Recent Windfall - NiCd Question

[Madscientist267]

I recently came into a small cache of 'surplus' NiCd packs from a friend of mine. They are brand new, in the box, but aged to the point that they no longer have a valid response when 'keyed' into the POS system.

I'm currently charging/testing them, and am looking at disassembling the packs and reconstituting them to my own needs.

The question is, I am after 12V nominal, which is a dead easy calculation of 10 cells per pack, but I am wondering that since the VPC is only 1.2V, would it be worth my while (read 'beneficial/detrimental') to tack on another cell in each pack to bring the nominal to 13.2V?

I used to play with NiCds way back in the day, but have been more in the lead acid game for so long that I've really kinda lost the finer details of packing them, and wondered what you guys think.

They're not super high capacity, nor will the packs I create be, but I can think of a few uses for them. I'm considering the extra cell to squeeze out that last drop as they die off when powering an inverter or some other voltage sensitive device.

I realize that the discharge curve is more or less flat, hasn't been that long -  - but the price was right, and I definitely don't have a bunch of 'Turbo Racing' RC cars to put them in...

Thoughts?

Steve

[zap]

I believe Bruce regularly tacks on an extra cell now for a "12v battery".

If you consider that a good sla will almost always show somewhere around 13.2v... then the extra cell makes perfect sense.
Most electronics that rely on a "12v" power source should be able to tolerate the extra power of the full pack with an "extra" cell... which will probably end up somewhere around 15.5 to 16v hot off a charge.

I'm sure Bruce will add to this...

[dnix71]

A fully charged nicad has about 1.35v/cell which makes a fully charged 11 cell pack just under 15, which should be okay. Discharged to 1.1/cell, which is a safe number, you barely have 12v. That's way too low for a lead acid battery, but okay for most "12v" appliances, since nicads have a pretty good current output until they hit bottom.

Most single cell nicad chargers take them to 1.45v/cell, but that isn't necessary in your case. The trick to making them last is to limit the current during discharge so the weakest one doesn't get flipped by it's internal resistance during discharge.

[Madscientist267]

Excellent, Smithers.

So that leads me to my next question....

Paralleling...

Best to do cell-for-cell (like a Lithium pack), or set-for-set (like lead-acid)?

IOW, cells in parallel, or sets of cells in parallel for increasing capacity?

Steve

[Bruce S]

Steve;
 You have my attention
There are some considerations that will nee to be addressed before you build up the packs and I'll explained why from the real-world test I continue with.
The build is actually going to be based on what type of load you're going to put on them.
If it going to be inverter based then It would actually be better to build them up as 14.4Vdc packs so when the needed rush of current draw doesn't bring the inverter into LVD, the inverters will take the higher "standing" voltage with NO problem.
I have tested this using little 150watt units that plug into cig lighter ports up to 750watt Vector MSW units.
IF they are going to be used for motoring, then certainly the 12-cell would be best due to the voltage sag that seems to happen during start operation.

IF the load is going to be straight DC then the 11 to a pack will be just fine, actually the 10-cell will be too. The Packs that both Norm and I built for our LED units in the Philippines were 10-cell and they are doing just fine and with enough sunlight recover just fine.
I have built units in different ways and they all seemed OK.
Currently the way Norm is doing them is to build up 5-cell units for 6Vdc serial then parallel them for higher current (his is at 3.9Ah/6Vdc) then serial them up for 12Vdc.
My units are built in two fashions, 10-cell for LED lighting and attic fans, stuff that can be run on lower voltages all the way down to 9Vdc.
12-cell units are for inverters. I build them as 14.4Vdc then add banks to add up to higher Ahs. I used to build them up as high as 20Ahr banks and actually have a few still in operation, but there can be problems with a big bank like those getting too hot during a large recharge and bursting a cell or two 
I now build them as 14.4Vdc 16Ahr packs and use the tin connectors up until the end where I sue 12g stranded wire for the connections to inverters. IF I need higher Ahs I merely build up another 16Ah pack.
My oldest bank of packs is currently hooked up to a 300 watt (600 watt surge) Vector that makes the morning coffee into our 4-cup pot  , this setup is hooked to a 60watt HF panel set. still going 

The really cool thing about the NiCds is that you can actually take them down to 1.0Vdc/cell and they'll be just happy little campers, anything past 0.9Vdc/cell and like dnix71 said could get into cell reversal problems.

