Power Statistics

What about a Diesel aggregate / generator? :smiley:

Also I personally would not plug my L5 into a energy source until I would be sure that it does not takes any damage.

Somewhere in this forum the advice was given to only use the original power plug and cable for charging because of the supported charging protocol. I hope that will change in the future so that we can use common chargers and cables. Unfortunately the plethora of USB chargers and cables seems chaotic and until now we can’t see from the outside what is supported and there will be (are?) cables supporting higher voltages for charging laptops or so. Not all devices can handle higher voltages. I guess the charging parameters must be negotiated with care. I would not be sure that all cheap cables and chargers do that correctly.

@JR-Fi : Damn ! you killed me with that !! :rofl: :rofl:
@amarok : You can still plug the Librem5 to a tesla ! 100Kw battery for your phone ! :smiley:

To get back to the subject :wink:
Can someone with the Librem5 can share how the graph is when attached to a regular powerbank ? (in use or in idle baseband On)

1 Like

@amarok, if you are interested in possibly having a more readable output and you are willing to enable ssh on your phone, you could try X11 forwarding to run your application on your phone but display it on another computer. That way you could use the remote computer to control the application window size.

For security’s sake, I suggest disabling the ssh server on your phone after you have run your tests though, especially if you will be carrying it in public and doubly especially if your modem is on.

1 Like

Those are all .png, except for the last one.

I also have a small solar charger (with built-in battery storage) that I keep in my car in case I ever need power for my Android. It’s only slightly larger than the L5. :slightly_smiling_face:

@Purism, Feature request: solar panel backplate FTW! Lol!


I think what you saw is this (or in this discussion) :

Which means no problem to use a regular charging system (5v / 2A), it will just take longer to charge
As I don’t know about the USB-PD protocol, but I assume it probably negotiate the power delivered so it can deliver more than the regular power of powerbanks or small chargers to be adapted to the destination device

1 Like

I referred to that thread but two posts before or so.

Meanwhile the thread has progressed and I encourage to read it to the and. I don’t want to say anything wrong.

The other part I wrote about the chaos in USB specifications and higher voltages is what I remember out of an article in German computer magazine C’t some time ago. Not even they were able to keep in the know of every detail. They wrote an analogously statement explicitly. I think higher voltage chargers are not very common, yet, but this won’t stay forever. Maybe they will be marked somehow. If negotiation go well and they don’t give to much power immediately everything might be fine. I am not a specialist. The fear or danger of those new cables was written in that magazine article.

1 Like

They are marked - usually. If they have a fancy-sounding name, they do most likely higher voltage, and sometimes higher current. Both can be risky. The L5 will only negotiate higher voltage with charges capable of Power Delivery, and the high voltage mode is tested best with the supplied charger.

Regular 5V chargers (USB Battery Charging) should be safe as long as they are not doing something crazy like providing the wrong voltage.


A case in point being the chargers for the Planet Computers devices. The charger for my Gemini isn’t marked with a fancy name, but has the following:

Output: 5V==2A/7V==1.67A/9V==1.67A/12V==1.25A

IIRC it uses MediaTek Pump Express.

When my L5 arrives, I guess that I won’t be trying this charger out on it :wink:. The Gemini might be OK with Purism’s charger though. The Pump Express web page has this:

The technology is compatible with the international standard of ‘USB PD 3.0 programmable power supplies’, allowing standard USB PD 3.0 fast chargers to boost Pump Express 4.0-enabled smartphones.

I mean - it is going to work, since the L5 is using a “compatible technology” too. Using the supplied one is a recommendation to those who want to be extra safe.


I’ve ordered SMARTCOBY Pro (with two LG-INR21700 battery cells) anyway.

I have been using the Baseus Power Bank 65W Baseus with the Baseus PD USB-C cables with the Librem 5 and never had any issues over the last year. I use the power bank in my backpack with the L5 when going on long rides. That was before the L5 supported standby. Now I hardly need it.

Nice feature is it shows how many amps are drawn during charging or using and charging of the phone.

Updated Power Statistics when compared to 2021 here are 2023 numbers.
Environment: wifi hks off, wwan 4G on, open vpn connected, various apps running (mostly .deb apps but flatpak is installed on phone), screen brightness 85%, used indoors, on battery starting with 100% charge to 0%, suspend on at 5min, blank screen at 2min, sparklan modem.

Steps to repeat during entire runtime:

  1. occasionally pick up phone turn on screen, unlock as needed,
  2. make sure its reconnected to 4G,
  3. check pureos store for updates once,
  4. start syncthing check web UI over firefox once and sync,
  5. read news ~10-15 times using Feeds RSS app, mostly Guardian news,
  6. receive ~15 messages,
  7. send ~10 messages
  8. when phone is suspended and you receive message notification unlock and check message,
  9. check email ~5 times

While doing that the phone lasted almost ~6hours at room temperature. The flat lines indicate suspend operation.

  1. Observation last 20% rate of charge drop increases, probably needs adjustment to limit current use as battery voltage starts to drop towards the end. Can this be done with a driver update?
  2. it didnt really seem to matter what i used the phone for as charge drop rate mostly remains constant, so no one app being abusive to the battery (which would be an outlier and shouldnt be included in this type measurement anyways)

New Power Statistic from today, Wifi On but disabled in software also in addition to above used:

  1. in gnome software install one app (and search for 1-2 apps),
  2. uninstall weather .deb app in terminal,
  3. launch flatpak weather app and check weather


  1. battery almost lasted 5 hours
  2. similar observations to above apply
  3. 12 hours is not possible on my L5
1 Like

You may want to try it with WiFi HKS on, but disabled in software. There’s currently an issue with needless SDIO card-detect polling when the killswitch is engaged which may negatively influence battery life, and 6h is rather short based on your usage description. It should be able to easily reach ~10-12h when idling with modem on and no suspend.