Building a BIG NiMH pack and trying to charge in parallel?

See post #14 for more details.

I'm wondering if there is any way to safely charge NiMH cells in parallel. I've heard that it may be possible to use a thermistor to reduce the current flowing into a string of batteries when they begin heating up. What if the batteries were charged only to a maximum of say 80% or just below when the cells begin to -dv?

For my application I would like 3 strings of 72v. What if each of my strings used 70 cells for a nominal voltage of 84v but I only charged them to 72v using two 36v chargers?

The cells are Tenergy subC 5000mah, and I have ~700 at my disposal.

Each string must be charged separately.
This is easily done by adding D-Sub connectors for string separation and power output..

Lets say you choose a female 25 pole D-Sub. There is also 37 poles and 50 poles available for higher currents
Each pin in the D-Sub connector can hold approx 5 amps of current. You will use 24 of the pins for the string connections.
Solder each separate NiMH string terminal to a number of pins in the header.
In this case use 3 pins positive and 3 pins negative in parallel for each string (3x5Amp = 15 amps of current handling capacity/string).
When the D-Sub is open all strings are separated from each other.
Now you plug in the male 25 pole D-Sub that is connected to three separate NiHM chargers.
Each charger output has to be soldered to one set of 3 pos and 3 neg pins in the connector (going to a NiMH string).
Each string is now being 100% separately charged by their own NiMH charger.

Meanwhile you have to replace the controller battery connector with another male 25 pole D-Sub.
This connector has all its positive pins joined and negative pins joined together for a regular 2 wire battery output.
When you plug the connector into the battery D-Sub it reassembles the separate strings to one big parallel battpack.
The D-Sub pins (the 24 pins now used as 12+12 parallel) can hold 45amps or more. That should be enough for your needs.

Did you get it or do I have to draw a picture?

if you parallel the cells on the cell level the risk is much less of there being a problem.

However you can do as you intend with strings but you would have to either charge at a trickle charge rate empty to full in 10 hours fastest to be safe or charge to a voltage below the peak voltage so that you don't get voltage drop off.

i Have reliably charged Vapextech 12v sticks in parallel before using the zero peak function on the charger.

Charging using a CV method worked well for me when i ran NIMH as long as you periodically trickle the cells to balance them up.

what i did was Delta peak charge the cells and record the voltage the cells peaked at and then CV to a voltage somewhat below that voltage.

Chilledout, what kind of charger were you using?

Honk, not sure I followed your post, could you provide a drawing?

I also saw another post by a member who used a schoktty diode, but I'm not really even sure what that is. I think another solution for me would be to automate the charging process, so that after the taking ~2.5 hr's to charge the first string a timer would trigger a switch to start charging the second and eventually the third string. Not exactly sure about how I'd go about doing this but it doesn't seem too complex.

Ideally I'd still like to charge all the batteries at once in parallel using some clever circuitry so i can fully recharge overnight without having to constantly disconnect and move chargers. Another benefit i see from charging them all in parallel is that each cell would be charging at a lower C rate.

Anyone else care to chime in? Can the thermal runaway process happen if the cells are balancing whwn not in use or when they are being discharged in parallel?

Do you just want to be able to charge them overnight?

In that case simply charge them at a C/10 rate and your packs should be fully charged after 12 hours.

This is the safest way to charge packs in parallel.

so if you had a 10ah pack that would mean charging at 1 amp.

I still recommend paralleling at the cell level

The charger i use is an duratrax Intellipeak Ice.

SHARKBITEATTACK said:

Honk, not sure I followed your post, could you provide a drawing?

Click to expand...

Well, I don't really have the time for a drawing right now but I'll do better than that.
Here's some pictures on my present eBike battery using 120pcs of 2.2Ah NiMH AA cells.
It's built as 12 separate blocks of 10 cells each. Each cellblock of 10 cells is 12V 2.2Ah.
It might look like all cells are connected but they are not, there's only 10 cells per string.
Each cellblock within the pack goes to a header. The header provides separate connection to each block of 10 cells.
Now all 12 blocks can be connected as needed. Parallel, series, or parallel and series.
Minimum voltage at 12P = 12V 26,4Ah. Maximum at 12S is 144V 2.2Ah. At 6P2S the output becomes 24V 13.2Ah.
I use it connected as 4P3S, meaning I get a 36V 8.8Ah batterypack. Total pack weight is 3.5kg.
When charging the pack I just insert the 12 channel NiMH charger into the 24 pol header.
Each string of 10 cells is then being charged completely separate from each other.

