Driving high power led

[kuon]

I need to drive high power uv leds QLUV07M3QCV  which are 3.7V 700mA.

I have a 24V power source, so at first I went for the simple route and I put 5 of those in serie with 7.8Ohm 5W resistors. It was working find with my bench supply with 700mA current limit.
But the power supply I use on the machine is not precise enough and it did overshoot enough (to around 25V) for my test PCB to burn (I was wearing UV safety glasses and I noticed too late, my LEDs were burned and they cost a fortune).

I think I should drive them individually because their forward voltage varies about 10% which causes a significant error on a strip of 5 (of course, I realized all this after burning everything).

Anyway, I looked at the LEDs drivers, but there is not a lot of possibilities at 700mA current.

Do you have any recommendation?

[kuon]

Well, I found a few switching regulators, like this one https://www.digikey.ch/en/products/detail/diodes-incorporated/AL8862FF-7/14825822 which should do the trick with good efficiency, I can keep them in a strip of 5, with current sensing it should be properly regulated regardless of their forward voltage.

[mariush]

NCL30160 : https://www.digikey.com/en/products/detail/onsemi/NCL30160DR2G/3062085

Datasheet: https://www.onsemi.com/pdf/datasheet/ncl30160-d.pdf

Features
• Integrated 1.0A MOSFET
• VIN Range 6.3 V to 40 V
• Short LED Shutdown Protection
• Up to 1.4 MHz Switching Frequency
• No Control Loop Compensation Required
• Adjustable LED Current
• Single Pin Brightness and Enable/Disable Control Using PWM

See figure 10 on page 9 in datasheet... shows graph with 5 leds at 700mA, between 90 and 95 efficient with 24v input voltage.

AL8661 : https://www.digikey.com/en/products/detail/diodes-incorporated/AL8861QMP-13/10130663

Datasheet: https://www.diodes.com/assets/Datasheets/AL8861Q.pdf

 Wide Input Voltage Range: 4.5V to 40V
 Output Current up to 1.5A
 Internal 40V NDMOS Switch
 Typical 5% Output Current Accuracy
 Single Pin for On/Off and Brightness Control by DC Voltage or PWM Signal
 High Efficiency (>95%)
 LED Short-Circuit Protection
 Inherent Open-Circuit LED Protection
 Sense Resistor Short-Circuit Protection
 Over Temperature Shutdown
 Up to 1MHz Switching Frequency

AL8843 - https://www.digikey.com/en/products/detail/diodes-incorporated/AL8843QSP-13/10668306
Datasheet: https://www.diodes.com/assets/Datasheets/AL8843Q.pdf

Wide Input Voltage Range: 4.5V to 40V
• Output Current up to 3A
• Internal 40V NDMOS Switch
• Typical 4% Output Current Accuracy
• Single Pin for On/Off and Brightness Control by DC Voltage or
PWM Signal
• Recommended Analog Dimming Range: 10% to 100%
• Soft-Start
• High Efficiency (Up to 97%)

IS32LT3954 (3A driver) - https://www.digikey.com/en/products/detail/lumissil-microsystems/IS32LT3954-GRLA3-TR/14308387
Datasheet: https://www.lumissil.com/assets/pdf/core/IS32LT3954_DS.pdf
IS32LT3952 (the 1.5A version of above) - https://www.digikey.com/en/products/detail/lumissil-microsystems/IS32LT3952-GRLA3-TR/12675539


PAM2863 - https://www.digikey.com/en/products/detail/diodes-incorporated/PAM2863ECR/4033265

ALT8080 - https://www.digikey.com/en/products/detail/allegro-microsystems/ALT80802KEJJTR/9655954

etc etc

[kuon]

Thanks.

Actually I found a few now. I think I had a bad filter on digikey first when I first searched, I couldn't find anything about 500mA.

[Benta]

This is the simplest and most reliable way (but it will dissipate a couple of watts on top of the LED power consumption):

[Zero999]

The LM393, an inductor, a voltage reference, a MOSFET, a Schottky diode and a few passive components can be used to make a buck LED driver.

Vellemen make a kit with can be modified to work off 24V.
https://www.velleman.eu/products/view/?id=366712&country=us&lang=en
https://www.velleman.eu/downloads/0/illustrated/illustrated_assembly_manual_k8071_rev1.pdf

If you only need to run if off DC, then the bridge rectifier isn't needed, so I've not shown it on my schematic.

[kuon]

Thanks a lot for your great suggestions. In the end I went for the IS31LT3360 to reduce component counts, but having the actual discrete schematic is very educational, thanks.

[thm_w]

I need to drive high power uv leds QLUV07M3QCV  which are 3.7V 700mA.

I have a 24V power source, so at first I went for the simple route and I put 5 of those in serie with 7.8Ohm 5W resistors. It was working find with my bench supply with 700mA current limit.
But the power supply I use on the machine is not precise enough and it did overshoot enough (to around 25V) for my test PCB to burn (I was wearing UV safety glasses and I noticed too late, my LEDs were burned and they cost a fortune).

I think I should drive them individually because their forward voltage varies about 10% which causes a significant error on a strip of 5 (of course, I realized all this after burning everything).

This doesn't make sense.
If you have a power supply with a 700mA current limit, that is half decent, it will limit the current to 700mA. Does not matter if you set the voltage to 25V, does not matter what the LED forward voltage variation is.
Even if you set it as 25V, with no current limit, the 7.8 ohm series resistor should limit current to about 830mA.

Something else went wrong here.
Did you let it sit for a long time and overheat without adequate heatsinking?

[kuon]

I need to drive high power uv leds QLUV07M3QCV  which are 3.7V 700mA.

I have a 24V power source, so at first I went for the simple route and I put 5 of those in serie with 7.8Ohm 5W resistors. It was working find with my bench supply with 700mA current limit.
But the power supply I use on the machine is not precise enough and it did overshoot enough (to around 25V) for my test PCB to burn (I was wearing UV safety glasses and I noticed too late, my LEDs were burned and they cost a fortune).

I think I should drive them individually because their forward voltage varies about 10% which causes a significant error on a strip of 5 (of course, I realized all this after burning everything).

This doesn't make sense.
If you have a power supply with a 700mA current limit, that is half decent, it will limit the current to 700mA. Does not matter if you set the voltage to 25V, does not matter what the LED forward voltage variation is.
Even if you set it as 25V, with no current limit, the 7.8 ohm series resistor should limit current to about 830mA.

Something else went wrong here.
Did you let it sit for a long time and overheat without adequate heatsinking?

Yes the LED did not burn because of current, they burned because I put a 3W resistor because that's what I had and didn't take the time to size it (I'm an idiot), the resistor was on the other side and I didn't notice it become red, the heat conducted into the PCB and the LED died before the resistor.

[thm_w]

The resistors is dissipating ~5W, the LEDs are dissipating ~10W.
10W is a lot of power for a PCB, from the LEDs alone, even ignoring the resistor. Is it an aluminum PCB?

[kuon]

The resistors is dissipating ~5W, the LEDs are dissipating ~10W.
10W is a lot of power for a PCB, from the LEDs alone, even ignoring the resistor. Is it an aluminum PCB?

Yes it is quite a lot of power, and I did underestimate the dissipated power in my first design. I did proper calculations for my second design. Now the PCB holding the LEDs has no other components except the LEDs and a connector, I put plenty of VIAs under and around the LEDs and a big heatsink on the back side of the PCB with a fan. I had this setup running for a day, and the PCB is stable at around 55°. I guess an aluminum PCB would have been better but it is quite expensive for my hobby project.