Charging: I trust the GHurd dump controller to do this task, they have NEVER failed me, actually I set it to about 16Vdc, this makes them about 90% fully charged, but safer than trying to take them up to full charge which is actually 1.47. When they get up around the 1.37/cell area you can feel the heat as the internal resistance to charging kicks in. I like safe  hearing one of the packs pop made me drop my cold beer the first time around.

Hope this helps!!
Let us know what you're gonna do and we can certainly help
Bruce S

[Madscientist267]

Hmm... interesting.

One problem I see with a 14.4V pack on an inverter though is that from what I recall, the inverter that I intend to use will cut out somewhere just above 15V, I think 15.5...

That being said, I wouldn't be able to use these as a 'floating' on-demand source. I've run some numbers and come up with a few constants that appear to put the voltage of these packs well over 17V when full and floating.

The highest constant coincides with the NDV point at full charge, and is 124% of the nominal voltage. On a 14.4V pack, this is ~17.86V.

Another is more based around the 1.45 VPC at full, before NDV occurs. 121% - rendering ~17.42V for a 14.4V pack.

Both are well above the cutoff.

The third, 113%, is the settled voltage after the pack has sat for several hours; makes ~16.27V for 14.4 nominal.

Granted, this is OTV, and I haven't tried it yet with a 12 cell pack on the inverter in question (second 6 cell is still doing a recovery charge) - but I can't recall if the inverter will 'trip' before it ever powers on to make the voltage drop from quiescent current draw.

You mentioned that you've never seen a problem, but is this after significant settling time, or can you start your inverters reliably with a pack fresh off of a charge?

I'm not going to be drawing insane amounts of current from the inverter, so a simple doubling or tripling of the cell count for AH capacity should be fine, wouldn't it? Thinking this way since there would not be significant voltage dips, and so a 11 cell layout might work better?

You guys are the experts on this one, LOL, and questions, yep, I got 'em.

FYI the packs in question (currently) are 7.2V 1700mAh. The other set is 9.6V 700mAh, but I'm probably going to use those either as is, or to directly power DC loads rather than an inverter.

Steve

[ghurd]

They really do not stand at >1.4V for very long after a charge.  Been a long time since I did much experimenting with them, but they drop a lot in just a few minutes?

The automotive type inverters (Vector, Statpower, etc) list the high voltage cut off as "About 15.0V" or "Around 15.5V"?
It seems they do not put a lot of effort into making it exact.

And my "general experience summary for the concept" is with 3 identical inverters, one will work fine at 16.5V, one goes into over voltage shut down at 15.5V until the voltage drops and it restarts, and one smokes at 15.3V.
Most of them do not have an issue until well past 15.0V.  Maybe nearer 15.6~15.8V typical?
G-

[birdhouse]

steve-
i'm no expert with these packs, in fact i've only used em in an RC car as a kid, but:

have you thought about building a ten cell pack, then have two extra cells attached with a small switch?  basically connect the pack to your inverter with just ten on, then shortly after the voltage has dropped on the ten cell pack, flip the switch to turn it into a 12 cell pack? 

infact, you electrically smart folk could probably make the above system automated with some fets, or resistors, or something like that...   

maybe this is stupid, but just a thought that popped into my head.

adam

[Bruce S]

