I have built an Arduino controlled macro photographic stage setup and I use a Mosfet switch circuit (see attached) to switch two DC 10Watt LEDs on for 2 seconds that act a flash while a photo is taken. The system works well to photo insects.

The 10 Watt LEDs are powered via a buck converter set to 11V/0.9Amps. The system works rather well, measuring the volts/amps across the LED Lights terminals give zero when off and peaks at 10.2v/0,85amps during the 2 seconds of being on with LEDs connected as expected.

The system does not work that well for black or dark objects so I swapped the original LEDs for 2 50Watt LEDs to increase the light levels.

This works in a similar way powering the LEDs with a boost convertor set to 30V/0.6Amps. Measuring across the terminals during the 2 seconds peaks at 29.8V/0.6Amps However, when off there is 15.8V/0Amps across the terminals and a few of the individual LEDs on the 50Watt chip are just starting to light although not current is flowing as measured with Digital Multi-meter on mAmp setting.

I’ve checked through the mosfet threads but can find an explanation of a voltage drop across the source/drain when a Mosfet is off so can anyone explain why is there this voltage drop for the 50Watt LEDs. I would have thought is should also be zero as the 10Watt LEDs as the Id & Dsd are still in the safe operating area according to the IRLZ44N data-sheet.

Is the Mosfet leaking across source and base?
Should I be using a different Mosfet? If any suggestions?
Are the resistors I’m using wrong?


Thanks Daniel.

 

Welcome to AAC, @DrMac
What resistor do you have in series with pin 1 or 2 of the PC817?

 

The way you have it wired requires the opto-transistor to be turned on when the FET is off.
With a 2k load resistor and a 5V supply the collector current is 2.5mA maximum. According to the PC817 datasheet the transistor saturation voltage is (possibly much) more than 1V at such a low current in the opto-diode, so the FET could be turned partially on if the opto-diode isn't causing enough current through the transistor.
Try reducing the collector load to 1k.

 

You could try to disconnect the e of opto-transistor from the Ground and take the R5 away and in series the opto-transistor between R4 and Vg(pin 1) of IRLZ44N.

 

With a 2k load resistor and a 5V supply

Err... It has a 30V supply (±15V)

 

Err... It has a 30V supply (±15V)

Then R4 is too small, it could use 6.8K~10K.

 

Err... It has a 30V supply (±15V)

My bad. Your eyesight 's clearly better than mine .

 

Welcome to AAC, @DrMac
What resistor do you have in series with pin 1 or 2 of the PC817?

Thanks for all the resposes.

1st I have 220 ohms on leg 2 of the optocoupler so the arduino can send a signle to the opto couple and switch on the mosfet.

What you all seem to be indicating is to tune R4, I'll breadboard the curcuit, have a play and let you know the result

 

My bad. Your eyesight 's clearly better than mine .

Apologies the supply to the LED is 0 to 30 volts DC, should have cropped of the +-15V

 

The datasheet for that mosfet says the leakage (Idss page 2) 25 uA at 25 C and 250 uA at 150 C. It's very possible that
your LEDs can light a bit in these current ranges. You might put a 10K resistor across the LED terminals. This should
be enough to sink the MOSFET leakage current so that the LEDs don't light.