Voltage across individual cells

Could someone please tell me what the voltage across each cell should be when the battery is fully charged?

That's a curly one, but I would consider 1.43v and above to be fully charged, though I tend to either Ah count or monitor temp to be sure

At present my battery is about 40% charged and is showing 1.4v across all the cells which are accessible on the top of the batteries......except ONE..... this is the one at the back left which the motor controller goes to. This particular cell is showing 0.01v .... I presume this is not normal and indicates a bad cell? Could someone confirm this for me please?

That cell is dead as a door nail.
you can replace it, but if it's only one, don't bother, just bypass it.
Since it's on the end, it should be much easier.

I'd like to strip the battery packs and check each cell.

By the way, I just used a multimeter on volts and put the prods across the cell terminal without undoing any of the interconnects or anything. Is this the way to do it?

yup, that's the quick and dirty way to do it
You will find the dead cells, but probably not the dieing cells.

Matt

The results are one dead cell and 8 swollen cells which swelled so badly they broke the metal strapping and cracked off both ends of the plastic crate. They were the rear battery, right side and at the bottom.

Any idea what caused this concentration of swollen cells in one module?

No. I don't know. Not only are cells swollen but they have blue-ing from heat where the lid of each cell is crimped on.

Looking at the battery crates, it seems a bit daft to me that only the tops and bottoms have holes in them. I'd have thought the sides would too, in order to allow heat from the side modules to disperse. I may actually modify them by boring large holes in them> I htough about fitting 12v side fans too... computer ones..

Look at it a bit longer to understand the airflow through the battery. Holes in the sides would make things worse. The sides actually need to be taped up to close all holes/air leaks, otherwise the bottom cells get less cooling.

I'm a bit suspicious that the entire module failure of 9 cells is at the back battery, at the right, at the bottom.... where perhaps least cooling air runs?

Thinking out loud here of course....

The impellers should be sitting on top of the black battery boxes, sealed by the foam strips. They suck air through the black battery boxes. The air needs to enter through the bottom of each black battery box, so that it passes between all cells. If the black boxes let air in at the sides, then the bottom layers of cells get less cooling air flow.

It does not matter much if some air gets in through the white sides of the battery compartment.

I would check the serial numbers and production dates of the cells to determine if maybe this entire 9-cell module is much older or from a different batch than the rest.
See: http://visforvoltage.org/forum/7291-vectrix-down-under-whats-happening-new-zealand-and-australia#comment-41722 for how to interpret the Serial numbers on the cells.

I would check the serial numbers and production dates of the cells to determine if maybe this entire 9-cell module is much older or from a different batch than the rest.
See: http://visforvoltage.org/forum/7291-vectrix-down-under-whats-happening-new-zealand-and-australia#comment-41722 for how to interpret the Serial numbers on the cells.

I checked. Seven of the defective cells were made 14 June 2006 with individual numbers ranging from 117-155. Two others were made the next day on 15 June 2006.

Guess what? The rest of the battery packs comprise cells made in July 2007.....13 months later than the defective ones.

So... the blown module contains cells all made within a two day period in 2006

The rest of the batteries comprise cells made 13 months later and they are still OK.

10/10 to you on this one.....

The age difference alone would probably not have caused this complete failure of the module. I assume that it was inserted into the battery with a different SOC than the rest and then probably over-charged at 10A.

The method I used to install replacement cells:
(this is only a suggestion as I am sure there are many ways
to do this but this is how I did it - with no problems)
ensure the new/replacement cells to be installed are all at
the same potential. This is accomplished by simply connecting
them in parallel (DO NOT DO THIS WITH BAD CELLS) and leaving
them sit overnight. The high cells will bring the low cells
up and the low cells will bring the high cells down until
eventually there is little difference.
(+/- 0.05v; and I suppose it is up to debate if they
will ever be exact but this is close enough)....
Measure the per cell voltage for the replacement/good cells and
then multiply by 100 (number of cells in the bike). i.e. 100 x 1.3 = 130.0;
following: if there are only 99 cells in the pack then 99 x 1.3 = 128.7
Discharge the bike's battery to this per cell voltage and then
perform the replacement/installation of the good cells, observing all
necessary safety precautions of course.
Perform a charge cycle.

