texas instruments buck boost

[EightSixZero]

Has anyone used Texas Instruments tps63020 buck boost for making one batt vari volt mods? I`ve made several based on their ptn04050c boost chip and nuck's fistpack wiring diagram, both the slide switch style and the up/down switches with a digital pot and even one with a thumbwheel but they have trouble going below five volts. Even with a diode on the vout line i cant get below 4.7 and if i add a resistor between adj and vout on the chip i can get down to 4.2 but then the high end is like 4.8 or less under load. dont get me wrong the 4050c chips arent bad but since they`re boost only i think the buck boost will be better suited and give a larger useable voltage range. Not to mention its 4 amps which is great for the dual coil tanks i exclusively use these days. The chips will be here Monday so any experience/tips anyone can share would be great

 

[meatsneakers]

While the voltage range should be larger, the rated amps are closer to 2.3a on a 3.7v - 4.2v li ion for our usual vaping voltages:

 

[Java_Az]

the 4050c is a module almost ready to go just add caps and a resistor to set the voltage. The tps63020 is just a chip your going to have to add quite a bit of stuff to make it work 3 resistors 6 caps and 1 Inductor. It also comes in a 14pin QFN power pad package. Which is pretty small and tough to work with without experience. The chip has pads on the bottom so you pretty much need a rework station to solder one onto a PCB. Some folks have used toaster ovens with success also. They are so small you cant just solder wires to them so designing and making a PCB for the circuit is pretty much your only option. To give you an ideal the chip is 3/16 of a inch by 1/8 of a inch with 14 solder pads in the little area.

EDIT: Adding this picture of a QFN type chip i did last year to give you an idea of how small these chips are next to a 10 cent piece.

 

[EightSixZero]

I've worked with some of the tssop digital pots with success using a magnifying glass to
make sure my connections weren't bridged so I understand the smaller packaging being more of a challenge, as well as having to build the entire circuit instead of the essentially plug and play nature of the 04050c. The added complicity is what I'm looking for in order to keep making these fun instead of the tediousness of doing the same circuit over and over. if anyone has a wiring diagram using one of these (the 63020) please share so I have a good idea what to expect. thank you all for your input.

 

[CapeCAD]

I've worked with some of the tssop digital pots with success using a magnifying glass to
make sure my connections weren't bridged so I understand the smaller packaging being more of a challenge, as well as having to build the entire circuit instead of the essentially plug and play nature of the 04050c. The added complicity is what I'm looking for in order to keep making these fun instead of the tediousness of doing the same circuit over and over. if anyone has a wiring diagram using one of these (the 63020) please share so I have a good idea what to expect. thank you all for your input.

Serch TI for TPS63020EVM-487, this is the evaluation board for this device and they include a schematic and layout you can work with.

However, also look at the chart in the documentation Efficiency Over Li-Ion Cell Range (Figure 4) and see if this meets your expectations.

 

[EightSixZero]

capecad- thanks for the input, i appreciate it. the efficiency for the load range and voltage I expect to use is in the mid 80-90% range (3.8-5.5v 2-2.5a). I'm going to look into how this stacks up to the 050c to be sure but that doesn't sound all that terrible to me. the 050c gives me one or two days of heavy vaping off one aw18650 so if its in that ballpark im sure I'll be pleased. I intend on building one for my own testing purposes, and if its any good I'll use that setup in place of the 050c for all future builds.

 

[CraigHB]

The thermal pad underneath the chip is critical for heat sinking. You will not get the rated output without a good thermal connection to that pad.

I do all my PCB soldering by hand. What I do for thermal pads is put a couple large(ish) vias under the chip so I can solder the pad from behind. The vias must connect to a ground plane to facilitate the required heat sinking. In order to solder the vias to the thermal pad, they need to be plated which is not something you can do with a home-grown PCB.

As you can see from the photo posted earlier, the pins are quite small and connections need to be made with traces/pads that are only .25mm wide. I use a 10x stereo microscope. Unless you have very good vision, I don't really see how it would be possible to hand solder a QFN package with any less magnification than that.

The other option that would not require as much magnification would be so use an oven or hot plate to reflow solder the chip. I have no experience with reflow so I don't know how hard or easy that is to do. I also don't know how reliable that method would be. You still need to use plated vias to connect the thermal pad to a ground plane, but you would use a bunch of small vias instead of a couple big ones.