Power supply to LiPo plug instead of VBUS

boron
argon
xenon
Tags: #<Tag:0x00007fe21c76ce50> #<Tag:0x00007fe21c76cbf8> #<Tag:0x00007fe21c76c7c0>

#1

I know to externally power a Boron/Argon/Xenon device it is recommended to use VBUS. However, I was wondering if it was possible to power the device through the LiPo plug using a 5V PSU with the correct LiPo plug header. This is assuming we will never use USB in conjunction with the PSU. The reason for this is for ease of assembly.


#2

5V is within the Tech Spec:

image


#3

Speaking about the Boron specifically there may be cell over voltage protection enabled in the PMIC considering it expects a 3.7v LiPo battery there and 5v is a over voltage fault condition for the battery.

I would not supply 5v to the Li+ input on the Boron.


#4

I ran long term trials w/ Boron LTE with 4.95V on the Li-Po Connector. No problems at all.
However, that was from a Battery Source, not a 5V PSU.

But if easy assembly is the driving force, why not use a 5V PSU with the correct USB connector, ( like this ) ?


#5

Interesting.

Considering the PMIC handles charging of the battery they must not have enforced the battery over voltage protection feature that I know is built into the PMIC settings.

I still wouldn’t recommend it and instead would recommend he go with your USB Plug instead.


#6

the 5v to battery connector solution might cause over voltage. i’m skeptical of the max input voltage spec itself because the spec does not seem to have standardized measurement criteria. but to get past that why not use a adj. buck converter to dial the voltage to whatever? to me the more important issue is to have the psu source surge protected especially if in rather remote installations. i read many posts that indicate a device becoming nonfunctional for no apparent reason that was powered by a psu and very rarely if ever is a spike or surge considered a possible cause. device destruction can happen as easily as a very short power outage. i’ve seen it by just flipping breakers on/off quickly. using a buck converter at least gives you the chance of frying that instead of the device in case of some power disturbance.


#7

@All thanks for the replies! Thanks for pointing out the cell over voltage protection. Rereading the Boron datasheet it seems that 5V is sort of too much as they recommend the Li+ Pin input to be:

Remember that the input voltage range on this pin is 3.6 to 4.2 VDC.

@Rftop USB connector is a bit more complicated than just x2 wires. Moreover for what I am making I also do not want to add anymore “length” to the particle board. The USB connector is also unfortunately alone the length axis of the board. Hence I was wondering about the LiPo connector on the side.

I currently have a 5V 2A PSU soldered to the VBUS of a 2G/3G Boron unit. Aside from an assembly pain, I am also hesitant to debug the unit via USB while being powered by VBUS. Has anyone tried connecting to USB while the device is being powered by VBUS? I know that LiPo is fine since it has charging/discharging circuitry.


#8

Why? There is a VUSB pin (1 wire) and one of the GND pin (1 wire) - no more

I read people advising against using USB and VUSB at the same time but as long both sources are fairly stable and “exact” at their 5V level, I’ve had little issue so far.
However, adding additional diodes for back-feeding protection and/or a USB cable with a deactiveted Vcc line are an option for the “faint hearted” :wink:

How does your entire setup look like?
Are you using “flying” wires?


#9

Why? There is a VUSB pin (1 wire) and one of the GND pins (1 wire) - no more

@ScruffR I guess USB connector was a bad choice of words.

@ScruffR when Rftop mentioned this he linked a micro-usb powered wall charger on Amazon. So I assumed when he said “USB connector” he meant the micro USB male plug and was suggesting me to use the micro header so I can just plug/unplug the device and then just solder the Vcc/gnd lines to my PSU. This does solve the "assembly problem as I don’t have to solder directly to the particle pins. Cons is that this adds length to the board which I don’t want since it won’t fit in the enclosure I am using.

VBUS/VUSB pin is what I currently use now and is soldered to my PSU. That I agree is just 2 cables like you said :grin:

Meh, faint hearted… I guess I have a spare particle board, an RPi and a fire extinguisher. I’ll build the binaries on my laptop, cloud flash the device and check the serial logs via a RPi for debugging. If it blows up at least I won’t loose my laptop and will look into diodes and deactivating the Vcc line in the USB. Good suggestion!

