Heltec V3, USB battery bank, and solar panel

I’ve got a Heltec V3, up on my roof, hooked up to a USB battery bank. Everything is hunky dory. Every week I bring the USB battery bank inside and charge it up.

I thought about hooking up a solar panel to the USB battery bank to keep it topped up while it’s powering the Heltec. That’s fair enough. The solar panel does charge the battery bank (with no Heltec). Lovely.

But, when I hook the Heltec to the battery bank (with the solar panel connected to the battery bank) the Heltec causes the battery bank to keep flicking off and on thus rebooting the Heltec.

If I take the Heltec off the battery bank the battery bank carries on charging fine. No restarts. Plug the Heltec into the battery bank and it and the battery bank starts flicking off and on.

This has happened with two different panels and two different battery banks.

Any ideas??? I’m stumped.

1 Like

Connect the node via tiny battery plug directly from the liion battery and charge the power bank from 10w solar panel to 5v usb connector.

1 Like

Double (maybe three times) -click the button on the power bank

Being able to work as a UPS isn’t a feature that most USB battery banks have. (Most cannot charge and output at the same time)

1 Like

Well, now you have me wondering what a double/triple tap of the button might do…

1 Like

Double = send position,
tripple = toggle gps on/off

I have my Heltec V 3 up high connected to 3000Mah battery and ran the usb cable down to a usb port. That way don’t have do any manual recharging.

Plug the Heltec into the battery bank and it and the battery bank starts flicking off and on.

I guess you are experiencing a peculiarity which exists in many, but not all, USB powerbanks. I have this on-off behavior with recent versions of INUI B5 powerbanks, whereas my Anker 335 Powerbank (PowerCore 20K) does not exhibit this behavior. The Anker powerbank provides steady power to my low-power devices.

Just yesterday I came across an explanation for this on-off behavior: some power management IC models have an auto-off “feature” to cut off power to connected devices when those devices draw too little current; e.g. for the IP5306 powerbank IC this threshold is supposedly below 50 mA.

GreatScott! has a tutorial how to hack-fix this problem.

I asked Anker about the 335 Power Bank.

My question:

" Dear Sirs,

Does the Anker 335 Power Bank (PowerCore 20K) switch off automatically
when current drawn is low? e.g. 70 mA, 150 mA or similar ?

I am looking for a power bank that just stays on all the time even if
the current drawn is low.

I saw someone report that this bank remains on even if the current
supplied is low. Is this correct?

Thanks for any help,

Yours faithfully,

"

Their answer:

Hilarious. “We only sell them.”

I got one off Amazon, used, 28 euros… so will test if it stays on all the time.

Thanks Abdull for the info. I found that very useful. I purchased a Anker 335 from Amazon for 30 euros (it was listed as “used” but was only slightly scratched and electrically A1). I tested the device because you said it powers your low power devices fine. I was looking for a power bank that just keeps going and going when running e.g. the RAK node. They mostly all switch off when the current drawn is as low as that drawn by the RAK or similar systems. So I used a so-called “keep-alive” circuit to see when the Anker 335 will stay on for ever. The parts I used are these:
USB current and voltage meter https://amzn.eu/d/efswIkj
keep alive load USB Battery Pack Keep-Alive Load
In the test the “trickle charge” mode the Anker has available (allowing low current but only for 2 hours) was off.
I found that, in order that the power bank does not turn off when the current meter is removed and only the keep-alive circuit is used, I had to set the potentiometer on the keep-alive circuit (when current meter was in) so that the meter read 75 to 80 mA in the steady state (the number displayed fluctuates slightly).

I do not think the Anker cut-off limit is as low as 80 mA because some power was drawn by the meter. But this is a guide to how to set it and to see what sort of duration approximately one might expect.

Therefore the Anker 335 might power the RAK system with the keep-alive circuit set as above for about 11 days. 20000 mAh / 80 mA / 24 h = approx 10 days , in theory. I have not tested this in practice yet.

The current reading from the RAK when there was no keep-alive circuit was 15 mA. It can run for about 2 weeks in practice and 17 days in theory (10000 mAh / 15 mA / 24 h) on a 10000 mA LiPo battery directly.

So using a battery pack with the sort of keep-alive as above is not better than using a smaller LiPo directly or smaller LiIon directly in my view. Hope what I said make sense.

I will be testing other keep alive circuits

and

to see if the above holds then too.

1 Like

I would not use a Heltec board for a solar charged node for 2 reasons : its relatively high energy consumption versus the consumption king that the RAK wireless boards are; and secondly because it does not provide any input to be able to directly connect a LiPo/LiIon battery cell to it. That is why instead of trying to spent lots of time and money on finding and/or hacking a USB power bank that keeps on permanently, you’d be better off buying a RAK wireless board that has a direct input to connect a raw lithium battery cell (or mutliple of these cells in parallel). I have a RAK node that is powered like this with 4 cells in parallel, directly plugged to the board and it lasts a few months without recharging ! With the addition of a solar panel, you make your node permanently autonomous without ever needing to recharge. I have published DIY build videos for this on my youtube channel here : https://youtu.be/scx2zJT-Ufw?si=p0ap3l4eLwHFPHmP

The capacity of each of the four cells is 3000 mAh in video. I get about 2 weeks of RAK off 3 18650 3500 mAh batteries when using the waveshare D solar power manager to provide usb c input power to RAK. Many months off 4 seems very good. Batteries should be balanced (all about same voltage) before spot-welding, right?

About 2 months, with very little usage

Yes, that’s better to avoid eventual large current rush from the most charged cell to the lowest charged one. Since they are soldered in parallel, balance will be automatically maintained afterwards