Need help selecting a DC-DC LED driver

For an efficient step-down driver, that is likely the best one. Being a step-down, it will not give 3V out at the lower Vins, which means the LED will be less bright.

The TPS63060 may be an option.

Each of those devices has LED driving app notes.

Hi,  We looked at the  TPS63060 but found the efficiency is lower than the TPS62150.

As a quick test we drove a singe LED with the TPS62150 constant voltage mode. We have not yet modified the eval board but will do soon using the recommended circuit in the data sheet and the app notes. The output current was about 800ma. at 3.3V.

I know you do not drive an LED in constant voltage mode but it was just a quick test.

Question... Driving the LED in CV mode we noticed that the input current increased considerable as the battery voltage dropped from 12V down to 5V. Can we expect the some thing to occur when we mod the eval board for CC operation?

Thanks,

Paul

Yes, this is conservation of power. Since the output power is the same, the input current must go up as the input voltage goes down. Vin * Iin * eff = Vout * Iout

Hi Chris,

I cannot believe I asked this question, of course your answer is right. Too many long hours make you forget things.

There is a question I have regarding the TPS62150. In SLVA451 it shows using this part as a single LED driver on page 2.

I understand placing a resistor on the SS/TR pin lowers the FB pin to a lower reference voltage to reduce power dissipation across Rcs.

We intend to use a D/A on the SS/TR pin with the 154K resistor. What I am not clear on is fig. 3 on page 6. If there is no voltage on this pin the current through the LED is near zero. As the voltage increases on the SS/TR pin the LED current goes up. So at system startup some voltage must be present to turn on the LED. Is this correct?

Thanks,

Paul

It's not clear what your question is. But if you look closely, there is some offset on the SS/TR pin, such that 0V there is still some mV at FB and some current through the LED.

You can drive the pin with a DAC, but it will have to sink the 2.5uA current source to properly control the SS pin. Note that you can't apply a voltage on SS/TR which is greater than the voltage on Vin, in case your Vin is removed.

Hi Chris,

Referring to the schematic in the app note I did see that the graph showing the voltage applied to the SS/TR pin does show about 50ma or so of current when the pin is at 0V. That is no problem for our application.

The DAC will have no problem sinking 2.5uA. The system will be powered by a 2.0V regulator so the DAC voltage will never exceed this level. So if Vin goes to 0V so will the DAC supply voltage.

Also I noticed that the efficiency graph  for analog dimming for a Vin of 4V and 17V. We plan to operate on the low end at 3V, this will be the cutoff voltage for the system, will this cause any problems with linearity?

To clarify my original question I was asking that the DAC has to have some voltage above 0V to cause a few hundred milliamps of LED current when the system is turned on.

Thanks,

Paul

Usually, 3V is below the forward voltage of the LED. So, you will not be able to deliver the desired current.

No, you don't need a certain voltage on SS when the device turns on. You can power everything up and then use the DAC to get the appropriate output current.

Hi Chris,

We are using a LUXEON TX LED that does have a 2.5V min and a 3.0V max Vf. I was referring to the Vin of the switcher as being the shutdown point when the system turns off. We are using a micro to control all of this and monitor the battery voltage. Regarding my original question regarding the efficiency graph for analog dimming. It shows a Vin min of 4V. Is there any problem if we operate with a Vin to the switcher at 3.0V? Would you know if this will drastically alter the dimming linearity?

Thanks for all of your help.

Paul

The 4Vin is due to the LED we used and the drop across the IC, inductor, and sense resistor.

The IC can handle 3Vin. The linearity will change simply because with 3Vin, there is no way to put 3.0V across the LED due to the same drops in the circuit. So, you will get a lower LED current.

Hi Chris,

That is what I thought. The high 3.6V Vf of the LED would limit the low end Vin to the switcher. We will dim the LED when the battery voltage gets near 3V. The diode as I said we are using has a lower Vf so this should give us a little more margin.

Thanks,

Paul