Battery Capacity Measurement???

[Sajjadkhan]

DONE: I have build an Efficient Sealed Lead Acid battery charger which can charger 200AH battery in less than 5 hrs. Its a micro controller based. It provide constant current as a first phase (constant current ~= battery AH rating) till the voltage reaches to 13.8v then in second phase voltage is doped to 14.4v and finally charger enter into standby mode.

Change i wana:

Since the charger is being commercially used, i want to modify it in such a way that it can connect minimum of 5 batteries and max of 10 batteries. So if one battery is charged it switches to the next one.......

Problem:

As there would be multiple batteries so there AH rating can be different and the charger needs to know the AH rating of the battery for its first phase.How to determine battery AH rating using some circuitry or etc.... any help will be appreciated.

 

[amjadali56]

can you post your scham for good sack?

 

[atripathi]

You may want to limit current being used for charging. Have not heard or seen a circuit which can help "measure" AH of battery.

 

[ea.arun]

battery SOC measurement we can be achieved using charge/discharge algorithms....not heard of AH measurement...

 

[RCinFLA]

You cannot really tell the AH of a lead acid battery without discharging it some amount. The closest, damage minimizing test, is to discharge it by about 30-40% of expected AH then let it sit for a couple of hours to allow battery to reach equilibrium rest state. Then measure terminal voltage and check it against expected SOC open circuit voltage.

If open flooded cells, best test for SOC is measurement of specific gravity of electrolyte. Cell voltage is very close to SG +0.845 volts. 2.12v rested voltage is a fully charged cell with SG of 1.27 to 1.28 at 25 deg C.

90% SOC = 2.103v, SG = 1.258
70% SOC = 2.062v, SG = 1.217
50% SOC = 2.017v, SG = 1.172

The charger could measure the rate of rise in voltage with a given current and adjust its rate based on voltage rise. The issue with this is it only works if battery is good. A bad battery may rise too quick if it has high internal resistance due to grid corrosion or too slow if there is high internal leakage.

For a three phase charger, the first constant current phase should be maintained until battery reaches bulk voltage (14.3-14.5 vdc for a six cell battery). Typical maximum bulk current rate is 10-15% of AH rating. An AGM battery can go to 25% AH rate. It then maintains bulk voltage regulation until current drops to about 2-5% of AH battery rating. Alternate approach is to just time the absorb constant voltage period. If current termination is used there should still be a max timeout to prevent a battery with high leakage never making it out of absorb due to current never dropping to 2-5% AH level.

 

[Sajjadkhan]

The charger could measure the rate of rise in voltage with a given current and adjust its rate based on voltage rise. The issue with this is it only works if battery is good. A bad battery may rise too quick if it has high internal resistance due to grid corrosion or too slow if there is high internal leakage.

Yup battery health is an issue here.... Your statement gave me an idea. Well it will not give you an exact amount of capacity but the reaming capacity. Just like in our laptops remaining capacity is measured as change in voltage after every 10sec when. Determining the rate of discharge can lead you to how much capacity is remaining. e.g. a high capacity battery will discharge slow along time compared to a low capacity battery. So current can be adjusted according to (Current = 100 - Capacity remaining). The load for discharging would be some power resistor and will be constant. For this i have to perform various discharge experiment on various sized of batteries, acquiring their graph and then model it for micro controller. What do you think?

Typical maximum bulk current rate is 10-15% of AH rating. An AGM battery can go to 25% AH rate. It then maintains bulk voltage regulation until current drops to about 2-5% of AH battery rating

I have a doubt in this because i have followed a website when i developed my first fast charger about 1/2 years back.

Fast charging of sealed lead acid batteries. ( Para 7 and 8).

Till now its working fine. And batteries are still healthy. But yes i have tried that charger on SLA batteries not flooded type batteries. I am charging these flooded batteries for a 2 weeks...lets see how well they perform!

Also i found a video: YouTube - ‪fast charging SLA batteries in 19 min‬‏

He is charging a 38AH battery with more than 60 A current...wow

 

[RCinFLA]

You won't get homgeneous recharging at large charge rate. At high current you will only get to 50 to 60% SOC before gassing starts. You will creat oxygen/hygrogen and vent gas reducing the electrolyte. This will reduce battery longevity.

