I have been trying to optimize my power usage and noticed that the worst device in my rack is my Mikrotik CCR2116. It has a power factor of 0.46 (based on Raritan PDU) which means it is using more power then my single socket Intel 4214 with platinum power supplies when idle.
I have found an old post on Mikrotik forum that sounds like it does not matter, for devices < 100w. As there are no standards. It is an old thread though.
I was shocked to find this out since all the reviews talk about how low power usage the Mikrotiks are, which is definately not the case if you factor in the power factor. I am guessing the power supply or fans are cheap.
I am leaning towards the power supply to try to replace with a Meanwell EPP-100-15 which has PF > 0.95 and 91% efficiency. Has anyone done this before? or have models of power supplies that worked, by the looks of it its 2" x 4" power supply but might be just bigger to ensure only their supply works.
The original one should be this: https://mikrotik.com/product/up1302c_12
so 10.8 A @ 12V
10.8x12=129.6W
it seems like largely over-dimensioned, if it is true that the CCR2116 is 60-83W.
There is another thread:
If I understand correctly, you seem to miss what power factor is measuring. Your power consumption does not increase because of a low power factor. (Neither will more heat be generated, nor will your electricity meter spin faster, nor will you pay more.)
With a decreasing power factor, a higher so-called apparent power is present. This has influence on how to size conductors, transformers, generators, upses, protection devices, etc. They obviously have to be over-dimensioned (compared to a situation with a device having the same power with a unity pf) and therefore infrastructure providers don't like them. The circuitry for correcting the pf is both costly and increases size, so regulations are balanced in order to not effect your phone/laptop charger/led lightbulb, but still require higher consumption devices to incorporate these solutions.
(In all likelihood, your laptop charger, monitor, etc. will show similar values. If your meter is working correctly, that is.)
One of the things to be aware of is that power factor can be lagging or leading according to whether the load is partially inductive or partially capacitative. And if the load is electronic, the power factor can be entirely misleading, sometimes exceeding 1.0. If it is leading, you can compensate with a lagging load And vice versa. In factories, old school fluorescent lights would often partially compensate for motors.
For electronic power supplies, you may find that the reactive component of the load remains fairly constant over the whole range of real power, giving a horrendously bad Power Factor at low load, which actually becomes quite respectable when the power supply is near full load.
Overall, for 100W loads, as already explained, it is not worth worrying about.