To me this seems easier than trying to adjust the new cells to the
existing pack level... however you decide to do it please let us know
your results...thanks.

Hi James
Have fun playing the voltages but they don't tell you much.
When I need to replace cells, I flatten the pack with the bike by riding it down to a crawl.
I then remove the batteries and break them into the strings (8s and 9s) of cells.
I put each string under a dc load of 300W (some halogen bulbs).
Then the voltage becomes very usefull.
As the cells discharge(a farily quick event as they are nearly flat) the bad cells give up first..
I change any that are signifcantly down (with cells with same SoC).
I also change any that are bulgded as they are always down on capacity.
This method is fast and brings the entire pack back to significant capacity.
If you start charging and discharging cells you can spend a lot of time (weeks) and in the end, you get pretty much the same result.
I am Australia and the ambient temps here are not kind to NMHIs packed like this.
I have seen a 56deg Batt temp coming off charge if the bike is in the sun.
The stock software brings the batt fans on at >36deg, that is the room temp here for a lot of the year !!
So we eat cells..I have changed about 30 cells in 4 years on two bikes.
Mik's batt cool fan mod and The Laird's SW are fantastic, and make a significant differance.
And try not to use regen into a hot pack...

http://visforvoltage.org/forum/7471-abcool-40-vectrix-auxiliary-battery-cooling-system

Could someone please tell me what the voltage across each cell should be when the battery is fully charged?

That's a curly one, but I would consider 1.43v and above to be fully charged, though I tend to either Ah count or monitor temp to be sure. Matt

How would I do the Ah count on a cell then, Matt? I've checked voltages and they are the same on all cells. HOwever, I gather that voltage is not the only indicator of health, right? Is there also a correct reading for resistance across each the terminals of each cell? Or any other reading that my multimeter is capable fo detecting?

Have a look at this thread, particularly around the linked post.

Detailed testing only makes sense when you have no access to replacement cells or if you need to recondition them in the process.

Looking for bulging cells is usually the most practical way. I learned this the hard way, it took months to test all the cells!

http://www.endless-sphere.com/forums/viewtopic.php?f=14&t=6277&start=30#p101043

You and I both Mik, I don't even use the cycling chargers any more, The lairds Software does a pretty good job of equalising the packs in the bike. James, one thing is a must but, you must make sure all the cells are at the aproximatly the same SOC, otherwise your first charge will fry your new cells.

... you must make sure all the cells are at the aproximatly the same SOC,...

there is nothing like a balanced pack - maximum range with maximum performance.

...otherwise your first charge will fry your new cells.

I suppose there are many variables which influence the outcome but I don't
necessarily agree with this... If the cells are good they wont get toasted on
the first charge - more likely what will happen is, over time, the unbalance
in the pack will become more pronounced and eventually the lower cells will go bad.
I am sure there are many scenarios and the heat of the weather can/does have a major impact.

I am refering to the inital SOC at time of fitting to the pac. Something like, If you flatten the pac to facilitate easy identification of low capacity cells, and replace say 10 or 20 of them with fairly well charged cells, then let the standard vectrix charger software loose on the pac, the new cells will be fully charged long before the origional cells, and the Vectrix charger will slam 12 amps into them for as long as it takes to achieve its goals, and these cells will get very very hot. and if per chance they are not on a temp sense point it will continue on blissfully unaware. Charging these cells hot is mainly what kills them and I try to avoid any charging near or over 40deg if at all possible..this includes the use of regen if I know the pac is that hot..

I've used a 12v battery charger to get the voltage on the replacement cells the same as the rest. I set the car charger to 'low' and left it on for about 15 minutes and it got them all the same.... 1.32v.... as the rest of the pack...... ...

Did you check again after a few hours or days?

NiMH cells can have very significantly different SOC with the same open voltage. The discharge curve of good NiMH cells is very flat between 70% SOC and 30% SOC. In other words, you can have cells with 40% different SOC that look the same if you just look at the open voltage. And the stock charger will kill the fuller ones or severely damage them immediately.

