[SOLVED] I need some help understanding LED constant current SMPS

[monteie]

I got a hold of a Philips LED light bulb, and I decided to play around with it to further my understanding of a constant current SMPS. As a test, I split the LED ground wire so I could measure the current being sunk into the LED string, 0.6 A in my case. As another test, I decided to leave the LEDs disconnected from ground, and instead placed a few power resistors between the supply and ground; I choose a resistance that would produce 0.6 A at the same output voltage seen by the LEDs under normal operation. Therefore, I expected to measure 0.6 A through the resistor, but I didn't, instead I measured 0.2 A.

From my understanding, if the current sense technique is working properly, the device should attempt to adjust its output voltage until the load sink 0.6 A. In my case, the output voltage of the driver dropped to hit the 0.2 A mark. If an r_sense was used, this doesn't make any sense to me, so perhaps there are other sense techniques that care if the load is a resistor or a diode? Can anyone explain to me why the constant current supply is not sinking 0.6 A into the resistor?

Thank you, and best regards

 

[iffee100]

I got a hold of a Philips LED light bulb, and I decided to play around with it to further my understanding of a constant current SMPS. As a test, I split the LED ground wire so I could measure the current being sunk into the LED string, 0.6 A in my case. As another test, I decided to leave the LEDs disconnected from ground, and instead placed a few power resistors between the supply and ground; I choose a resistance that would produce 0.6 A at the same output voltage seen by the LEDs under normal operation. Therefore, I expected to measure 0.6 A through the resistor, but I didn't, instead I measured 0.2 A.

From my understanding, if the current sense technique is working properly, the device should attempt to adjust its output voltage until the load sink 0.6 A. In my case, the output voltage of the driver dropped to hit the 0.2 A mark. If an r_sense was used, this doesn't make any sense to me, so perhaps there are other sense techniques that care if the load is a resistor or a diode? Can anyone explain to me why the constant current supply is not sinking 0.6 A into the resistor?

Thank you, and best regards

A schematic of the SMPS is required to explain it.

 

[chuckey]

Some (most lighting) SMPS. use a saturable coil to monitor the current and set the switching PWM and need a minimum current before they work properly. They typically work over a 5:1 current range. As this SMPS is designed for just one load its range may be smaller +- 10%?
Frank

 

[monteie]

@chuckey This was indeed the issue. My confusion came from assuming secondary side sensing when in fact most lighting applications use primary side sensing, which is extremely prone to loading.

To aid others who find this thread, attached is a useful application note on Primary Side Sensing Flyback converters.