Hello.
I`m considering solution for power outages.
I already have 12V 2A router.
Low power routers are not expensive but does downgrading to 5V 1A router means 2x or 1.5x additional powerbank time in real life scenario?
Laptops and phones mainly discharge during usage (screen, wifi, cpu) so twice effective power consumption does not mean twice battery life, it depends on performed tasks. Does this analogy works for routers?
Will low power router save twice battery life?
It should give you more runtime on a power bank. 12v 2A is 24W, and 5v 1A is just 5W, so the power usage is way less.
However, a less powered device can be less powerful and has less wireless range. I don't know the numbers, maybe not that much, but certainly a bit of less range.
Another thing is that the power usage is when the device is at full load. My RT3200 is 12V 2A but I read that it consumes way less than that most of the time. And even at full load, it doesn't use the 24W.
EDIT:
To make context, I have a mini DC UPS for my RT3200 and fiber modem and I can get up to 3H 15m of runtime. This DC UPS has a 8800mAh battery. A router with less power requirements should give you more time. I live in Venezuela where power outages are the norm.
There is no general answer to this, no way to avoid just trying. Actual mean power draw is not necessarily related to the rating of your power brick at all.
each USB 2.0 port means the PSU has to account for 0.5A for the connected device
each USB 3.0 port means the PSU has to account for 0.9A for the connected device, more of a margin in practice, if the vendor is pushing the idea that magnetical HDDs are supposed to be attached
beyond the external PSU, there is another stage of dc-dc voltage conversion onboard of the router (3.3V on older ones, a variety of voltages with 1.8V as SOC voltage for modern ones)
there is a per ethernet port power draw, which varies between 10/100/1000/2500/5000/10000 MBit/s, quite significantly so
contrary to older routers, modern ARM based routers will have quite variable power draws based on the load
power bricks come in standard sizes, they are not custom designed for the exact requirements of the router, it will vary how close to the actual demands the vendor made its choice.
especially in the older days, the actual onboard voltage regulation was rather wide band, meaning the same router might have been shipped/ powered with 5 V, 9 V or 12 V power bricks, depending on what was cheaper at the time (this generally isn't declared, you'd have to dissect the PCB and reverse engineer the components to get an idea)
the current rating of your power brick is only the maximum design limit, it does not at all imply that the router will actually chug that much from it (at least not all the time/ under normal circumstances)
before even thinking about the UPS side of things, what its losses and/or native battery voltage are
Try, measure and time it yourself, there is no alternative.
If you already get a router, just use the AC adaptor from factory, whatever it's 12V or 5V.
A 12V router could not be used under 5V, because of the internal RF circuit needs 12V to work maybe.
If you consider a travel router, 5V is must, especially which can be laptop USB powered.
Do additional antennas drain more power?
One good option for 5v router I found is Xiaomi Mi Router 4C. But it has 4 antennas. And it also comes with different chips. One of chips doesn't works with Openwrt. How to find out the type of chip on the Xiomi router?
The qty of antennas, donot related the DC power consumption / Voltage.
Xiaomi Mi Router 4C, 5V 0.5A, only one SoC chip MT7628DAN, I used many the model. Very good for 5V USB powered. But the WiFi signal, donot expect much under such price.
I did not meet this. what chip, the SPI flash? no problem.
Whatever PSU you choose it must always be able to deliver the maximum current needed at any time at any temperature in a time shorter than the demand at the router. At the same time the output voltage need to be as stable as needed to avoid brown-out conditions.
On top of this there is also the level of output ac ripple and noise (I don’t talk about audio noise!) at max current that is a real matter with primary switched PSU. Especially with radio equipment.
Then we have protection systems in the PSU like isolation between primary and secondary circuitry and isolation to earth. And the router can sometimes expect the PSU to be the over current/voltage protection for the router in case of short circuit.
And then it is nice if the peak efficiency is located at the operational current but usually if you don’t build the PSU yourself then this point is a mayor compromise.
How good is MikroTik hAP Lite TC RB941-2nD with OpenWRT? It is real winner on power consumption with 5V and 0.6A (0.5A for TC version). I don't need large coverage wired network is main task.
It is not listed on TOH page but on the battery powered devices. https://openwrt.org/toh/mikrotik/rb941-2nd
If you've read and understood @slh's excellent post, you should know that the power output of the supplied power supply and/or the power input specifications of the device are estimates at best. There is no way round testing!
I power my Fritz!Box 7520 and a Raspberry Pi 4 from an Eaton 3S 36W Mini-UPS. Eaton provides a nice runtime graph, displaying the expected runtime depending on the load here. Based on the table linked by @spence, I should have a consumption of about 20W, giving me a runtime of about 25 minutes.
As you can also see in the table, the power consumption varies heavily with the amount of processing going on. The same Raspberry Pi 4 can draw 12W on average when running Home Assistant (I do that) or about 2W when running just an NTP server.
The power rating marked on router/router PSU is only mean for maximum, but in many cases you will be using way less power.
For example, I have GL-INET MT1300/MT3000, both claimed to use up to 5V 3A, however without using USB port on router, just a simple travel router with WAN plugging in, I can see that it almost never uses > 3W power (tested with USB-C power meter)