Having a hard time with a SparkFun part

So I found this part: SparkFun LiPo Charger/Booster - 5V/1A - PRT-14411 - SparkFun Electronics

I want to use it to power the wearable computer, but I'm running into many documentation problems.

For one, the part says that it has no undervoltage protection, which is rather annoying. However, my batteries have this feature built-in (like most LiPos), though I have no idea what the cutoff voltage is. I need this measurement so I can accurately calibrate the voltage sensors so the cutoff voltage displays 0%, and the full charge voltage displays 100%.

The second problem is the switch on board. It says the maximum rated current is 600mA, but the load on the wearable varies depending on what the main computer (Pi 0 W) is doing. There's not much concrete information on what the Pi can reach in terms of power consumption, so it's hard to tell if an external switch is required. Should I just build a prototype and measure the current draw? Change the switch based on that?

Another issue is the ENABLE pin. What does it do? How is it any different from just flipping the switch off?

If anyone has answers to anything here, it would be much appreciated.

Hi
Sorry if I got it wrong, but I found this schematic and the datasheet of the ICs used in this module.
See if they help.

PAM2401.pdf (479.6 KB)
MCP738312.pdf (823.1 KB)
LiPoChargerBooster5V1A_v10.pdf (88.4 KB)

RV mineirin

1 Like

Lithium battery charge level in percentage is not easy to determine from only the voltage. Cell phones track charge and discharge events and use them to improve the accuracy, for example. The PI questions belong on a PI forum.

1 Like

@ruilviana I'm sure someone somewhere knows where the right information in those datasheets is, but I'm struggling to figure out how to use them to answer my questions.

@aarg I'm not going to implement something that complicated. As for the Pi thing, I might just have to benchmark it myself. The maximum rated current that a Pi 0 W pulls under heavy load seems to be ~300mA, but power stats on SBCs tend to vary.

EDIT - I think I phrased my intention sub-optimally. Basically, I want to tell the Arduino that the battery's cutoff voltage is 0% and the full charge is 100%. So when I see "1%" on the display, I know I should probably turn it off.

EDIT - I also think I know a way of measuring the cutoff voltage. Just plug in an LED, wait for the LED to turn off, and then measure the raw output from the battery. Not the protected output from the wires, the actual lines that go from the cell into the protection circuits.

The cutoff voltage can be determined experimentally. The functioning of the ENABLE pin is actually explained on the schematic you linked to.

1 Like

Looking now...

So it's just a redundant way of turning off the thing?

It's a way of turning off the thing...

...that leaks power. Because the power controller still stays on.

Whatever. Looks like the best way to turn this thing off is just to slap a beefy switch between the battery and power manager.

EDIT - So that solves issues 2 and 3, but trying to get the battery cutoff voltage is still a problem. There's always the LED method that I mentioned before, but I want an accurate load estimate. I think the only way to find this is to build a prototype, power it from a direct USB cable, measure the current doing different things, make an average, and then measure the voltage. So that's solved too.

Be sure. Most (99 %) of my lipo cells have no low voltage cutoff feature…

a7

And therin is a problem.

A lot of Lithium batteries do indeed have their own internal cutoff circuits, but often that cut off voltage is in the circa 2.4V region.

Now the battery internal cut is likely there as a safety measure in that its there to attempt to keep the battery safe when the normal expected device low voltage discharge cutoff is faulty.

Maybe allowing a Lithium battery to constantly discharge down to 2.4v is 'safe' but I would not wear one.

Neither would I. And there is the problem.

I COULD implement a low battery warning on the Pi's display by sending a key combination, but even then it would be up to me to turn it off before the cutoff.

There is no way to turn the wearable off automatically, because then the Arduino that's in charge of everything turns itself off, which takes the power away from the transistor, which turns the thing back on, which turns it off again, etc.

Because you didn't find or design a decent power management circuit.

a7

1 Like

I'm not an EE. I just cobble other people's designs together to make something cool.

If you have a recommendation for a small power management board with a voltage cutoff of 2.8-3V, let me know.

EDIT - It appears that the battery has a cutoff of 3V. So I'm probably fine.

A good way to test a battery for protection circuitry is to short it. If the wire melts, there's no protections circuitry. If nothing happens, you have protection.

Mine don't melt things, so they have protection circuits.

Thats an unusually high voltage internal battery cutoff voltage.

Provide us with a link, sounds a useful battery.

1 Like

One other point, do you know of any commercial device, that uses Lithium batteries that depends on the batteries own internal cutoff ?

The normal situation is that you want a device to shut itself down gracefully by the device itself detecting an impending low battery situation.

Relying on the batteries own internal cutoff can result in the device oscilating between on\off as the battery gets low.

1 Like

Search 1s lipo 600mah on Banggood. That's where I got it.

The cutoff voltage is probably because it's a drone battery, which use very high current.

I could theoretically tell the Pi to shut down at, say, 5% using a keyboard shortcut. Then again, in my (rather extended) experience using a Pi as a daily driver PC, they are totally fine being abruptly powered off, though it could get annoying if you have work to save, hence the requirement for a low battery warning. It's something I'll have to play with.

My combat robot does this as the voltage gets low. The motors start twitching and the power light flashes.

EDIT - I used a Pi 4 2GB as a daily driver for three months.

Searched. Turned up a crap ton of cells, many of which I know to be without any protection circuitry.

Please provide a link to the battery you actually have.

a7

1 Like

I bought them on a whim a few months ago. They were on sale (looks like they are again). My information comes from the following review and the graph the used attached:
image
image

EDIT - Despite the false advertising pointed out, the cells should be fine for this use case, as they shouldn't really exceed ~1.5C.

To summarize (this post to be marked as solution):

@srnet prompted me to find the battery cutoff voltage and use low-battery warnings and a dedicated shutdown rather than relying on said cutoff voltage.

I decided that the necessity of a seperate switch would have to be determined using a current draw experiment.

@ruilviana and @aarg dug up what the ENABLE pin does with some datasheets, finding that it's a way of turning off the output of the power manager but keeping the power manager on. Why this is useful, I have no idea.