24V in, 5V out, n > 75%

I'm looking for a low cost DC to DC regulator IC that will be powered from 24V, with a fixed 5V output @ 80ma that has greater than 73% efficiency. Any suggestions?

I have used the MCP16321T (really like this $1.11 part) in the past and this is advertised as n=95%, however this is with; Vin = 6V, n drops to 62% when powered from +24V

I like the simplicity of this device, just add an inductor and a couple of caps, no external FET's required. I just can't live with 42% loss.


Max1750 is in the ball park, but its 2x cost of the MCP16321 https://www.digikey.ca/product-detail/en/MCP16321T-ADJE/NG/MCP16321T-ADJE/NGCT-ND/2835013
MAX15062, n=85%, $3.43
LTC3631, n=87% $4.33

1st create a free login to www.TI.com then open
https://webench.ti.com/webench5/pow...&O1I=0.08&O1V=5&op_TA=30&application=SWITCHER

The highest efficiency was 92-94% for TPS54061-Q1

I presume, you are only reporting the MCP16321T datasheet efficiency values and didn't try to modify the design for 24V/100mA operation.

Although the part is not necessarily optimal for low output current operation, the bad efficiency in the said operation point is mainly caused by a too low output output inductance. You should try with 50 or 100 µH.

FvM said:

I presume, you are only reporting the MCP16321T data sheet efficiency values and didn't try to modify the design for 24V/100mA operation.

Although the part is not necessarily optimal for low output current operation, the bad efficiency in the said operation point is mainly caused by a too low output output inductance. You should try with 50 or 100 µH.

Click to expand...

FvM,

Bear with me, power supply design is not a strong point for me, and this part is desirable based on cost. Can you explain why increasing the inductor value will improve efficiency?

I'm guessing it has little to do with losses in the inductor:

PL = Iout² * LESR

PL = 0.08² * 0.033 = 211uW

How would I determine efficiency with 50 or 100uH?

You actually have to avoid the inductor current from going into discontinuous conduction mode (DCM).

Unfortunately, this Microchip datasheet is low, pitifully low on design details.

Check any of the datasheets from TI, specifically those who were created by National prior to being acquired by TI, and there is a wealth of design information and step by step formulas. If one is not into the mood of actually taking paper, pencil and calculator to actually design the supply customized to one's requirements, they have powerful interactive simulators.

The late Bob Pease once wrote an editorial: "Why write application notes?" He made several arguments, but the most important was: "They help sell components".
Apparently Microchip has not learned this lesson for this particular component, which seems oddball considering some other very complete datasheets and app notes they have.

Check other vendors. STMicro is lower cost and they have good app notes.