Constant current source from a single LiPo cell at 5A

Hi,

I need help identifying a constant current source circuit to power a laser diode at about 5A from a single LiPo cell.
The Vf of the diode is about 1.9V.
I'm not interested in off-the-shelf stuff. I'd like to build it from scratch.
Is requesting help for specific chip selection something I can do or it is against the rules?

Thanks,
RI

This wont work, so a lower voltage dropout is needed/.

3 terminal regulator with Iout=1.25/Rsense

1V dropout. 1.25V current sense, 1.9V LED demands 4.15V in min

DEVICE OUTPUT CURRENT*
LT1083 7.5A
LT1084 5.0A


Similar to this


Since the power dissipation in the LAser diode will be high, the duration must be short, so a not only must it be current limited but it must be a fast pulse.

The simplest solution that does not require feedback but assumes a constant load cell voltage of 3.7V is to use a low RdsOn switch and a logic level N-ch FET with a series power resistor to limit the current. If you use a precision reference voltage then you can regulate the voltage to the gate with feedback for thermal compensation on repeated use.

Otherwise A 3.7V-1.9V =1.8V drop @5A= 360 mΩ for the total R including RdsOn of switch
https://www.digikey.com/product-detail/en/SIB422EDK-T1-GE3/SIB422EDK-T1-GE3CT-ND/2442103

I would rather have switching instead of linear.
Power dissipation is going to be a big deal. This is for a handheld unit.
5A on a .25 resistor is over 6W I'll have to dissipate on top of the diode.

This simulation illustrates how a buck converter would do what you want.

It is current controlled (via the sense resistor located below the led). The led parameters are similar to your laser diode.



Control IC's are commercially available. You might still need the external mosfet/ transistor.

You might consider interleaving two converters. It will reduce stress on all components, including the battery. The output will be much smoother.

So, if I understood correctly, the transistor would be taking the entire 5A current through it, right?
Isn't it a bit much?

What is the 0.17V and where is it coming from?

newhobby said:

So, if I understood correctly, the transistor would be taking the entire 5A current through it, right?
Isn't it a bit much?

Click to expand...

Yes, it can develop a few watts of heat. You must make sure you turn it on fully, so it is in saturation.

What is the 0.17V and where is it coming from?

Click to expand...

A reference voltage, applied to one input of the op amp (or comparator).

By adjusting it, you change the amount of current through the load.

Can this be applied to a FB pin?
How can I turn LTM4608A into a constant current source?

Sorry, I have no experience with control IC's.

LTM4608A is made by Linear Technology. They produce a free simulator called LTSpice. It supports many of there products. Several here recommend it.