As you might have noticed I use a regular 24 pol pinlist for battery connection.
It works OK but there is always a risk for polarity shift when inserting, if not being careful.
In my next battpack with 180 cells I intend to use a 37 pin D-Sub that's keyed....much better and safer.


The battery holders being loaded by 120 cells. The holder core is maid from regular PCB.


Another view. The cells are connected by adhesive copper tape onto soft rubber.


Top view of the fully loaded battery


The batterybox is made from 0.7mm thin PCB. Not sturdy but very lightweight and watertight.


Side view of the silver painted batterybox.


Cellblocks shoveled into position......


Lid is closing....


Battpack is finished. All positive 12V channels within the header is facing the viewer.


The 12 channel charger and its powersupply is attached and working great.


Close-up of the 12 channel charger. When a LED is switched off = String of 10 cells is ready.

Wow honk! That is by far the most creative battery pack I've seen! Thanks for explaining it all.

I'm glad to be of service to you. 8)
Perhaps you can build something similar and really make good use of your 700 cells.

Nice! Where did you buy your pcb, copper and rubber? I want the same!

Hi.

For the battery holders you can use any 1.60mm standard FR4 PCB etched clean from copper or some other material with enough stiffness and properties.
Then measure and cut it accordingly to the size of your cells. I use a string of 10 cells in two layers of 5 pcs.
You can etch away the copper film but leave some copper for good cable output termination of each cell string.
For soft cell contact I bought a 1mm thin rubber sheet being cut into same size as the PCB holders using a sissor and then fastened by dual layer tape.
You also need some adhesive conductive copper tape and this is cut as strips to connect the cells in strings of 10 when using NiMH. Your choice of cells might differ.
One more thing, you need to insert a thin sheet of plastic A4 film in between the two large cell blocks to prevent the cells from any contact, if not a short circuit may occur.

The box was simply made from 0.7mm PCB.
I just cut the PCB into pieces to form a box of the right size. Then I soldered a thin 0.1mm copper sheet lengthways onto the edges and bent it into 90 degree angle.
The next piece was soldered to the bent copperheet and so on for the rest of the box pieces.
The lid is just kept in place by silver tape, no screws. It's almost invisible when compared to the silver painted battery box.

wow wow wow

Thanks for all these details. That's excellent. Time for me to check price in France and to buy some of these items

http://www.all-battery.com/rectangularnicd24v5000mahbatterydcpowerpack.aspx

same stuff.. 40a max discharge..can I buy 3 and serial them for a 72v5000mah pack? Would be great on a HT motor that draws very low amps....5ah on a HT motor with high voltage would get you like 10-12 miles at WOT

SHARKBITEATTACK said:

I'm wondering if there is any way to safely charge NiMH cells in parallel.

Click to expand...

Yes--if you design and build something to:
--monitor the voltage and temperature of each cell, constantly
--disconnect each paralleled string/cell from the rest if it becomes fully charged per that monitor
Then it would be pretty safe to charge them in parallel.

But that's gonna be pretty complex, and expensive, and large, and heavier than needs be, because you will also probably want this during discharge.

See this wiki entry for links to the many discussions about this, for more information. It's been discussed to death.
http://www.endless-sphere.com/w/index.php/Batteries_NiMH_%26_NiCd
http://www.endless-sphere.com/w/index.php/Chargers_NiMH_%26_NiCd

There are simpler ways to do it, assuming all the cells are the same, and haven't changed as they age, but if you want a definitely safe way, that's the best I know of.


Paralleling them on cell level, as is discussed above, will also work, but only as long as the cells are identical. Once they age enough that one cell is full and suddenly drops in voltage while the others are just about to reach their peak voltage, then that cell is essentially shorting out all that energy from all of it's paralleled brethren, and now has to dissipate the heat not only from it's own end-of-charge reaction, but also from all of them, too. If the charger is not setup to detect that this group has now dropped in voltage, then it will also continue pouring current thru the group, adding even more energy to the problem.

