1
\$\begingroup\$

I am searching for a Li-polymer 1A charger with a 3V regulated output. I have found Linear LTC355x serie but can't, after reading many "LDO vs Switching" debates, understand the use of both on the same chip. LTC3553, for example, offers a 200mA buck and 150mA LDO.

I need to drop the USB 5V and battery 3.3-4.2V to 3.0V with a 170mA maximum draw. Space on the PCB is a premium so I'd prefer a combined chip. The PCB also contains 868mhz and 1575mhz antennas so a linear converter might be better ?

Should I choose between LDO or buck (I couldn't find a only-LDO or only-Buck chip) and disable the other ? Should the buck be the main source and the LDO the backup/standby source ? Is it only used to output different tensions, for example, 1.2V and 3.0V ?

EDIT : Thank you for your answers. They got me thinking about the VINLDO pin and I was wondering if feeding the LDO from the Buck to obtain a clean output is worth a try ? Assuming :

  • Battery safe tension range is 3.0-4.2V and VUSB is 5.0V.
  • All devices on the PCB could use 2.8V.
  • Buck is needed to obtain an efficient conversion from > 4.0V.
  • LDO is needed to obtain a clean output.
  • => The buck could output 3.0V to the LDO which in turn would produce a clean 2.8V ?
\$\endgroup\$
  • \$\begingroup\$ Linear LDO for logic, switching buck for power. \$\endgroup\$ – Ignacio Vazquez-Abrams Dec 10 '13 at 3:05
  • \$\begingroup\$ i didn't understood question properly. But, if 5V to 3V or 3.3-4.2V to 3V conversion at 170mA, Peak power dissipation =2*0.17 = 0.34W, LDO's can handle this easily. \$\endgroup\$ – user19579 Dec 10 '13 at 4:08
1
\$\begingroup\$

Looking at the application circuit, it seems that they have made on channel an LDO and another a Buck to save cost and component count.

enter image description here

  • Core need higher current and lower voltage than I/O. Buck works better for that than an LDO.
  • I/O needs somewhat higher voltage and less current (usually). LDO would be adequate.

Why not 2x bucks? Because a buck requires more components (inductor and better capacitor) than LDO. In this case, the combination of buck and LDO is a compromise of efficiency vs. cost and size.

It's not uncommon to use an LDO to power RF communication, analog front ends, because these can be susceptible to the EMI. At the same time, the digital section can be powered from a buck.

\$\endgroup\$
  • \$\begingroup\$ Thank you ! I edited the question to continue the discussion. \$\endgroup\$ – davew Dec 11 '13 at 16:15
0
\$\begingroup\$

Another point is that an LDO "simply works" when you start power, while buck regulators require an oscillator to start and their control loop is more complex.

So some applications that require standby power, but only turn the main power supply on when something's happening can make good use of this device.

As the standby power requirements may be very low, the LDO's lower efficiency is no problem. Indeed at low current they can be more efficient than the buck, which needs a certain level of power to run its oscillator etc.

One example application might be an USB-powered device, which is allowed a small current initially, but after negotiation, can draw more current and fire up the buck converter.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.