltage than than the others an hour after the stick was taken off the overnight 350MA charge.
I put a 3 A charge across the low cell, and instead of raising in voltage, it dropped, as it began to short.
I dug out the stored energy demo from the hybrid car class. We use it to show why you need to turn off the switch and wait for the capacitors to discharge before touching anything. The 6100uF cap has copper buss bars, and a voltage doubler circuit to get 350VDC from the 120V AC line.
I charged the cap to ~100V and discharged it right on the heavy brass wingnuts.The cell was still shorting, hit it again @250VDC and short is gone, and then for good luck, I charged the cap to 350VDC and zapped the whole subpack.
You can see that each discharge generates a divot in the brass nut and capacitor buss bars. Wear ear and eye protection as it sounds like an M80 going off, and molten metal can spray out if you don't make the contact fast and hard.After confirming that the cell voltage was rising at about the same rate as the other cells, I put the whole subpack on a 3A charge until the cells were at 90%, then finished with a 350MA for several hours. Final cell to cell voltage while under charge was within 0.05V
Will see what happens overnight and do more testing tomorrow.
I need a high voltage high current switching device to replace the unpredictable contact resistance and exploding metal with my present technique.
Photos
http://www.99mpg.com/Projectcars/mimapackwhack/
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #7
I whipped up a subpack tester that measures, graphs, and datalogs a subpack under different conditions.
The shorted cell has recovered to about 80%, and was the first to drop out.
Wrote a quick labview battery monitor program to watch the cells discharge through my 35A nichrome ribbon load.
Tomorrow I will add AH computations, and I/O to disconnect the load when any cell drops out.
I will build more loads until I can get the full 100A discharge, and will use my old Lincoln DC stick welder to get a 50A charge.
May as well test them at the real world limits.
Some photos:
MIMA Pack Whack and rebalancing the battery - MIMA Honda Insight Modified Integrated Motor Assist
To really know what is going on, we must look at the cell level, as thats where the problems are.
 

·
Premium Member
Joined
·
9,977 Posts
Wow.... Nothing like shorting a high cap, cap. BANG!!!!! Just make sure you do not get any body part between or on the terminals.

Thats interesting you can shock a cell back to life. Wish my friends were as smart as Mike. I recall some guy back in high school saying if you leave dead batteries out in the sun it gives them half their life back, alkline and recharge.
 

·
Registered
Joined
·
1,616 Posts
The one I have heard over the years and it doesnt seem to go away, is the idea that sitting a battery on concrete will discharge it. Total balderdash. But a lot of people believe it.
 

·
Registered
Joined
·
1,616 Posts
Back in the 1960s I worked for Geotechnical Corp. (later became Teledyne Industries) We looked for oil. One way was to throw dynamite off the back of a boat in the Gulf of Mexico while running at a constant speed and then measure the seismic returns. It was very dangerous and many a boat full of dynamite with crew was blown to bits.

We developed a system that used a huge capacitor about 20 feet long by 4 feet high by 4 feet deep. It was charged up and then discharged across an 8" gap in a device that was towed behind the boat. To initiate the discharge a wire was fed down to the gap and as the current of the towing caused it to trail back to the other side of the gap, an explosion occurred and the salt water was immediately turned into steam. It was the equivalent of many pounds of dynamite but was much safer....if you stayed away from the capacitor while it was charged.

Our first testing of the device was in the back parking lot in a fiber glass box full of sea water that had been built for the test. The first test blew the box to smithereens!!! So after that the testing was done in an above ground swimming pool that we assembled there. The sides were more flexible so it didn't blow apart.
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #13 (Edited)
I heard that one also. I think it started back in the days of electric cars in the early 1900, when they used less than perfect containers for the battery cases, and some corrosion of the concrete would happen? With today's plastics, it is not an issue.
That's interesting that the seawater explosively vaporized the water, I wonder if one could make an electrical discharge steam engine?

