SparkFun Forums

[ccool3]

I've been working on developing a power management circuit for my data logger project using a Single Cell Lithium-Ion Battery, a MCP73833 Li-Ion Charge Chip, and a DC-DC converter to regulate the power supply. I would like to create a regulated 3.3V using the battery as the voltage supply, but I've been having a hard time finding a buck/boost IC to do the job. My design criteria are: 1) an IC that can be prototyped (no no-lead packages), 2) regulated 3.3V output, and 3) at least 100mA of current output. Does anyone have any suggestions for a buck/boost chip that fits these criteria?

I have tried the MCP1253-33X50 (http://ww1.microchip.com/downloads/en/D ... 21752a.pdf), but I have not had success in getting the power specified in the datasheet. I mounted the chip on an etched circuit board, used capacitors specified, and tested the circuit provided in the datasheet (pg. 12 - Select to Vin for 3.3V). I am only getting 20mA instead of the expected 100-120mA. When I check the output with a scope, there is very little noise (~5-10mV) for the 3.3V response. If someone is familiar with this chip, I'd appreciate suggestions for trouble-shooting this problem.

Thank you for your time and input.
-Crystal

[mindspan]

I've been looking at similar parts lately, and have pretty much decided to go with one of Linear's integrated PMIC's. ltc3556, ltc3557, ltc3567 and ltc3586 all have integrated buck-boost, bucks, ldo's, a good power switching scheme and make good use of available input power when charging (mine will occasionally be plugged into USB).

Alas, they are only available in QFN, but i've had pretty good luck prototyping with QFN by using a paste syringe and hot air rework (and better luck with a stencil and hotplate). You can also try making the pads extend out from under the chip a bit so you can get the tip of an iron on them.

Aaron

[waltr]

I'm using an LTC3440 Buck-Boost converter from a single cell LiPo to power a PIC18F25J10 and a Kent display. This chip works very well.

[MrPotatoHead]

A single cell Li-Ion cell is 3.6-3.9V. Since the voltage curve is almost flat until full discharge, you could skip the regulator. Virtually every 3.6V IC will work perfectly fine at in that range. A 1000mA cell from sparkfun, http://www.sparkfun.com/commerce/produc ... cts_id=339, is a 0.2C battery. Average discharge current is 1000 * 0.2 = 200mA. 100mA requirement is solved. Once the cell voltage goes below 3V, your circuit will brown out (assumption on my part). That will provide a basic protection for cell under voltage (over discharge).

Or. You could use something similar to current project.

I need a 3.6V supply from a 2 cell Li-Ion pack (~7.2V nominal). I am using a TPS62200, http://focus.ti.com/lit/ds/symlink/tps62200.pdf, from TI. It is in a SOT23 package, so you can easily hand wire a prototype. I just solder leads to the package pins. The efficiency will likely be bad with such a bad layout, but who cares. There will probably be some ringing in the output, so put a big cap there to filter it out. In the final version, you can order a 3 dollar PCB from batch and those problems will be gone. For a 3.3V output, the input needs to be 3.6V or greater. You will lose a lot of your battery capacity, but it is simple solution.

Or. You could use this part to use the full battery voltage (2.8 - 3.9)

First, this one is a QFN package type. Sorry. I could not find another part that I am familiar with. A TPS63030, http://focus.ti.com/lit/ds/symlink/tps63030.pdf, is a buck/boost converter. Beyond being the wrong package, this one does everything that you want. TI might make an EVM that you could use as a daughter card. ... Again. Sorry. I know that is not really a good answer for you. I just threw it out there for the sake of discussion.

[48X24X48X]

Hi,

You can try using LTC3440 or LTC3530. I have use them in wireless sensor network application, and they worked fine. For 100 mA output current, you can find quite a number of them in the market. If your quiescent current is important, I can conclude that most of the buck-boost in the market are at least >= 30 uA in quiescent current rating. So, if you are looking for something like 1 uA, you might need to revert to LDO but you'll lose a bit of the last energy left from the Li-Ion for voltage lower than 3.3 V.

[MichaelN]

A single cell Li-Ion cell is 3.6-3.9V. Since the voltage curve is almost flat until full discharge, you could skip the regulator.

I agree; or you could use (say) a 3V LDO instead if you need a regulated supply. Most of the nominally 3.3V chips I've used are fine at 3V or less. You save a lot of components, board space and cost over trying to use a buck / boost arrangement.

[ccool3]

Thank you all for the helpful suggestions.

I tried the LTC3440 chip and had good success achieving the regulated 3.3V criteria. I still wanted to get the MCP1253 chip working, however, because I liked the smaller requirement for external components. I modified the layout on my etched circuit board to match the recommended layout in the datasheet, and that appears to have solved my problem. The chip is working great and is supplying the regulated 3.3V, 120mA . Apparently I just needed a little more patience in my circuit set-up.

Thanks again,
-Crystal