I’m trying to fit the particle inside a standard 2 gang electrical box with some custom circuitry. This is a prototype for a company so I can’t give any more information than that. I do need to make quite a few units for testing/development purposes so I am trying to simplify assembly. If this works out we will be looking at perhaps laying out a custom board. For now, yes “flying” wires…


#10

Have you considered using a FeatherWing Proto board for the prototypes?


#11

Yes, I’ve looked at those. Both on the particle store as well as Adafruit site. The components on the custom circuitry will definitely not fit on a featherwing board. We would need a custom PCB with a slot for the particle… I also wished Particle sold us headerless Xenons/Borons/Argons but I know the reason why they do not.

But back to the main topic for powering the Particle via LiPo/VBUS. In summary…

  • LiPo needs 3.6-4.2VDC. 5V not recommended. This is a pity since most PSUs are within the 3.3V or 5V range. I’ll look at sourcing another PSU if need be.
  • Debugging via USB while VBUS/VUSB is powered should work if the 5V levels are stable. Else, diodes + disable VCC on USB while powered by VBUS.

I’ll keep you guys posted if this works or blows up but if you have any suggestions I’m all ears =)


Disable charging on Boron 2g/3g turns off cellular
#12

@RWB, The PMIC cant perform charging when the Battery Input is the only Power Source (this project).
It could also be disabled in Firmware. I agree, it’s not the best solution - but an answer to the OP.

@slacker89, I’m sorry for the confusion from my post. I didn’t know space was a major constraint. I meant simply use the 2A “Wall-Wart” straight to the Boron, as-is.
Can you use a CE Labeled 120VAC:5VDC that fits in a standard convenience receptacle box ? That leaves you the other half of the 2-gang box for your Boron + circuit, and has the chance of being Code Complaint and Safe.

Getting the Boron to consistently operate 1" away from the 120V transformer might be a larger hurdle than where to land/connect the 3V3/5V. I have no experience with that situation, but I’m sure others here have.


#13

I know, it’s just that 5v on this pin should trigger a PMIC shut down due to battery over voltage protection. I just would not want any protection features triggered unexpectedly now or in the future should Particle decide to enable this over voltage feature due to future changes or in response to something that happened to a unit out in the field.

It would be better to just feed this pin 4-4.2v to avoid all that if you have to use this input port.


#14

@Rftop no problem. I did not mention this in my initial post so no need to apologize :grin:

Yup I can, and this is what I have at the moment. I have a 110/120VAC to 5VDC connected to the Borons VUSB/VBUS. The connection is soldered via “flying” wires which is fine but just makes assembly a bit more delicate. Hence the orginal post of wondering if I could use the LiPo socket instead. I did a quick look around and I have yet to find a 120VAC to 3.6-4.2VDC PSU for this. I’ll do a bit more digging but typically PSUs are 3.3VDC or 5VDC, will let you guys know if I find one.

Note: The 5VDC PSU also outputs 2A so there is no need to rely on a LiPo for power even if you are using the 2G/3G Boron versions since they can spike up to 1800mA when transmitting data.

@Rftop Yup the PSU is CE. Note note that since I have custom circuitry inside, the entire device as a whole has to go through re-certification.

Just an update on this, tried this last night. I’m able to debug the board via serial and all while being powered by VUSB/VBUS. I did this through the RPi to be safe and all is well. Yay!

Hmmm if time permits and I’d like to try this for curiosity, maybe boot it in loops multiple time to see if it boots up fine and leave it on for a few hours etc… But yeah I don’t think I will use this for the final proto-type since it is not recommended by particle. I don’t know the long term side effects as well so its risky. Like @RWB said, if particle decides to check for over voltage in a new deviceOS, this will affect devices as well if they were to be updated.


