Designing a 60Ah lithium ion battery charger

Having a/c source of 220V,50Hz I intend to design a charger for 48V,60Ah li-ion battery.
I want to evaluate design cost with and without transformer.
Any inputs?

at what rate you want to charge it?...1C? (60 aMPS)

iF YOU ARE PREPARED FOR VERY SLOW CHarge then a cheap low power current regd isolated offline smps would be nice

As above, I suppose it depends on the required size, and if its just one for yourself, or for commercial mass production.

My own home brew standby mains charger (for a solar battery system) is just a transformer/rectifier/choke with SCR phase control on the primary.
Big, ugly, robust, and reliable.

Your storage energy is 3000 Wh and if you want to charge it in 1 h, you need a 3000 W charger !

So charger time is a big cost factor.

Production cost 10 cents/W in volume, prototype $1/W for ballpark.

It is a make/buy decision. If you have done this before many times, you can improve on cost, otherwise not, much worse, where cost includes your time.

To buy, cheap chargers run at $0.10-0.20/W smart chargers with load balancers $1/W and up so unless you have done this before, buy is a better choice.

If you want to be innovative DIY, use an AC motor/3ph>DC alternator for 48V and regulate the voltage with field current. These might be used commercial truck alternators
48V,180A alternator $295 used. https://www.ebay.ca/itm/John-Deere-...057596?hash=item43d1772e3c:g:KD4AAOSw3KFWde8Y

but I like warp's butt ugly , cheap reliable method.

If its for yourself, a suitable transformer, or even a whole clapped out ancient battery charger could probably be sourced on e-bay for parts and completely rebuilt.

An example of what you could get for only ten dollars:
**broken link removed**

Thank you all for reply,
I want to start in power electronics having built small rating chargers for mobile phone I want to make a DIY charger of the above mentioned rating and needed help design the ratings of bridge rectifier,filter circuit,chopper.
I will be charging it at 0.8C.

I have doubts like:
If the charger is required to be charged at 0.8C i.e 48A of current then charge time will be 1.25hours.
Calculations I have followed are:
Vdc=9220*1.4142)/pi=99.034V
Now as it's pulsating D/c hence filter circuit for good filtering would be : C-L-C or L-C how do I find ratings of these?
Idc =48A (0.8C)

I have a little black book of design details back from when I was designing commercial phased controlled battery chargers, mostly for lead acid applications.

Lithium would not be much different in the power stages, but obviously the control settings for precise required battery voltage would be different.

First requirement is that you need to cater for incoming mains variations that may be up to +/-15% of nominal mains voltage.
Plus 15% is easy, the regulator takes care of that.
But minus 15% needs a higher specified secondary voltage if its going to work.

For lead acid 48v system you might require 60v for occasional battery equalisation charging.
Not required for lithium, but I am giving the figures I have here only as a basis for your own calculations.

So the transformer would be nominal mains voltage (220 in your case) to 81v RMS secondary.
Actually only 68.9v RMS is required, but the extra allows for -15% incoming mains with a bit to spare.

The output of a choke input filter is average, not RMS, so we only get 62v average from 68.9v RMS. Allowing 2v drop for the bridge rectifier we get our required achievable 60v dc.
In retrospect 2v seems low, but that is what I have written in my book, and I know that the result actually works.

This voltage was reduced as required by phase control in the primary winding to whatever the required real charging voltage is set to.
But the power system was sized to reliably reach 60v at -15% incoming mains and at full rated load.
That may be excessive for your requirements ?

You can juggle these numbers, but its what I used to use for lead acid batteries.

For 48volt 50 Amp system (50Hz) the choke would be 3.072 mH at 80 Amps saturation, which would measure as about 3.74 mH with zero dc.
Output filter capacitor would be 20,000uF

Only single stage filtering was normally used, and the above will give less than 5% RMS ripple at full load.
If you need two stage of filtering, use another choke the same, and 10,000uF second capacitor.

That should at least give you a starting point.

Thanks,
I even thought of using half phase controlled rectifier (transformer less)
Where as per design equations I could figure out firing angle =81.125
**broken link removed**
The above circuit along with subsequent filter circuit when fired at 81.125deg give voltage approx=48V.
The procedure I followed was to evaluate V(peak)*sin(w*t) integral from @ to pi for positive half and @+pi to 2*pi for negative half cycle equating this whole integral to 48V(desired Vdc) .

Warpspeed said:

For 48volt 50 Amp system (50Hz) the choke would be 3.072 mH at 80 Amps saturation, which would measure as about 3.74 mH with zero dc.
Output filter capacitor would be 20,000uF

Click to expand...