I ran the zapped subpack through a 350MA charge, and let it soak with the charge on and complete for 5-6 hours. I worked on the test program while waiting, and have an accurate AH measurement, individual low voltage drop out detect with audible feedback as well as control for a large contactor to turn of the discharge. I will build a load support system for making the high current connections to the subpack. 100A @ 6V is 600 watts, so it will get pretty warm.
An interesting effect of the charge, is that after the charge and the cells settled for several hours, the difference in max voltage that I saw up to the last charge has disappeared, and all cells are pretty closely matched. Direct evidence for cell balancing??
I worked on the dataloging, and now have the individual 6 cells, total subpack voltage, discharge current, and AH as well as a time stamp being stored in the spreadsheet file.
I will run the subpack through a 100A discharge and record the process later today.
Looks promising.
Spoke too soon. The "bad cell" was still the limiting cell in the subpack, and it seems to be getting worse on each cycle.
I connected a starter relay as the load disconnect, and it seems to handle the 100A fine, as well as tweaking the data logging software. I have a plan for a cheap 50A charger for the subpack, which will also use the starter relay to shut off the charge.

No time to play today as we were connecting the CAN cable to Jon's Insight2, and I have 4 MIMA orders to fill, I may actually sell all of the systems if things keep up at this rate.
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #14 (Edited)
getting better data

Had a bit of time this evening, so I ran a discharge trace using the improved software, and large 48V contactors. I can turn on a 30A charge and watch, or do a 70A discharge and watch. I remounted the nichrome resistors and put copper jumpers so they can be put in parallel. I ended up with one 5 strand resistor for the charge current limiter, and 8 strands for the discharge. The charge power supply is one of my 12V E-wheel AGM-31 batteries.
I put the cell traces on the same trace, which was set up to autoscale the time axis so all of the trace is on screen at the same time.
The trace clearly shows the difference in cell internal resistance (dc shift), and the capacity (discharge slope). Cant wait to try this on a "good" subpack.
http://99mpg.com/Projectcars/mimapackwhack/
 

·
Premium Member
Joined
·
9,977 Posts
Thanks for taking apart a cell. Id love to hear you insights on how it can be rebuild, reused or died. I read you looked for burn marks, holes, etc from zapping, but did not find any. Does the chemicals wear our? Does the insulator fail?

Was there any vents? I know you said it was tightly sealed.
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #17 (Edited)
I have a lot more cells to examine to answer some of those questions, I will post the results.
I had to open one, just to figure out the best way to do it, and to see how it is built. Now that I know, I will improve the technique, and get an idea as to what is normal and what is not. I think I lost too much energy in the Zapp to give the cell enough current, and have been researching how others have made the high current switch.
The simplest one was a copper hammer and plate, and the most complex was a giant SCR, which I just happen to have in my stuff pile. The thing weighs 5 lbs and is 4" in diameter.
It was for firing a 1KW CO2 laser.
It is hard to beat a copper hammer for raw simplicity and reliability, but using an scr would let me use the zapper as a cell spot welder, so it may be worth the extra effort.
I am tuning and calibrating the subpack tester so I get the most accurate data, and I will start abusing another stick. When I find another shorted cell I will give it a more respectable zapp with a better switch.
I used a dremmel with a grinding wheel to grind off the welds and separate the remaining cells, and I was able to make useable NIMH "D" cells for my Solar charged Led emergency light.
http://99mpg.com/mikestips/foreversolardeskla/
Can't stand to waste a useable battery cell.
 

·
Premium Member
Joined
·
4,389 Posts
Discussion Starter #18
The subpack tester is working pretty well now that I made a few modifications. I built preamps for the cell voltage and temperature circuits so I would get more resolution in the traces. The temp could still use more gain. The temperature sensors were mounted in a thermally and electrically isolated way to the bottom of the copper U clamps, with only the temperature sensor touching the cell, so the response is pretty much instant. The charge circuit can now be switched from 36A to 6A, to 350MA so I can gently top off the subpack after rapidly charging to nearly full. I will use the voltage drop between the 6A and 36A charge rates to make a cell impedance measurement. I cleaned up the controls and tucked in the temperature graph.
Now to get my batmobile pack on the bench and see what we see.
MIMA Honda Insight Modified Integrated Motor Assist - MIMA Pack Whack and rebalancing the battery
 

·
Administrator
Joined
·
12,445 Posts
That looks like a lot of work, just pop it in a box and send it over :wink:

Nice graphs can you share the source for the graph program? Not the program itself.
 
1 - 20 of 151 Posts
Top