Steve;
 The 1700ma ones should make a nice setup  the ones I use are 1300ma.
Packs: Here's my experiences, both will work. I still have 10-cell packs that work just fine down in the 300watt max load area. From the ends of the packs I use 12ga stranded wire to go to the inverters.
The reason I when with 14.4Vdc packs was due to the voltage nag I would get when attaching them to larger inverters, i.e. 750 units, they would if on a full charge power up, but complain ( internal cap screaming while charging up) and then if I put even a small load such as a 700watt Microwave built for installing into Semi's and the 900watt inverter companion, they would sag enough to go into LVD.
Once I went up to 14.4Vdc I stopped having that problem. I do not allow them to go into full charge , the cut-off for mine I have set at about 16.5 at this point I don't have to worry about the internal heat build up, the batteries get above the 90% full mark and I'm okay with that safe level.
NiCds can be tricky little buggers and that cutoff zone which is described in a great number of PDFs is considered a point where there's actually a voltage dip, then full charge. And after popping a few large packs (20Ahr) I backed the charging off to the 1.35 - 1.37 and I've had NO problem since then.


The other 700ma packs: I would be checking the devices you will be running, I have 12Vdc fans that run attic venting and they run perfect on 9.6Vdc NiCd packs,  culled from 19.2 packs, actually the fans run quieter, of course that's from the slower speeds.

Norm does some guru type stuff with these too, he may jump in here and add his experiences too.
GHurd is absolutely correct about the sitting voltages.
One of the other little tricks I've done with 10-cell is to attach the larger Ahr NiCd packs to a SLA that had the Inverter attached to it. I attach a GHURD dump controller to the NiCd pack set for it to dump at normal range of ~14.0Vdc. The NiCds get charged right along with SLA, but the SLA handles the in rush from the inverter.
SLA for me are very hard to come by, unless using old car battery, NiCds are much easier to obtain.

How many of those 1700ma puppies did you come into?
Cheers;
Bruce S

[Norm]

Steve,
If you are planning on using them for 12 volt.....
The basic idea is to try and reconfigure with as little
resoldering as possible....
What are these from? and what do they look like
cordless drill paks ....in 6cell stacks one cell stacks...
More detail send me pics of what you have....
if they are 6 cells in a pac.....I would simply put 2 pacs in series
and charge them to be at 15 -16 volts max with Ghurds dump
controller or whatever voltage max that the inverter likes.
The basic idea is don't tear it apart any further than you have to.
"don't fix it if it isn't broke"
"don't break it if you can't fix it "

so send me some pics or even text sketches.....dying to see them.
Norm.

[Madscientist267]

Okie. Quick and dirty, it is. But this is the current count.

Threw it together in a hurry, as it's late and I'm exhausted. Long day.



When I was posting the earlier stuff, I didn't have all of them to count up.

Most of these are still untested. In fact the good majority.

Unfortunately, there is only actually a single 1700mAh pack, I wasn't paying close enough attention when I glanced them over the first time. Wondered why the second pack took so much longer to charge... The first was actually a 1200. 

I'm not overly positive that this is the grand total that is available just yet. This is part of a purging process for old stock at the local electronics store. They came off the shelves and were being put into the store's recycler, when I was asked matter of factly like if I had any interest in them. Damn skippy! 

The fringies of being a regular.

There are a few SLA's for old camcorders, and a couple oddball NiCds here and there.

If that 3.3AH NiMH tests out, I got just the place for that, but let's just say I won't be able to take the mini-me laptop with me on a plane ever again...  LMAO

So, in a nutshell, I'd rather not mix and match the cells from different capacity packs for obvious reasons, and since there are only two 1200mAh 7.2V packs, I'll probably just tie them together in series for use as a down and dirty "I need a quick shot of 120V now!" supply in my solar "survival" kit.

Now for the 9.6V packs... Hmmmmm... Whole different story. Thinking lighting...

More as it comes...

Steve

[Norm]

The first bunch in the upper left corner looks like 8 AA rechargables hooked in series
right?
14.4volt packs would be nice......
You could take 2 of them and split them vertically down the middle, giving you 4 half paks
to add onto the 4 full paks.


