I've designed a switching controller using this chip for use with a large LED array. It's still in a prototype stage, so I'm using a potentiometer/voltage divider to adjust the voltage at Iadj to adjust the current.
Vin = 48 V
Vo min = 28 V
Vo max = 37.5 V
Iomax = 10.5 A
Here is the schematic.
Here is the top layer. the whole board is 90 mm x 70 mm
And the bottom layer. The only thing going on here is the ground plane for the power side, and the gate trace (nice and wide) for
I probed the pins for the IC. Here are the DC values.
Vin: 48.18 V
SS: 3.151 V
RT: 1.10 V
PWM: 7.2 V
COMP: 4.503 V
Iadj: 0.332 V (note that this depends on the position of the potentiometer)
Imon: 0.011 V
DDRV: 7.61 V
OVP: 0.577 V
IS: 0.006 V
GATE: it's kind of hard to give a DC value, as this is the switching pin
VCC: 7.6 V
I am operating this in a buck configuration.
All of those values seem about right, and the GATE pin is switching at what looks like a fairly high duty cycle.
Here is what I am observing: if I hook it up our LED array, it draws about 13 A before R8 inevitably burns, which I guess is a good thing, since that's an inexpensive and easy part to replace. This is independent of the position of the potentiometer.
If I hook it up to my eload on CR mode (constant resistance), Vout is constant, and Iout is proportional to Rload, again independent of the position of the potentiometer.
I separated analog and power grounds, connecting them through a jumper (R1). As a result, the analog signals are clean. There is virtually no noise on SS, RT, PWM, COMP, IADJ or IMON.
My question is why is Formula 31 from the datasheet not holding true. Rcs = 15 mO. If Viadj is 0.332 V, that should limit the current to 1.58 A Why is is driving 13 A? Even at when I bring Iadj down to 0, it still drives full current.
