Yes, of course, RTFM is warranted in there somewhere. ![;)]( "Wink")
I did see the official docs along the way, but the numbers they publish were all higher than what several sources were saying they saw so I assumed the official docs were being conservative or using max values to avoid over promising. Case in point, for the Pi 4B they have a typical active number of 600mA where I only see 400 max with my workload, and they have idle at .6A (600mA) and I've never seen mine idle that high. Also though, I've turned off WiFi and BT from the beginning, and I'm using the Lite OS, and either disabling or turning off HDMI so I don't know how much my end result compares with their test case runs.
More information is good and that's what I'm after. Considering what the docs say, and with the one data point provided here (523mA for the 400s with USB mouse plugged in), I'm WAY under those numbers at this point, so maybe the sources I found out there are just on crack.
I am currently running some tests on a Zero W (first gen) to see if I can make that work. Right out of the gate it's pulling 100-150mA (0.5-0.8W) so at those numbers I could run for at least 4.5 days on the larger power banks I have, which would be grand. In order to use the Zero though I have to modify my workflow and fully remove a step where I add a text overlay to each image. I'm still contemplating that, but probably not the end of the world. The text overlay is helpful in troubleshooting and provides a bit of context for the timelapse. I'll have to see if there is another way to get that done that doesn't hit so hard in CPU and time cost. (Image capture currently takes 4 seconds, adding the text overlay takes an additional 7 seconds.)
It remains to be seen if I can use the same Pi-UPS and cellular modem hat on the Zero, but it has the same GPIO header so I'm hopeful. I also got a response from the cellular modem vendor (Sixfab) telling me how I can fully power off the cellular modem between uploads with a GPIO pin instead of trying to unbind USB, which didn't save any power anyway.
Thanks for the replies so far. Hopefully I can get a few more data points to support the idea that the Pi3 and Pi4 running at 300mA is the real lower limit. If someone out there has been able to get one down to 150 or 200mA though (on and at idle, not inactive and shutdown), I'd be interested in how they did that. I didn't think about disabling CPU cores until today, but quick search seems to indicate that might not work out either.
I did see the official docs along the way, but the numbers they publish were all higher than what several sources were saying they saw so I assumed the official docs were being conservative or using max values to avoid over promising. Case in point, for the Pi 4B they have a typical active number of 600mA where I only see 400 max with my workload, and they have idle at .6A (600mA) and I've never seen mine idle that high. Also though, I've turned off WiFi and BT from the beginning, and I'm using the Lite OS, and either disabling or turning off HDMI so I don't know how much my end result compares with their test case runs.
More information is good and that's what I'm after. Considering what the docs say, and with the one data point provided here (523mA for the 400s with USB mouse plugged in), I'm WAY under those numbers at this point, so maybe the sources I found out there are just on crack.
I am currently running some tests on a Zero W (first gen) to see if I can make that work. Right out of the gate it's pulling 100-150mA (0.5-0.8W) so at those numbers I could run for at least 4.5 days on the larger power banks I have, which would be grand. In order to use the Zero though I have to modify my workflow and fully remove a step where I add a text overlay to each image. I'm still contemplating that, but probably not the end of the world. The text overlay is helpful in troubleshooting and provides a bit of context for the timelapse. I'll have to see if there is another way to get that done that doesn't hit so hard in CPU and time cost. (Image capture currently takes 4 seconds, adding the text overlay takes an additional 7 seconds.)
It remains to be seen if I can use the same Pi-UPS and cellular modem hat on the Zero, but it has the same GPIO header so I'm hopeful. I also got a response from the cellular modem vendor (Sixfab) telling me how I can fully power off the cellular modem between uploads with a GPIO pin instead of trying to unbind USB, which didn't save any power anyway.
Thanks for the replies so far. Hopefully I can get a few more data points to support the idea that the Pi3 and Pi4 running at 300mA is the real lower limit. If someone out there has been able to get one down to 150 or 200mA though (on and at idle, not inactive and shutdown), I'd be interested in how they did that. I didn't think about disabling CPU cores until today, but quick search seems to indicate that might not work out either.
Statistics: Posted by archaic0 — Mon Aug 18, 2025 2:43 pm