 But first check all 6paks charge them up and test them out, you can even check 2 in
series at a time still in the package on the ends with very firm pressure and very sharp
pointed probes should read 1.75- 3.2v .

Actually the easy way out would be a couple of LED lights for the 2 batches of batteries
parallel the top = 1 pak 9.6v...6000ma,   enough LEDs to discharge it and still fairly
full brilliance after 10 hrs. for example.

Those SLA camcorder batteries don't sound bad either ....12volt 2 amp/hr. ?
Norm
top and bottom corner of left.
Parallel them

[Madscientist267]

There are 3 SLAs as of the moment, two are 2AH, the other is 2.3AH.

I have serious reservation however that these will ever see anywhere near full capacity. I've managed to make some progress with recovering SLA from lengthy sitting, but it's nowhere near an exact science. It's always been hit or miss, but with each 'resurrection', the results get better each time. The common key so far with the success stories with these is EXTREMELY slow charging, generally less than C/100. For small capacity (<5AH) it seems that closer to C/200 is even better). It's a game of patience with them; A 2AH can take a week or better at 5mA...  

But the NiCds seem to be bouncing back rather well. I figure the raw chronological age is the only factor in capacity loss, and probably works out to somewhere between 80 and 90% of rated. Since they are all 'brand new', there is little (if any) cell balance issue to deal with, and slow charging is just a precautionary measure. Some of them so far have been a little 'loose' in terms of impedance when the charge was first started, but after they bite and start clamping down (a few minutes at ~C/100) they appear to be taking the charges nicely.

I'm also curious as to whether the NiMH pack is going to behave differently than the NiCds when I go to charge it. I'm not overly impressed with NiMH as a chemistry for a few reasons. But, if it is recoverable, that extra capacity will be nice, for sure.

I'm considering modifying some 120V LED lights to run the LEDs straight from the packs as-is, and just making dedicated lamps for the packs, as you pointed out. Another hairbrain scheme that ran through my mind was to make a single one that runs from 120V DC (100 cells in series). Rumor has it that there are more of these packs on the way.

The LED bulb I have in question claims a draw of 100mA (a 7 watter), so it would be good for ~7 hours (assuming the 700mAh packs and that all are healthy). Not sure of the contents of the innards of the bulbs though, but on at least one version (a 4 watter) during an unrelated test some time ago, I was able to run it at full brightness with half wave DC (going through an SCR for control), so I'm going to assume that there's a bridge in the bulb and the AC just ends up as DC when it gets inside. If thats the case, straight DC should work just fine.

The only thing I can't figure out with that is whether I should use an RMS value (thinking so), or the peak value for DC input (120V vs ~160V). I imagine if it's some form of buck converter inside the bulb, it won't be that critical... ?

I'll know more after I pick up the next wave of packs, and see what I'm working with...

Steve

EDIT - Yes - the 9.6V packs are sets of 8 'AA' cells

[Norm]

I have a theory that a wall wart with just half wave rectification would
work well on a small SLA like that would shake things up a bit and desulphate
better than a full wave rectifier  maybe?
Yes ....I have a couple of old Camcorder batteri es like that I'll have to che ck them out
someday....
I don't really like too high of a voltage on batteries 4 12volt in series
is all I care about....
......I remember that one guy did mention that a treadmill motor ran a
table saw rather well with 90 volts.
Norm

[Madscientist267]

Just keep the charge rate REALLY low. Patience is the key with a lead acid. Even then, don't expect anything near full capacity.

Steve

[Madscientist267]

Thought I'd bump this up with an update...

A few of the NiMH packs that I have are exhibiting the occasional shorted cell.

Not too much of a problem (yet!) since I remember a trick my dad planted a seed in my head for some 30 years ago...

His favorite trick was to take the offending cell and wander out to the car and hit it with the jumper cables a few times to break up the crystals.

I wasn't really interested in isolating and removing the offenders from all these packs, so I had another hairbrained scheme - the Ghurd Controller! 