If it's enough energy, it could cause a fire, and there is nothing you can do to stop the energy from flowing into that paralleled cell, either, unless you disconnect it from them. But even if you had relays of some type between every cell to be able to do that, you still can't easily detect *which* cell just shorted out the group, unless you can measure very small voltage drops across the cell interconnects.

Even if it's not enough energy to cause a fire, it *will* prematurely age the cells that get hot, and make the problem with the already-aged cell that started the whole thing worse, so that each time it happens, it'll happen sooner. The good news is that eventually it'll reach a point at which the rest of the cells haven't reached a state where they have much energy to contribute to the problem....

So its a rainy day and and I can't go biking so I'm looking at my batteries debating what to do with them. Since I have the time I'll explain in a little more detail about what my goals are.


Here's the current setup. 36v, 2808 9C, infineon 40A controller. I've been charging these batteries for the last 2 weeks using a homemade charger made from three 12v 1A wall wart transformers hooked in series, and then trying to balance each individual 12V pack with a light duty automotive trickle charger. So far this method has worked mediocre. I'll get about 5-10 miles of range with pedal assist and then one of the packs will drop to 10V while the others are still hanging in the 12v range. I'm eventually going to bump the voltage up to 72v but I want to add more batteries in parallel first because the current draw would be too much for the batteries.

View attachment 4
Here's an individual cell from tenergy's website. http://www.tenergy.com/10514

Features and Benefits

-High quality Sc size, 5000 mAh Nickel metal hydride (Ni-MH) battery.
-Up to 30 Amps high drain current rate, ideal for making various battery packs for heavy duty applications such as remote control (RC) toys, electrical guns, power tools and backups.
-Solderable taps enable cost effective and easy way to make your own power pack of different voltage and shape.
-Each cell is individually checked before shipping and all the batteries are matched.
-Ultra high capacity, 40% more running time than Ni-Cd Sc battery.
-Very long cycle life and Rapid battery charge up
-Significant savings (60 % or more) from any retail stores.
-Battery tested based on International Electronic Commission (IEC) standard to ensure capacity, quality and life time 6
6 months warranty.

Technical Specifications

-Dimension: Height 43 mm, Diameter 23 mm
-Weight: 71g per cell
-Capacity: Min: 4600 mAh
-Voltage: 1.2V
-Single Cell Resistance <=5 Mili Ohm
-Standard charge 0.1C × 16 hrs
-Rapid charge 0.5C×2.1hrs approx
-Standard Discharge .2C

The Tenergy website is kind of vague on the details of this cell. Also they say that rapid charge is rated at 1C which would be 5A but other sites say .5C or 2.5A.



Here's what the batteries look like with the shrink wrap off. They are 10 cells in series for a nominal 12V. I have no idea how much current those tabs are rated for but they are stacked 2 high. (These are not the tabs that can be ordered with the cells. A battery shop assembled these.)


Here's my current Battery rack. It's 2 strings of 30 cells that get hooked up in parallel. (36V 10Ah) I plan on keeping this and putting these packs in series once I go to 72V.


Finally, here's the rest of the batteries I have and I plan on incorporating a good amount of these into the bikes main triangle. I would like at least 3 strings of 72V which would give me a batery pack thats a little over 1kWh. I have 3 of these battery modules (pictured) and they each contain 240 cells arranged in a 80S3P (96v, 15Ah) config. They also have a center tap in the middle (small red wire) so that they could be charged using dual 48v chargers. Unfortunately, all of these cells have all been tab welded together and were welded in parallel first and then series! This makes them impossible to break the strings apart by simply snipping wires/tabs. The vehicle these came off of has had problems with cells overheating and bursting in the past. I would guess that having all the batteries hardwired and charging in parallel may have something to do with that...

http://www.tenergy.com/01016

I recently just purchased 2 of these NiMH smart chargers and they should be arriving any day in the mail. They charge at 3A which I think would be perfect for charging a 15Ah pack (3 strings simultaneously). Now I know this cannot be done with conventional methods, hence the reason I'm looking for a solution using some clever circuitry. *Ideally I want a robust system so that I can charge all the cells at once at a low C rate and be able to charge from empty to full overnight without risking any fire hazards. I think I may be chasing after a goal that is unattainable, and if I have to charge 1 string at a time and then manually disconnect and reconnect the next string then so be it. I guess I'm just lazy and want to be able to charge as simple as possible :wink:

Also, the strategy of charging in parallel at C/10 has been brought up. This seemed like a good idea until I came across another members post (Jeremy Harris I think?) who was doing this and woke up to the sound of his NiMH cells exploding

SHARKBITEATTACK said:

-High quality Sc size, 5000 mAh Nickel metal hydride (Ni-MH) battery.

