A diagram setup of parallel and series connected Ni-MH cells

[numberonebikeslover]

It is how i'm about to do with Ni-mh Cells (of 38000 mAh and 1.2V) to get a 12 V pack.

i'm using diods (D1, D2, D3) to connect each series, so that each serial is irrespective of the other, and any change in once serial (voltage/charge) may not effect the other one. the same way again D4, D5, D6 are used to connect each with the charger for the same purpose.

plz let me know to go on with it, or any suggession would be highly appreciated.

Sincerely Yours
Naeem,
NWFP, Peshawar (Pakistan)

 

[Toorbough ULL-Zeveigh]

D1,D2, & D3 are good.
D4, D5, D6 don't accomplish anything.
You absolutely must separate the series strings to charge them individually or you risk burning down something.
D1-3 save you from having to physically disconnect anything while allowing you to charge individually, that's about the simplest you can get away with.

This is what happened to Jeremy Harris who also tried using charge diodes.
Other than the fact his diodes are hooked to the negative end of the batttery, the series circuit is functionally identical to yours.
http://endless-sphere.com/forums/viewtopic.php?f=14&t=3241

 

[fechter]

If there was a big hefty power resistor in series with each of D4-6, then you might get away with charging from a single supply. They would need to drop a few volts at charge current, which means the charger needs to be a slightly higher voltage. If they drop enough voltage, then it will behave like 3 separate chargers and prevent current hogging. They will get quite hot too.

I'm not sure what value would work or be safe, and it would depend on the charging current.

3 separate chargers would be much safer.

 

[Drunkskunk]

If there was a big hefty power resistor in series with each of D4-6, then you might get away with charging from a single supply. They would need to drop a few volts at charge current, which means the charger needs to be a slightly higher voltage. If they drop enough voltage, then it will behave like 3 separate chargers and prevent current hogging. They will get quite hot too.

I'm not sure what value would work or be safe, and it would depend on the charging current.

3 separate chargers would be much safer.

If I'm figuring this right, even then it might not be safe. If its running at currents high enough to charge all the packs at a reasonable rate, and one string has a bad cell, or has gone unbalanced, then all the current could dump down that string. Path of least resistance.

I suppose if you used thermosensors, you could prevent that from happening to a degree (No pun intended ) But the safest solution is what Fechter recomended. one charger per string.

 

[fechter]

With just diodes, as soon as one string gets full (or you have a bad cell) and its voltage dips, it would hog all the current and overheat. With resistors, only part of the current will go down the lowest leg, but it would still be more than the other ones. The higher the voltage drop on the resistors, the more even the current distribution will be.

Thermistors on each string would be a good idea.

 

[numberonebikeslover]

It is for the information of all the concerned technical people working on Ni-MH Cells charging that i've found this useful link on the internet which contains a circuit diagram of a Chip, namely MAX712 which is specially designed for smart charging of NI-MH Batteries

http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1666

Sincerely Yours
Naeem

 

[Dr. Shock]

This is a pretty interesting product range. I looked quickly at their other battery related stuff and it looks like they have some lithium chemistry chips as well.

http://www.maxim-ic.com/products/power/

Also maybe some items of interest to Fechter and the boys working on PIC based BMS?

I'll have to look at this stuff when I'm not tired.

Good eyes, Naeem.

 

[fechter]

They keep coming out with new stuff. Most of their lithium base chips are for Li Co and have too high a voltage.

Texas Instruments and National Semiconductor have similar items.

 

[monster]

numberonebikeslover if you want to have that set up i would get a 36v charger and then install a series/parallel switch on your 12v strings to switch between 36v and 12v for charging and discharging. no diodes needed.

thats what i have done with my bike and it works really well. only difference is i have 2 strings of 60 1.2v cells of 7ah and i switch between 72v for discharging and 144v for charging. i like to charge at same voltage as mains supply (no charger needed) but maybe i am just strange.

my other comment would be that you need some more batteries if you wanna power an electric bike !

 

[numberonebikeslover]

thank you very much all of you guys for all the valuable information and help you all are putting her. surfing through internet i came accross this circuit diagram i would love to share it with all of you is as under:

http://www.angelfire.com/electronic/hayles/charge1.html
Schematic is here:
http://www.angelfire.com/electronic/hayles/charge1fig3.html

Moreover, an idea just came to my mind while typing this.

i had an automatic lead acid battery charger that monitors the voltage rais and stops on a certain voltage. so the main triguring source of this circuit is the voltage of the battery. it stops let us say at 13.2 Volt and it is adjustable through a variable resister nob.

i thought of using the same for my NI-MH cells to monitor their voltage and stop when it gains the required amount of voltages. "BUT" i don't know what is the voltages a single NI-MH cell gets to, when we call it as the cell is full. if some one could tell me, then i can tune my charger to make the charger to stop charging at that instent.

Sincerely Yours
Naeem

 

[fechter]

With Nimh and Nicd, the battery voltage goes up as the battery charges, then the voltage drops a bit when the battery gets fully charged. Most "smart" Ni chargers sense this voltage drop to signal end of charge. Cell temperature also rises quickly at end of charge. At this point you either terminate the charge or go to a very low current.
If you charge at a low rate (like somewhere around 0.1C) the batteries can absorb the overcharge without overheating.

I wonder if the PIC code is looking for the voltage drop. They make a ton of dedicated charger controller chips that do this already.

What would probably work well is to use the existing SLA charger as a current source and have separate charge controllers for each string. The SLA charger would need to output a higher voltage than the battery string to make up for voltage drop in the charge controllers.

 

[safe]

Four Voltage Monitor - MAX6782–MAX6788

The MAX6782–MAX6788 are designed to monitor two to four battery levels (1% accuracy) and assert an active-low output indicator when the monitored voltage level falls below the user-set threshold. Each battery level is associated with an independent open-drain or push-pull output. Each of these independent outputs can be used to provide low battery warnings at different voltage levels. Each of these monitored levels offers fixed or adjustable hysteresis in order to prevent the output from chattering as the battery recovers from the outputs that can source up to 1mA. The MAX6789/MAX6790 monitor four overvoltage conditions and assert the complementary overvoltage outputs when any voltage at the inputs exceeds its respective threshold. The MAX6789/MAX6790 allow each trip threshold to be set with external resistors.

Four Voltage Levels - Four Separate Outputs

These are some really cool chips... and they only cost about $2 each... 8)

Imagine for LiFePO4:

2.5V - Absolute Low Voltage Cutoff
2.8V - Low Voltage Warning Light
3.3V - Full Charge Light
4.2V - Charge Termination