Running RAK4631 without the RAK19007?

Yes, but it could do some things like sending an emergency message through an admin channel. Or it could lower the output power by 3 dB. Or it could ignore packets not coming from the admin node and so on. Also, shuting down the node at a certain undervoltage may save the life of the battery cell or increase its lifetime drastically.
Also it would be useful that the advanced user could set the voltage limits such as “100%” or “0%” because some will prefer LiFePo4 (100% = 3.6 V) over Li-Ion (100% = 4.2 V)…

However, meanwhile i temporary added the cable bridge and some Cs and run the module from a 3.3 V power supply:

The current measurement is disappointing. It is permanently just about 1 mA lower then together with the 19007. So the current peak obviously comes from the module itselfe, not from the blinking LED.
Good to know.
BTW even when the 4631 is shut down, it still consumes about 0.55 mA! So even then an empty battery becomes further discharged! And if the 4631 is shut down and connected to the 19007 it even continues to consume 2 mA. So, offline still consumes 50% of the online state.

To be honest, that’s not what i expected when i read "Ultra-Low Power Consumption 2.0 uA in sleep mode”
Could it be that someone translating from chinese to english does not know to differnciate between aA, fA, pA, nA, µA, mA, A, kA ?

Anyway, a TTGO Lora V2.1-1.6 consumes 11 mA, permanently. And the RAK4631 consumes just 4 mA. But i thought that the difference is much higher…

At least we see that the 4631 CAN indeed be run without the 19007 from 2.0 V to 3.6 V, favouring a LiFePo4 cell, while saving at least 1 mA or 24 mAh per day.

I see, external electronics are necessary!
Q.E.D.

When adding an efficient step-down converter, the RAK4631 can be run from something between 5…24 V.
It consumes just 2.2 mA at 13.1 V when idle and 27 mA when transmitting at 20 dBm.

nice. better attach a complete antenna :slight_smile:

Totally fine to build your own devices, just be aware that good hardware takes a lot of time and experience to do it right. Hardware issues created by alpha custom boards that have not been tested well are not the responsibility of the firmware developers to fix.

1 Like

Since firmware version 2.1.2(ish) i noticed that all my RAK4631, powered by this method, showed 0% and 0.00 Volt.
The solution is to add a voltage divider to Pin40 (P0.05/AIN3). Actually a voltage of 2.4 V at this pin is displayed as “94% 4.14V” on the phone app.
There must be a programmed divider somewhere in the source code, something like line 112 in power.cpp. But actually that factor is not 2. Here it is just 4.14/2.4 = 1.725.

I’m now running one of the remote nodes directly from a 46160 LiFePo4 cell, https://aliexpress.com/item/1005004222128348.html
It may be worth to correct the values for 0% and 100%…

1 Like