Click to expand...

I'd personally doubt that 5Ah rating, though I guess it is *possible* to put that much in that size. I had F-size cells that were (when new) 13Ah. C-size that were (when new) 9Ah. But the sub-C that I had, even the "good" ones, were at best about 2.5Ah. They were all made a few years back, though, so tech could've improved.




-Standard Discharge .2C

The Tenergy website is kind of vague on the details of this cell. Also they say that rapid charge is rated at 1C which would be 5A but other sites say .5C or 2.5A.

You might well be able to charge it at 1C, but they may get pretty hot that way, which individually isn't an issue, but buried inside a bigger pack certainly can be. It's worse when you consider the energy dump / heat conversion that happens at full charge inside each cell, because that heat is in addition to the heat generated from cell resistance.

I've forgotten exactly what my packs charged at, but I know that I damaged the F cell packs when the thermistor in the pack became intermittently disconnected from the charger, which wasn't more than 2-3A max. The charger was looking for the sudden heat increase at the end of charge in order to stop charge, but since the thermistor wasn't connected inside at that time, apparently (had been at start of charge, or it would've refused to start), it kept pouring current in there and overheated/overcharged the whole pack. It lost a huge amount of capacity, and increased resistance a lot, as afterward it sagged a lot more than it had before.

Unfortunately, all of these cells have all been tab welded together and were welded in parallel first and then series! This makes them impossible to break the strings apart by simply snipping wires/tabs. The vehicle these came off of has had problems with cells overheating and bursting in the past. I would guess that having all the batteries hardwired and charging in parallel may have something to do with that...

Click to expand...

That's pretty likely, unfortunately. As noted before, as soon as any single cell in a parallel group reaches full charge, it's voltage drops, and it's partners at a higher voltage *plus* the charger itself are now pouring current into that already-full and already-hot cell, making the problem worse, unless there is a thermal sensor on *each cell* that the charger will respond to by stopping charge once that happens.

(and if there is enough energy in the other almost-full cells to dump into the full one that's now lower voltage than they are, it won't matter that the charger isn't putting current in, cuz they still will).

*Ideally I want a robust system so that I can charge all the cells at once at a low C rate and be able to charge from empty to full overnight without risking any fire hazards.

Click to expand...

As noted by others like Jeremy, and above, it doesn't matter what rate the charger is charging at, or even if it's running at all, if the paralleled cells that aren't full yet have enough energy to dump into the full one(s).


It *is* possible to do what you want, safely....but you'd have to do it like I described, and it's complex and probably large, (maybe bigger than the pack?) and probably expensive if you buy the parts new.

amberwolf said:

SHARKBITEATTACK said:

-High quality Sc size, 5000 mAh Nickel metal hydride (Ni-MH) battery.

Click to expand...

I'd personally doubt that 5Ah rating, though I guess it is *possible* to put that much in that size. I had F-size cells that were (when new) 13Ah. C-size that were (when new) 9Ah. But the sub-C that I had, even the "good" ones, were at best about 2.5Ah. They were all made a few years back, though, so tech could've improved.

You might well be able to charge it at 1C, but they may get pretty hot that way, which individually isn't an issue, but buried inside a bigger pack certainly can be. It's worse when you consider the energy dump / heat conversion that happens at full charge inside each cell, because that heat is in addition to the heat generated from cell resistance.

I've forgotten exactly what my packs charged at, but I know that I damaged the F cell packs when the thermistor in the pack became intermittently disconnected from the charger, which wasn't more than 2-3A max. The charger was looking for the sudden heat increase at the end of charge in order to stop charge, but since the thermistor wasn't connected inside at that time, apparently (had been at start of charge, or it would've refused to start), it kept pouring current in there and overheated/overcharged the whole pack. It lost a huge amount of capacity, and increased resistance a lot, as afterward it sagged a lot more than it had before.