One of my controllers is set up so that I can either dump to the internal resistor bank, or out to another battery. While lead-acid was originally intended for this, I later discovered that the impedance of a lead-acid is way too high for it to really be effective (unless the 'dumped-to' battery was low, which I wasn't willing to risk damage by leaving one that way long enough to make this practical).

BUT - NiCd and NiMH of the 9.6V pack flavor have a nice tight impedance, and the NiCds in particular tolerate the abuse of being overcharged (to a degree), and so why not kill two birds with one stone?

I had my doubts, even though I had seen it work on single cells as a kid. But going through entire packs to 'weed out' the bad cell was never really brought up. So far, the one I tried it on had the cell broken through within 50 pulses or so, and the meter all of a sudden shot up to where I expected it to normally be (based on the good packs). I'm going to let it finish it's entire charge with this method. 

Time will tell if the defective cells will hold a charge, but I figured it was worth mentioning since it's a rare case anymore that something he said so long ago sticks around to the point I can bring it to good use this far down the road.

Will keep you all posted; the packs keep coming, and so far I have about a 95% success rate with them right out of the box. The few that haven't been so nice seem to be responding to a little 'sledgehammer persuasion' 

Steve

[zap]

A few thoughts...

If the memory is indeed from 30 years ago then chances are it was NiCd, not NiMH.
I've got a couple dozen AA and AAA NiMH and the zapping trick has never done much for any of my cells.  It seems like it always work for discharge but once charged up again it seems it will always be shorted out once again.

I really think "they" have tweaked the formula for NiMH too much to try and get higher capacity and/or less self discharge.
I've had some Eneloops for a few years and they are good at holding their charge but their capacity seems to have gone down since new.  My others are branded Energizer, Duracell, and even La Crosse Tech.  I've had cells go bad in under 15 cycles.

I've also got about a dozen Rayovac that are probably 15 years old or more and have started going bad in just the last year or so.  These were one of the first generation NiMH and I wouldn't be surprised if some of these have seen over 500 full cycles and are still going strong.

NiCd on the other hand... like you said... seem to tolerate cycling and abuse much more.

[Madscientist267]

There's definitely a mix of NiCd and NiMH here. They're all labeled, brand new out of the box.

EDIT - LOL I just picked up on what you were getting at there - Yes, the original 'zapping' trick as introduced to me was with NiCd. I don't think MH even showed up until when, late 80's - early 90's?  

In my experience, MH isn't anywhere NEAR as robust as NiCd, with the exception of the memory issues, which MH seems to be devoid of.

I'm carefully planning the use of these packs, and from the looks of things, I won't have to do a whole lot of disassembly and reassembly of anything to get the voltages I'm after.

I'm throwing them all at one project more or less, and quite a contraption it will be when it is all said and done.

If I do manage to lose a few here and there, it's no big deal. Hopefully it will happen (if it does) during the testing phases and not after I've got them all boxed up and running.

So far, 100% success at punching the shorted cells through, although I did have to do one of them manually; the dump controller wasn't hitting them with long enough pulses to break up the crystals.

I'll know more after the tests, but so far so good. I can't complain either way. There's about $500 retail worth of free battery here (by the time it's all said and done), and so I'm not going to be heartbroken one bit if a few of them take a crap on me.

I've already set aside (in my mind) which ones had issues from the git-go, so that they can be put through a little more rigorous testing before I hand their prowess over to fate and see if they can hang.

One thing is for sure, the project I have in mind is so far looking very promising. I'm going to spare the details on that just yet lest I jinx it.  

I just wish I had access to more of these (and theoretically I would), but unfortunately it was simply someone's "laziness" that brought me my fortune to begin with. Every cloud, eh?

Most every other location where these would be found have certainly gotten rid of them by now to a more EPA "friendly" home. Rest assured, any tree huggers, they won't end up in a landfill when I'm done with them. Promise.  

