Hi
What is the idle* power consumption of L009UiGS-2HaxD-IN?
I have a setup where the device is locked in a small (4U) 10" rack cabinet, without ventilation. Previously I used RB2011 (I need a few 1Gb eth ports and 2,4GHz wifi + ext. antennas) and it worked fine, with acceptable temps, but the performance wasn’t high enough. Next I’ve installed RB4011 and of course it’s very fast, but it consumes (by my measurements) approx 11,2W idle* which is too much to dissipate, and the temps rises. I consider a change to mentioned L009 if it’s more energy efficient.
Thanks
*) - idle is defined as: configured and working device, few registered devices (wifi), small traffic (up to 1Mbit).
Measurements made with cheap wall power meter, and may be inaccurate.
What I wrote above is my definition of idle device, for the purpose of measuring power consumption.
Performance I need is full 1Gb routing with firewall, VPN and many many parallel connections.
So what is your expected busy/idle ratio? If it’s higher than 0.1 (or even less), then idle power consumption doesn’t matter as the whole physical setup will have to dissipate heat, produced during busy periods …
I really don’t know what’s the point of your question, but if you are so interested it’s about 95-98% idle (2-5Mbit in+out average) to 2-5% busy (day average).
The point of my question is that minimum power draw doesn’t matter if device actually draws higher power significant portion of time … as you explained your setup lacks heat dissipation, but you have to make sure that device doesn’t overheat during expected (extended) periods of time with higher activity. And thus minimum power draw might not be significant in your case.
You’re not asking the right question so you won’t get an answer.
Power consumption as you put it “in standby mode” is a variable that depends on many factors. For example, it depends on the number of connected ports, it depends on the number and signal strength of registered wifi clients.
Therefore it is not possible to calculate exact figures only by measuring. If 12W/h is a large amount of power for your rack, you should take care to increase heat dissipation rather than trying to reduce the output by 1-2W.
You can compare the consumption of the device by referring to the specification section Max power consumption without attachments
Suitable for your application are
L009UiGS-2HaxD - 14 W
Hap ac3 - 12 W
Hap ax3 - 15 W
RB4011iGS+5HacQ2HnD - 23 W
Reduce the power consumption as well as lowering the clock frequency and reducing the output power of Wifi. You can also slightly reduce the heat generated by selecting the power supply unit. The higher the output power the power supply is designed for, the lower its efficiency when working at low power. Mikrotik devices are equipped with power supplies taking into account possibly connected devices.
The leader in terms of minimum power consumption will be Hap ac3 with 12V 1A or 24V 0.5A power supply unit
I am asking the right question, but maybe you did not understand it properly. It’s not a pharmacy or chemistry lab. An error of 10% is totally fine and acceptable. If someone would said “I have measured 7,2W” and someone else wrote “In my case it was 6,5W” it’s perfect answer. Probably no one has this box, or no one has time to check the power.
By the way, about different power supply - is a good hint.
Not exactly an answer to your question because its a L009UiGS and therefore a leaner model but mine uses 4.5 watts according to my zigbee socket switch when close to no network traffic is seen.
I'm also interested in the measured power consumption of that model, as well as a similar model without wireless capabilities. Could someone please completely disable the radio (wireless) to see if it affects idle power consumption? I assume that disabling wireless should reduce power consumption by about 2 ~ 4 W.
My current router is RB750Gr3 or hEX and it draws only 2,8~2,9 W (including also DC-DC PoE convertor to 12V as shown in the image below)
according to my PoE switch which is providing power for MikroTik:
Also great candidate is hAP ax3 (2x more RAM compared to L009) but with useless wi-fi for me and higher price
UPDATE 2:
I'm also considering the following option:
Not upgrading my old Mikrotik router (hEX), but instead:
Connect a small miniPC Radxa Zero 3E (similar to Raspberry Pi Zero 2), which has a 1 Gb Ethernet port and 1 GB RAM (there are also versions with 2 GB, 4 GB, or 8 GB RAM).
This will expand the capabilities of the old router (I know it's not router-specific, but it's more relevant to self-hosting Docker containers).
Instead of buying the expensive hAP ax3 with 1 GB RAM (130+ EUR), I will use this 30-40 EUR SBC with 1 GB RAM, a 4-core ARM CPU, and 1x USB 3 (for storage).
The Radxa Zero 3E has a power consumption of only 1~3 W when running with a microSD card (without an external USB SSD).
This option makes a lot of sense for me. It will give me full control over managing it, and I will not be limited by RouterOS. However, there are other factors to consider, such as the increased effort required to manage the entire operating system and so on.
I initially wanted to upgrade my old router solely because the new Mikrotiks have ARM CPUs and are capable of running Docker containers. However, in my opinion, a router device is not ideal for self-hosting.