Click to expand...

I think these batteries that I have were purchased in 2010. Do you know what brand of cells you were using? On Ebay there's junk ass Chinese cells and the sellers like to inflate their real capacity numbers by 100%! Also, the Sub C cells tend to have higher energy density than cells of other sizes because they are specifically made for RC vehicles which demand high discharge rates. I think Lipo has stole alot of the battery powered RC market in the last few years though...

http://flashlightnews.net/forum/index.php?topic=2440.0
Here's a thread I came across from a flashlight forum where a guy tests 5Ah Sub C cells against 5Ah C cells. In his test he measures the amount of light the bulb puts out with respect to time. The Sub C Cells were able do deliver 10% more brightness due to less voltage sag but they discharged a little quicker. I guess in terms of capacity it would be a wash. I didn't understand all the jargon but I was trying to find find the power consumption of the specific LED they used to calculate the amount of Wh used.

Do you know if your charger had any other built in safety measures to protect against overcharge? The chargers I bought are specifically made for the batteries I have and have 5 "conditions" that terminate the charge.

Beautiful house. Put battery outside house and charge ( weber ) and keep hobby battery out of the garage as that's the garage is under the house. Hobby.

SHARKBITEATTACK said:

I think these batteries that I have were purchased in 2010.

Click to expand...

'
Did your store them really cool during these years?
The NiMH I'm using was bought in midyear 2007 but I never got the time to finish my project until early 2013.
Meanwhile they was stored at the bottom shelf inside my refrigerator at approx 5 degree C.
When I tested them this year they had maintained all their capacity, nothing seemed lost.
But a string of 10 cells I had laying around in room temperature lost approx 20% capacity.
Keeping spare cells stored cold is really good for shelf life. 8)

SHARKBITEATTACK said:

Do you know what brand of cells you were using?
<snip>

Do you know if your charger had any other built in safety measures to protect against overcharge? The chargers I bought are specifically made for the batteries I have and have 5 "conditions" that terminate the charge.

Click to expand...

I *think* the F cells were SAFT? I can't remember right now, though I know it's posted about in my DayGlo Avenger and/or CrazyBike2 threads. Those were the white ones.

I don't know the brand of cell for the C size, but they were from a Giant brand pack for one of their ebikes. Both of these came from an ES member (sorry I cant' quite remember the name) as well-used packs with known issues, and I didn't help them much in my learning process. :lol:

There were also some more F cells from EbikeFanatic (who used to be IanMcnally2 before some forum problems, IIRC, if you go looking for my posts about the packs in the DGA and CB2 threads) but I don't know what brand they were. The chargers came with them, 36V and 24V, IIRC. All they had for protection were thermistor for delta-Temperature; if they had delta-Voltage protection it didn't work very well, if at all. I think they were "hi power" brand. There was also a Tenergy 12V charger with both thermistor and voltage detection that did work, but I think I only used it on the "lighting" pack I built from some of those first F cells.


The sub-C cells were from powerchair packs for something called "E-Motion", IIRC. Round blue plastic containers that popped into the hubs of the wheels. Those packs actually had a BMS that (IIRC) monitored individual cells for voltage, and the pack as a whole for temperature.

On Ebay there's junk ass Chinese cells and the sellers like to inflate their real capacity numbers by 100%! Also, the Sub C cells tend to have higher energy density than cells of other sizes because they are specifically made for RC vehicles which demand high discharge rates.

Click to expand...

It's possible; they are also used for power tools often enough. All the old DeWalt tools pre-A123 used sub-C NiMH AFAICR. (I used to have a bunch of those packs before the house fire, awaiting disassembly and testing to pick out all the still-good cells, but I think they've disappeared with looters since then like a lot of other things).


I can't get that test link to load right now (free wifi around here doesn't work on all sites all the time) but it sounds about like I'd expect: higher discharge rates usually lead to (significantly) lower usable capacities--often a lot less than the "rated" capacity, because they usually rate that at a very low discharge rate--nothing close to what we use with ebikes.