How did you get those values?

If Half controlled rectifier or using a triac to obtain required d/c voltage is used then,
a proper firing circuit need to be designed but as will be using micro controller design of firing circuit can be simplified

Tony Stewart brings a good point, which I'm going to repeat:

Do you have experience designing these types of power circuits?

A 3 Kw supply will be very large and expensive. Dangerous too.

You will require custom transformers and chokes. The transformer I believe would weight on its own over 25 kilograms.
The DC choke it will be even larger; I have a 2.5 joule one (made by Hammond mfg.) which weighs 7 kilograms. You require 19.5 joule one which will be over 50 kilograms.
At the current cost of copper and magnetic grade steel, these two components alone will be well over $1200 US.

I've not even started with the filter capacitors nor the remaining electronic circuits.

So, while I applaud your desire to learn about power electronics, why don't you start with something much smaller? Say, 200 watts?
I hope that you are not like 99% of the newbies that post in these forum which ignore the suggestions from the grizzled veterans here.

scientist123 said:

How did you get those values?

Click to expand...

Its tied up with the requirement for <5% ripple which is a typical specification for commercial transformer/rectifier battery chargers.

Rule of thumb: L mH = 75,000 / Filter C in uF
This produces a natural frequency of around 19 Hz and a ripple factor reduction of about 30:1 at 100 Hz

Its not extremely critical, but if you have to design a whole range of battery chargers from a few hundred watts to tens of kilowatts using the same control board and PI control loop, it helps a lot to try and keep the ripple filter dynamics fairly similar across the range.

You will find many commercial designs end up around fairly similar values. its something that has evolved over decades of design and testing, and of course reverse engineering competitors products.

So its really a sort of consensus engineering based mostly on experience.

- - - Updated - - -

schmitt trigger said:

You will require custom transformers and chokes. The transformer I believe would weight on its own over 25 kilograms.
The DC choke it will be even larger; I have a 2.5 joule one (made by Hammond mfg.) which weighs 7 kilograms. You require 19.5 joule one which will be over 50 kilograms.
At the current cost of copper and magnetic grade steel, these two components alone will be well over $1200 US.

Click to expand...

That is exactly the situation.

Which is why I strongly recommend you seek out an existing suitably large fork lift battery charger being sold for scrap value, and rebuild it.
You will eventually pick up $1,000+ dollars worth of magnetics and a very robust steel enclosure for peanuts if you are patient.
These chargers mostly have a centre tapped secondary and two diodes.
If you fit a bridge rectifier it effectively doubles the output voltage. So do not pass up a suitably large 24v charger.

I wonder if MOT transformer can be rewound as a choke.

The problem with most transformers is they are dipped in transformer varnish and oven baked which glues all the laminations together.
It then becomes very difficult to dismantle without mangling all the laminations.

It must be fully dismantled because all the "E" and "I" laminations need to face the same way in a choke to produce an air gap.

I have winding specifications here for suitable chokes, but it would be far easier to start from scratch and purchase new laminations.

The above 3.74mH 50 amp choke requires 36 turns of 10mm x 2mm rectangular copper strip wound on a 4 inch stack of 2.25 inch laminations with a 1.5mm airgap.

And unless you have the right equipment and/or experience, putting the E and I laminations back into the transformer former is a *very* tricky job. I admit I've only done it once and I had all of the right equipment but it was a painful job and even then it 'hummed'!
Susan

Yes, its far easier with nice fresh straight clean laminations.

If its something you have pulled apart, the laminations will be bent and caked with blobs of varnish, and it will not go back together very neatly.

But doing it yourself will definitely be an interesting cost saving adventure.

I would still go for a scrap yard battery charger, its basically all there.
All you need is to scrape off the rust, hammer out the dents, and clean it up.

48V 60Amp-hour, so 16 Li-Ion cells, take to 4V max for charge = 64V max needed.

Charging at 10A requires only 480 watts approx, and gives a 6 hour recharge approx, much easier to buy a suitable transformer (& choke & all the other parts needed) if you want to go the 50/60 Hz way at this power level...

48V 60A Li-ion battery is made of several cells in series and parallel (I presume).

For optimal performance, cells must be charged individually and the charge state (how many Coulombs has been fed) must be recorded. Charging is perhaps possible in parallel but I doubt you will get good result by charging in series. Inequalities tend to increase with time.

The lithium ion batteries from autartech company come with a BMS where it charges individual cells in active balancing mode and a CAN interface is provided where SOC is been continuously communicated. So Now designing a charger for this capacity is what required.