#15

hello, i was tossing some ideas around in my head and wanted to check on some conclusions i’ve reached which might or might not be correct and this thread came up on a search and seemed somewhat related. conclusion 1 was that the Li+ pin can be both input and output. Li+ can be used to power the device but this is should be done only if there is no lipo attached at the same time or no usb attached at same time. i came to the conclusion because i noticed that a buck converter wired to Li+ for input when off and a battery attached the battery would backfeed ~3.6v to the buck converter, if the usb cable was attached with no lipo it also backfed voltage to the converter. conclusion 2 was that in the event that a battery was attached at the same time as input to Li+ serious [even fatal] damage could be done to the device via overheating. i came to this conclusion after i fried my boron somehow during my testing. another question is it is probably possible that an attached lipo might explode if Li+ is also energized at the same time? thanks for any comments.


#16

There are some differences between the Boron, which has the bq24195 PMIC, and the Argon and Xenon, which do not have a PMIC.

For the Boron, Li+ can be used as a power input.

  • You must never attach a LiPo battery to the JST connector and supply power to Li+ at the same time.

  • It’s best not to supply power by other means like the USB connector or VIN when powering by Li+.

  • If you disable charging in the PMIC, you could use USB for debugging with a normal USB cable.

  • The maximum voltage is 4.2V!

If you have a way to generate power in the range of 3.6V to 4.2V, Li+ is a good choice.

There’s a situation that can occur if you are powering by VIN or VUSB and do not have a LiPo battery attached. If high current use (like connecting to cellular) causes a voltage dip, the PMIC can believe it needs to boost the voltage output using the battery, but it can cause an even bigger dip when a battery is not attached before it realizes it and switches back to VIN only.

Also, in order to get the absolute lowest current use in SLEEP_MODE_DEEP you have to use Li+. Because the voltage is already in an acceptable range, one of the regulators in the bq24195 is turned off, allowing for lower power consumption in sleep.


Correct power supply for Boron 2G/3G without LiPo battery
#17

Ok, Rick, thanks for the information. growing up i never read directions and that has sorta morphed into never looking at schematics, until i need to try and fix something. which leads me to ask if there might be any usual suspects as far as fried components? i am gonna get a can of freeze spray this weekend and do the trick of the cellphone repair gods by looking for the hot spots as they melt the spray quickly. it’s hottest right around the battery connector/Li+ pin and the pins on that side with high temps in 170F-190F.sometimes the heat can be just a single capacitor but figured i might be facing something else. and another question is, the boron still worked. it was able to get online. at that point i figured maybe i had corrupted firmware so i tried the cli- cloud flash which worked for system-1 & tinker but on reboot would not the go online. it’s in some error state , blinks several green then blinks 1 magenta and repeats and i was wondering if there is some type of bootloader to system firmware mismatch that would keep the 1.2beta system 1 firmware, which was successfully flashed from working with the 1.1.0 bootloader, which was the build that was on the device? another problem is it is no longer recognized by windows at any point or state so usb or serial can not take place. that also leads me to think i fried one or both of the usb data lines. thanks for your time.


#18

You can also use simple alcohol (e.g. white spirit) and see where it evaporates first.


#19

ok, here in the u.s. i got 94% isopropyl alcolhol for clean-up so i’ll soak a q-tip and see how that goes. thanks for the tip.


#20

i was able to get the boron out of the several green blinks, one magenta, repeat cycle by increasing the voltage to 4.285v, technically still 4.2v. apparently the device was not getting enough volts or current to fully engage in cloud and/or safe mode. once it was online i was able to update the bootloader so it currently is at 1.2.0-b1. it runs fine at 4.25v now. have not measured current. that is through the Li+ pin. the battery also is able to run it ok. neither source heats up the board anymore. i guess that was the result of having multiple sources. the usb is dead. it gives no power to the device nor is it possible to use it for data after powered by other means. i reflowed the usb connector and that did not work. i’m guessing there might be a fuse but have yet to look. quite a bit of abuse and still mostly works. i’ll just use OTA flash. can’t remember but at some point is the bluetooth API going to allow for flashing firmware via bluetooth?