High Intensity LED Driver with MOSFET

Thread Starter

chuvihi

Joined Jan 18, 2023
3
Hi,

First of all I would like to wish you a happy New Year!

I'm new with Analog design and I'm trying to figure out how to improve the behaviour or my LED driver. My design is powered using a single Lithium Cell. First I have a Step-up stage to boost the input voltage (2.5V-4.2V) up to 6V using the TPS61089RNRR.
1674050713995.png

Then, I have a Attiny85 controlling a MOSFET to drive two High Intensity LED's. Each LED draws almost 3A with a VF of 3.1V. I have two in series supplied from the Step-Up stage showed above.

1674050674066.png

The circuit works perfect but I had some issues that I think are related to resistors precision and temperature.
Firstly, I've asembled different prototypes but at the beggining I've found that the current draw from the LED's was to high for the Step-Up and this was switching on and off. Then, I reduced the output voltage from the step-up in order to reduce the current, and this worked great but after some time the issue is repeating again.
I've checked the voltage output from the Step-Up and it is rising up (starting at 6.04V, up to 6.16-6.20V) so the current is also increasing and, I think, this causes the switching regulator to touch the current limit.

To solve this Issue, I'm thinking to add an OP-AMP at the gate of the MOSFET to control the current no mather the supply voltage rises up.

This is what my limited knowledge and research achieved, but I don't know, what kind of OP-AMP use, use high side or low side current sensing and how to improve the circuit.

I've simulated the following circuit but, in order to reduce the power losses on the sensing resistor (R7), I ended up with very low voltage in the inverting input on the OP-AMP. Then I had to apply a voltage divider on the non-inverting input in order to be able to dim the LED brightness.
1674051549984.png
Do you think the approach I'm having is correct or there is better ways to do this?
I really appreciate if somebody can help me to solve this issue.

Sorry about the lentgh of the topic. Thank you in advance for reading.

Many thanks.

Regards,
 

Papabravo

Joined Feb 24, 2006
21,159
You're fighting against yourself with this project. Boost converters are able to do their thing, but at a cost. When you boost the voltage, you reduce the available current. What is worse is that the inductor is compelled to handle additional current without saturating. Since LEDs require relatively large amounts of current you are stressing everything in sight. If the LEDs require 3 A, then the Lithium cell will be drained quickly since this application requires more than 3 A from the battery to boost the voltage and produce 3 A to the load. This is a faulty concept, and you would be better off coming up with the appropriate series-parallel combination of batteries to meet your requirements without the need for a boost converter.

One more thing. A 741 is absolutely the wrong part to use for driving a MOSFET. There are many alternative solutional and nearly all of them are superior to that choice. what you want is a device that can quickly charge and discharge the capacitance of the MOSFET gate. The wimpy output stage of a 741 CANNOT do this.
 
Last edited:

LowQCab

Joined Nov 6, 2012
4,028
If You connect the LEDs in parallel, instead of series,
then You won't need to boost the Voltage.
Boosting the Voltage wastes a lot of Power.

You need a "Switch-Mode-Current-Limiting-Circuit"
to protect the LEDs from over-current, and to
provide a dimming feature if that's desirable.
.
.
.
 

Thread Starter

chuvihi

Joined Jan 18, 2023
3
You're fighting against yourself with this project. Boost converters are able to do their thing, but at a cost. When you boost the voltage, you reduce the available current. What is worse is that the inductor is compelled to handle additional current without saturating. Since LEDs require relatively large amounts of current you are stressing everything in sight. If the LEDs require 3 A, then the Lithium cell will be drained quickly since this application requires more than 3 A from the battery to boost the voltage and produce 3 A to the load. This is a faulty concept, and you would be better off coming up with the appropriate series-parallel combination of batteries to meet your requirements without the need for a boost converter.

One more thing. A 741 is absolutely the wrong part to use for driving a MOSFET. There are many alternative solutional and nearly all of them are superior to that choice. what you want is a device that can quickly charge and discharge the capacitance of the MOSFET gate. The wimpy output stage of a 741 CANNOT do this.
Many thanks for your message Papabravo. The battery that I have to use is 5000mAh 21700 1S2P and I cannot change the configuration, that's why I ended boosting it up.
Also another problem that I needed to solve is to have an stable voltage source for the LED to avoid changes in the light intensity while battery is discharging.

Running those LEDs without current regulation is a bad idea, heatsink or not they will draw more current as they heat up.
Many thanks ElectricSpidey. Yes, thermally is more or less controlled, and as you pointed I need current regulation, but I don't have a lot of experience with Analog design and I don't know what is the best way to control the current. What do you think is the best option to do so?

If You connect the LEDs in parallel, instead of series,
then You won't need to boost the Voltage.
Boosting the Voltage wastes a lot of Power.

You need a "Switch-Mode-Current-Limiting-Circuit"
to protect the LEDs from over-current, and to
provide a dimming feature if that's desirable.
Many thanks LowQCab for your message. As I said before, I'm using a Switching regulator as a way to stabilize the voltage during all the battery voltage range and keep the light constant whitout dimming.

Many thanks.

Regads,
 

Papabravo

Joined Feb 24, 2006
21,159
Many thanks for your message Papabravo. The battery that I have to use is 5000mAh 21700 1S2P and I cannot change the configuration, that's why I ended boosting it up.
Also another problem that I needed to solve is to have an stable voltage source for the LED to avoid changes in the light intensity while battery is discharging.


Many thanks ElectricSpidey. Yes, thermally is more or less controlled, and as you pointed I need current regulation, but I don't have a lot of experience with Analog design and I don't know what is the best way to control the current. What do you think is the best option to do so?


Many thanks LowQCab for your message. As I said before, I'm using a Switching regulator as a way to stabilize the voltage during all the battery voltage range and keep the light constant whitout dimming.

Many thanks.

Regads,
Sad to say that what you are doing will continue to be a losing battle. We can't always get everything we want. You should not feel too bad if you cannot come up with a solution.
 

Thread Starter

chuvihi

Joined Jan 18, 2023
3
Sad to say that what you are doing will continue to be a losing battle. We can't always get everything we want. You should not feel too bad if you cannot come up with a solution.
Sorry I cannot get your point and what solution are you proposing.
If I understood you correctly, you are saying that the problem is using a DCDC boost converter to supply the voltage to the LED string and instead of that, connect the cells in different configuration, for example in series to get a voltage between 5V and 8.4V.
And then, with the mosfet and the Op-Amp control the current through the LED string?
If my simulation is right, doing this, the MOSFET will drain almost 6W when the battery are fully charged (8.4V)

Is this right?

Many thanks,

Regards!
 

Papabravo

Joined Feb 24, 2006
21,159
Sorry I cannot get your point and what solution are you proposing.
If I understood you correctly, you are saying that the problem is using a DCDC boost converter to supply the voltage to the LED string and instead of that, connect the cells in different configuration, for example in series to get a voltage between 5V and 8.4V.
And then, with the mosfet and the Op-Amp control the current through the LED string?
If my simulation is right, doing this, the MOSFET will drain almost 6W when the battery are fully charged (8.4V)

Is this right?

Many thanks,

Regards!
So how does having the boost converter improve the situation? The LEDs will consume the power they require for a given brightness level. The boost converter does not help you with this problem at all. it actually makes it worse.
 

dendad

Joined Feb 20, 2016
4,452
How about replacing R11 with a 0.47R resistor and wiring the LEDS in place of R8, then you do not need the opamp or FET as the boost converter will run as a constant current drive?
I think it would be a good idea to try this with a suitable load and meter in place of the LEDS to start with!
 
Top