Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

Small UPS with lithium batteries

Status
Not open for further replies.
Hi,

You could add a power path switch IC that has two outputs, one to charge the battery and one to power the load separately when the DC supply is available. Hunting around for Li-ion charger modules/development boards, etc. should come up with a ready-made board for your set-up that might fit what you have.
 
d123 said:
Hi,

You could add a power path switch IC that has two outputs, one to charge the battery and one to power the load separately when the DC supply is available. Hunting around for Li-ion charger modules/development boards, etc. should come up with a ready-made board for your set-up that might fit what you have.
Click to expand...

The power path switch IC seems an interesting concept (considering only what it suggests)...
Can you explain how it works exactly and how would I integrate it in the configuration?
 

KlausST said:
Hi

The problem here is that you never have a "true" charging mode, because the load alwasy draws current.
If your charger current is 2A into paralleled batteries, then the power is limited to 2A x 3.7V = 7.4W.
This 7.4W are used to power the load and to chargevthe batteries. If your load is more than 7.x W (as it seems to be) then the batteries never get charged.



A different approach:
I never use breadboards. I only rarely do simulations. I read the datasheets, do the mathematics. My first prototypes are on designed PCBs that are most equal to the end design. There may be only minor modifications ... and usually these prototypes go to the customers to verify the design.
With this prototype usually I even do the EMI/EMC (pre) tests.
Thus the end design has just minor modifications, low error rate from prototype to end design.
It takes more time "from idea to prototype" but less time "from prototype to end design".

Klaus
Click to expand...
Klaus brought up an important point that I forgot. Yes the batteries need to be charged by a higher current than the load drains, if there is no switching between the batteries and the power source. Normally this is the method used, because it also filters spikes and noise from the power supply, but the deficiency is the batteries die faster. I think the battery pack I posted with the charger should work for you still.

As for the PCBs I would like to ask Klaus, how many PCBs did he make until he reached to this stable situation ;D?

We are still overcomplicating things. I will make a circuit later in KiCad and give you the files.
--- Updated ---

According to your drawing, the battery is in the path of the charger, its not separated.

Do you like the circuit
1596581090472.png
 
Last edited:
Carry for cents bazar said:
Klaus brought up an important point that I forgot. Yes the batteries need to be charged by a higher current than the load drains, if there is no switching between the batteries and the power source. Normally this is the method used, because it also filters spikes and noise from the power supply, but the deficiency is the batteries die faster. I think the battery pack I posted with the charger should work for you still.

As for the PCBs I would like to ask Klaus, how many PCBs did he make until he reached to this stable situation ;D?

We are still overcomplicating things. I will make a circuit later in KiCad and give you the files.
--- Updated ---

According to your drawing, the battery is in the path of the charger, its not separated.

Do you like the circuit
View attachment 163298
Click to expand...

Well, it looks cool, but... that's chinese to me...
I remind you that I have no techincal knowledge of electronics, only hobby level empyrical knowledge.
 

Hi,

The circuit shows an automatic switch to the highest input voltage.

It has some problems..
Like:
1) it does not work when only the battery is connected. The Opamp is not powered then. It can't switch the battery to the load in this case. The load voltage still is zero, the Opamp voltage still is zero.
2) and the Opamp can't sense input voltages up to the supply rail. (CMIV range)

I'd start with two diodes, maybe schottky diodes.

But then new other problems arise:
* a charger is not a cinstant voktage supply. It may use charging techniques like constant current, maybe will stop charging when the output voltage (usually the battery voltage) reaches the "charging end voltage".
* How is the battery charged at all?

Carry for cents bazar said:
As for the PCBs I would like to ask Klaus, how many PCBs did he make until he reached to this stable situation ;D?
Click to expand...
Indeed just a handful. But I had good teachers. I've learned to work with datasheets and how to analyze circuits with engineers before I did my own designs.

Klaus
 

You need a BMS (battery management system) for the three Li-Ion series connected cells. The BMS decides the life of the three 18650 cells you are using. Try to use identical capacity cells. My personal opinion is that you can get better results with 4 LiFePO4 cells (they are safer and more rugged).
 

I am sorry Klaus and all. I don't like too long things. According to me you should use the circuit I gave you, the battery pack from my previous post with its own charger and a separate your own 24V charger. If you want to, you can improve my circuit, but I will not prolong this thread. I hate 50 page threads, you can't really find anything useful.

The idea is simple, 2 chargers, switch to the direct charger when the grid is on switch to the battery bank, when the grid is off. This way the batteries are charged automatically while the grid is on and you don't use them, unless necessary. This is why the pack with the charger is the best. The charger will do everything for you and you don't need BMS or anything else.

The downfall of this method is if the switching is not fast enough, your devices will power off and restart.

If you prefer the other method of the batteries always charging, just buy the pack I gave you and put as many connectors as you want on it.
 

Hi,

I'd go with c_mitra's suggestion of a BMS IC. Read up a bit about them yourself as there is nothing like learning.

You could use a load switch, a power path, and so on, but the BMS might be the simplest way.
 

Carry for cents bazar said:
If you prefer the other method of the batteries always charging, just buy the pack I gave you and put as many connectors as you want on it.
Click to expand...

The battery pack should be connected via a BMS (Battery Management System)- they are cheap. Some are very fancy.

Connect the charging voltage to the battery via a diode (the BMS has points to connect the charger). Connect the output of the battery to the DC-DC converter unit. If the load and the battery capacities are commensurate, there will be no problem.
 

Hi,

CC, the OP now has got several different ideas. Let him choose what he likes.
It may be your idea or any other idea, or a mix of ideas...

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top