 

[Sajjadkhan]

OK thanks. Now how to limit the current? I am posting the basic schematic which i have applied.
The regulator TL783 is same like LM317 but have a wide input range (upto 120V) but it can only give current up to 800 mA. Thats why i added transistor Q1. There are many transistors as Q1 in parallel but for simplicity i put just one. When the current across R1 is enough to produce 0.7V,Q1 turns on and contribute. When switch J1 is towards Radj, the regulator acts as constant voltage mode i.e. 30V for charging 24V battery. The resistors Rs1,Rs2,Rs3 (which actually up to Rs8 but for simplicity i limited to three) are like Shunt resistor which i made from copper wire. IN fig suppose it is of 0.03 ohms and three point have been taking dividing it into 0.01 ohm. When j4 is away from Radj then the regulator is in constant current mode. e.g. when J1 is closed and j2 and j3 are open then current will be 1.25/0.01 = 125Amps and so on. I this way i have limited the current. but i need 8 different constant current modes. and for this i need 8 switches, i can use 8 relays which i really dont like. the circuit will be big, noisy and bulky. With what i should replace them with??? I am thinking of Analog Mux but i couldn't find which handles more than 15V input, since in constant current mode voltages varies and can go up to 48V(in my case). and yes MAX IC's are rarely found in my country so dont recommend them.

 

[RCinFLA]

The current regulation is 1.27vdc / 0.01 ohms = 127 amps, 63.5 amps, or 42.3 amps. I doubt that is what you intended.

Going from 60 vdc to 27 vdc at any signifcant current is going to generate a tremendous amount of heat in the series pass device.

Darlington series pass will likely give you too much gain and not allow enough current through TL783 to keep regulator stable

 

[Sajjadkhan]

The current regulation is 1.27vdc / 0.01 ohms = 127 amps, 63.5 amps, or 42.3 amps. I doubt that is what you intended.

This was just an Example to make u understand my idea. I can take such points from shunt resistor to have current like 150A, 100A, 50A, 25A, 12.5A, 6A, 3A for charging batteries of 4Ah to 300+ Ah. You have to input battery capacity and circuit will select Amps less than Your input.

Going from 60 vdc to 27 vdc at any significant current is going to generate a tremendous amount of heat in the series pass device.

Yes you are right 60-27 = 33V Roughly will be across collector-Emitter and it will heat it up or destroy it if current is large. For that I have to make transformer with different output voltage so that my circuit will adjust its input so that Vce is less as possible. i.e. Transformer Voltage outputs = 12,18,24,28,32,38,42...RMS

Darlington series pass will likely give you too much gain and not allow enough current through TL783 to keep regulator stable

I'll doubt you here again:!:, its PNP not NPN so its current from emitter to base will contribute to regulator and will not take it away:wink:.Thats why i did not use single PNP because it will contribute high current to Regulator causing it to thermally shut down.Yes its gain is 1000 and but current through Q1 will be " (I = Vbe/R1) * Gain" where Vbe is Base-Emitter voltage and Vbe = I(Regulator) * R1. Transistor wont destabilize the Regulator because it is completely depending on Regulator current, it can not start with out this current and cant go above because this regulator current is limiting it. As load increases, current through R1 increases and Q1 outputs more current to stabilize Regulator voltage.Its a feedback. More i have applied this technique in many of my circuits and it works fine .

So any Idea about replacing the relays?

 

[Mr.Cool]

this is a bit of a curve ball.. but this problem has been solved entirely but using a different method. if you pulse voltage (actually just potential since there is no ground in the circuit) and keep the average current to a minimum (100 mA or less) you can charge the battery this way .. and since it is pulsed voltage then it does not matter how many batteries there are connected in parallel (with diode isolation). as an aside, the pulsed votlage method shakes off the crap that builds up on the plates during "normal" charge methods .. so the battery actually stays healthy. furthermore, there is no temperature rise in the battery (the major cause of aging) since the average current is ultra tiny.

something to think about

Mr.Cool

 

[nsaspook]

Pulse charging to restore lead-acid batteries is 99% snake-oil. With most non-traction batteries (SLA, car, most deep cycle types) even if you could remove the hard buildup the plates are so thin there's nothing left to recharge or the seperator sheets are warped and shorted. With forklift or golfcart types with much thicker plates that are designed for abuse it 's sometimes possible to restore a completely dead cell but almost never to full capacity.