A better method would be to charge each module with your 12V charger until the module reaches a certain voltage while being charged. Maybe something like 1.4V/cell while charging. It might involve watching them for a long time, depending on the current your charger puts out. Your 12V charger could possibly kill (and maybe set on fire) a module if you leave it connected for too long. That's if the charge current is above 3A.

The best method is to discharge all cells to empty before combining the battery back together. The advantage of that is that (unlike with balancing by fully charging) they will remain balanced = empty for a very long time.

Your suggestion to do small charges and discharges initially can help to reduce the damage caused by assembling a pack with unbalanced cells, but you will still do damage.

I've used a 12v battery charger to get the voltage on the replacement cells the same as the rest. I set the car charger to 'low' and left it on for about 15 minutes and it got them all the same.... 1.32v.... as the rest of the pack...... ...

A better method would be to charge each module with your 12V charger until the module reaches a certain voltage while being charged.

THat's what I did. I checked the voltage on all the modules and then got the replacement one to the same voltage .... in my case 1.32v......

If the open voltage does not indicare an accurate SOC.... what actually does then? Basically, I have 12 modules of 9 cells, all with a reading of 1.32v per cell and 11.88v per module.......

As far as I know (I work at a company that uses lots of batteries and tests them) the only way to really know the SOC is by charging the cells fully, and then do a measured discharge. At my work this is a black box connected to the cell and to a computer. At home I would use a simple lightbulb, a voltmeter and a stopwatch. That way you know exactly how many Ah (Amperes through the lightbulb times time in hours) the cell can give until the voltage drops below 1.2v or whatever you choose.
If you charge the same way, thus measuring Amperes * Hours that go in to the cell, than you know exactly what went in.

Just divide the Ah-out by Ah-in * 100 gives a percentage that tells you how much energy is retained. The rest went into heating the cell.
Divide Ah-out by Ah-according-to-manufacturer * 100 gives a percentage that tells you the health compared to a brand new cell.

At my work we throw out a cell that goes below 70 % in the last test.

Again, as far as I know, there is no way to find out how healthy your cell is with only a multimeter. Believe me, we would do that if we could. It takes time.

I am refering to the inital SOC at time of fitting to the pac. Something like, If you flatten the pac to facilitate easy identification of low capacity cells, and replace say 10 or 20 of them with fairly well charged cells, then let the standard vectrix charger software loose on the pac, the new cells will be fully charged long before the origional cells, and the Vectrix charger will slam 12 amps into them for as long as it takes to achieve its goals, and these cells will get very very hot. and if per chance they are not on a temp sense point it will continue on blissfully unaware. Charging these cells hot is mainly what kills them and I try to avoid any charging near or over 40deg if at all possible..this includes the use of regen if I know the pac is that hot..

It is curious that Antiscab made a bms for a Lithium pack but has left the Nimh pack to flounder.
It seems to me the Nimh pack would benefit every bit as much, if not - more so, having the same
kind of attention given by the bms.
Just a thought....

It is curious that Antiscab made a bms for a Lithium pack but has left the Nimh pack to flounder.
It seems to me the Nimh pack would benefit every bit as much, if not - more so, having the same
kind of attention given by the bms.

I didn't actually develop the BMS for the Lithium batteries I use, www.evpower.com.au did

What I did do was make a few plastic parts and a video showing how to use cheap large format Lithium cells in a vectrix as a replacement battery.

A nimh battery is a bit of a different animal to that of a lithium battery.

full charge tends to be determined either by Ah count from completely flat (the only time the SOC can be measured rather than just calculated) or by measuring cell temperature rise.

to be honest, refridgerating the nimh battery to keep all cells under 20 deg C will fix most problems (mainly to do with self discharge) and is a somewhat cheaper and more effective solution than the ~$2000 a BMS would cost (102 cells with both voltage and temp monitoring as well as shunting)

put simply, charging the nimh battery slower and to a lower voltage (1.43v per cell) gives the best bang for your buck.
any more than that, and going lithium is both cheaper and easier.
So that is where my efforts have gone

Matt