This whole project has made me take a closer look at the flooded variety as well... At least those are still made (although one needs a robust wallet and probably some kind of hazmat cert to acquire them). Such a shame too, I had the opportunity many years ago to pick up a whole load of flooded cells; I didn't know what I had in front of me at the time.  

Oh well, live and learn.

Steve

[Norm]

Steve ....google NiCad memory any number of articles on it
mostly it is a myth....
http://www.dansdata.com/gz011.htm
like this one for example....
Norm.

[Madscientist267]

Gotta tend to disagree on that - I have seen it take effect for myself.

It is not a prominent issue unless a cell is drained to exactly the same point each discharge, for several cycles in a row.

It was first noticed (or so the legend has it) in satellites where the batteries would repeatedly drain by the same amount while the satellite was in the shadow of the earth.

As long as they are discharged to varying degrees each cycle, the effect is minimal and can usually be 'erased' by a full, deep discharge.

Steve

[ghurd]

I agree with Norm.  With conditions.

I have 10 year old NiMh and older NiCd I use very often.
The trick, IMHO, is to charge them slowly (C/10) and never let them go over-voltage.

Also never let them sit unused for "years" (found some that were toast in the kids room that were not charged for 5~8 years?).

I am not so big on constant current unless there is a very very very very... sofisticated charger involved.

I am happy with my "charge it at no more than C/10 and float it at or below 1.4V/C".
I have a cordless phone that has the factory nicds, guessing 14 years old now.  Its twin (purchased at the same time) had the batteries replaced maybe 4 years ago (with USED batteries).
G-

[Norm]

I believe that the only problem with NiCads (and Ni-MHs also)is the users and the chargers
in the case of cordless phones ...most users have no idea what state of
charge the batteries are in....they just use the phone and 'hang it up' maybe
at night before they go to bed....in the charger....or maybe forget to put it
in the charger....or its in the charger all the time.

Another thing....in our case we have a cordless phone and I've noticied that it has
3 Ni-MH AAAs in the cartridge....one time I  noticed that the base of the phone
where the batteries are was very warm after a couple of hours....and needed
frequent recharges....but the voltage wasn't that low....
....but the fix was simply to shell out a few bucks for a new one (when the batteries
themselves only cost about a dollar)....life styles !.....Hah!
Norm.
Just rambling haven't had my second cup of coffee yet.

[Madscientist267]

Couple questions come to mind with this now that you guys bring it up...

On the C/10 rate issue, there seems to be varying opinion on whether or not this causes (or at least exacerbates) the dendrite issue, particularly in NiMH (leading to a shorted cell). You guys ever notice any correlation between the two?

The other, not that I have extensive experience with it, but I try not to float NiMH at all, they don't seem to tolerate any level of overcharge at all. My guess is that they gas more than NiCd, leading to dried up cells. Again, anybody noticing this?

I'm using constant current on one set of these now (NiMH), with my micro laptop, and had to add a thermistor to pull the charge rate way back as the battery heats up. It's a buck converter configuration with current limiting, which stays floored most of the time, only backing off when the voltage hits a certain point, or the thermie kicks in.

It's looking like I will need to also add an ambient thermistor as well to pull it back up, because when they are charged in a room where the temperature is above 75F or so, they don't get a full charge. The thermistor never allows the juice to get strong enough to get there. Conversely, they will easily go into overcharge if the room temp is lower than about 70F ish. The ambient thermie should be able to compensate for this if I do it right.

Guessing this is the 'very very very complicated' part of the charger you refer to G? 

NiCd seems to handle overcharge much better, although I'd tend to agree that it's not good to intentionally make them go into this state by design. But do you guys think that overcharge may be enhancing any memory effect? Or if not, how does it affect performance? Cycle life? Capacity per charge? Capacity over time?

Problem is with the laptop charger design, I need to give them as much juice as the laptop's power adapter can handle to bring the charge back up as quickly as possible. As it is, from dead, they take 18 hours + to reach full (need to adjust the sense resistor in the current limiter).

