How to power with SAFT LS14500 or ER14500 directly?

[jayjay]

Hello guys,
anyone has experience using one of these batteries: SAFT LS14500 or the ER14500 or similar Li-SOCl2 with 3.6V? can i safely power the board by connecting it directly to the Vin and GND pins? or should I still go through the Battery input?

any recommendations please?

target product:HTCC-AB01 and HTCC-AC01 (cube capsule)

[sir-sundays]

hi jayjay,

I had the same question but did not get an answer:

AM01 - CubeCell Module - Best Battery?

Hi guys, hope you are all doing well. I have a question about the AM01. What is the best battery to power the AM01 Module? We thought about using an LS17500 but then we saw that the AM01 only supports a max voltage of 3.5v but the ls17500 has a voltage of 3.6v to 3.7v. Are there any batteries that can be attached to the module directly (ultra-low power design ) or is some kind of regulator needed in any case? – SirSundays

I hope to find one here. In our own prototype, we use a voltage converter right now. But as soon as we send something with LoRa the voltage drops anyways. So maybe it will work fine. Maybe we just gonna test it. The worst thing that can happen is that we lose one module.

–SirSundays

[jayjay]

Hello @sir-sundays,
i took the risk and plugged it, and here are the details of my findings:
For the Cube cell HTCC-AB01 board:

1. Battery (SAFT LS14500) measured 3.66V
2. Plugged in to VIN and GND, worked as expected and it drew about 9ua to 10ua during deep sleep (mode LoRaWAN.sleep() ) basically exactly as advertised by Heltec, good job guys !
   3.Next i Plugged it in to 3v3 and GND and it worked as expected but this time it only drew 2.5ua during deep sleep mode, exactly as advertised! Great

For the Cubecell HTCC-AC01 Capsule board:
Since it comes with the charger board attached i plugged the battery as shown in here and here are my findings:

1. one capsule worked but drew high current about 20ma during deep sleep (don’t know why) but i think something is wrong with the charging piece since i changed it with another and the capsule worked as advertised and only drew 9.5ua
2. another capsule worked without issues and drew 11.5 - 13.5ua during deep sleep.

left to test is plug the battery directly to the 3v3 pin on the capsule without the charging piece to see if the current draw will go down to 2.5ua as advertised…

So the verdict is: IT IS SAFE to plug 3.6v directly to the pins
cheers,
Jay

[jayjay]

left to test is plug the battery directly to the 3v3 pin on the capsule without the charging piece to see if the current draw will go down to 2.5ua as advertised…

i did this and it worked as advertised and only drew about 2.5 ua while plugged into a 3.6v battery

[kater_s]

I tried with an AB01 board and Nordic Semi’s Power Profiler Kit II module here. Using a regulated voltage of 3.3V, in deep sleep the board draws 9.5 mA (not µA!) over the VIN pin and 9.6 µA over the VDD (internal 3.3V) pin. VEXT ist correctly turned off. In my setup I use a SAFT 14500 LiSOCl₂ battery (3.7V), a Schottky diode with a forward voltage of about 0.2 V and the LiPo connector on the bottom of the board. This connector also draws 9.6 µA @3.3V or 9.9 µA @3.6V or 10.1 µA @3.7V, measured directly with the PPK II.

To prevent voltage dips during LoRaWAN Tx phase I have a supercap (1 F) in parallel at the LiPo connector. Once charged, this does not draw any extra current.