Another variable that I'm catching on to a pattern here with is that there are numerous production locations for these - Japan, China, Spain, Mexico, Canada, you name it. The Chinese versions have all exhibited at least one shorted cell after their lifespan on the shelf. So far I've been able to punch through them all, but time will tell if their performance will ultimately be affected. They are also younger, and all are NiMH. So far the cads are all happily taking a charge. Fingers crossed.

Steve

[Madscientist267]

Well, a little bit of an update...

The NiMH packs are all functioning correctly, and deliver very close to if not full capacity. They seem to have been awakened very well and I don't anticipate any problems with them...

The NiCds are actually the ones putting up any kind of fight; shorted cells (when discharged too far), and "leakage", meaning that some of the packs in the big array (10AH) are losing their charge on the shelf and pulling down the healthy ones with them. Provided the charge is used within 24 hours or so, self discharge doesn't make up a whole lot of the "usage". After 72 hours or so, the charge is only about 70%ish.

Another problem I have with them is NDV taking hold on certain select cells, robbing the others of a full charge. To compensate for this, fans were added to the box, and now I even go as far as to chill the pack during the later stages of charging. This does not come without it's own set of problems... See here for more on that.

Here are the pics with the construction details for the 10AH NiCd array:


The individual packs that make up the array. There are 20 of these in the box, bringing the total cell count to 100.



The spacing method for providing a gap for cooling air to flow between the packs. It would have been better if these could have been wider, but space is at a premium... Radio Shack doesn't make a bigger project box!



A shot showing the positive side wires connected and chased down the sides of the packs. The negative is similar, as one might imagine.



The packs mocked up in the box. As you can see, it really is a tight fit.



The wire chasing across the top, running between each pack. There are two wires running on each side on the very ends.



Detail of the jumpers tying each set of two packs together in series.



Here, the main wires have been connected to the individual tethers going to each string. Not immediately visible is the fact that in spite of massive imbalances in the packs, the wiring is set up so that it does not contribute to the problem. 5 positives on one side are the same length as the 5 negatives on the opposing set, resulting in electrically equivalent leads for all 10 strings. IOW, each string has a long and a short lead. They are offset at opposite ends of the pack so that 5 have long positives and short negatives, and the other 5 have short positives and long negatives.

The diode is a blocking diode to allow the charger to power the fans, but the fans stop if the charging stops. I'm still divided on whether or not this should actually be there. My original thinking was good, but I completely forgot the configuration of the meter box when I designed that part of it, and the result is that the volt meter reads high. How high depends on how much charge current is flowing. The fix was to make up another mating connector that tethers the battery to the meter box which essentially shunts across the diode, defeating it entirely. Oh well, live, learn.  



Another shot after dressing up the other end of the main wires.



Mocked up before putting the fans in. The tether shown is the first one, and allows charging through one port, and discharging through the other. As mentioned before, this was replaced with a single connector tether.



A shot of the cooling fans.



The 4 pin Molex power connector.



The air intake holes. This was fun - A Dremel is a beautiful thing...



Everything in place, just need to put the cover on.



A shot of the completed battery.


And here is the 9AH NiMH "prototype" 12V nominal array:


This thing performs virtually flawlessly. When it goes into a box (of similar size), there won't be any need for cooling fans. Very happy with this so far.

Steve

[Bruce S]

Steve;
Very Nice! zip ties are a lovely thing
I'm not surprised about the discharge relating to the NiCds. I am, however more than a little concerned about the packs that have the cells that are trying to go to Zero. These you might think about replacing when possible, they usually are not reliable in the long run and some on my packs have caused disastrous results  .
The NiMh units will have a much longer standby life, and the only think I can think of being their downfall is their number of full cycles before they are unreliable.

The boxes are a very good idea. A dremel is a fun little jobber fir sure!!
Looks like you're having a little fun
Cheers;
Bruce S