MacBook – Why does a MacBook throttle without a battery

When measuring the usage of power adapters, it is important to keep in mind the efficiency of the PSU when making calculations.

For PC builders, it's important to ensure that you choose a suitable PSU for the components you intend to use, to ensure it's big enough etc. As such, the rated capacity is shown as what the PSU is capable of delivering, rather than what it uses. So, for example, a 600w rated PC power supply for a desktop computer that is rated at 80% (considered the minimum in order to get the "80 plus" certification) will actually use up 750w at the wall. 750W * 80% = 600W.

For laptops, the maths is often considered the other way around. As you are unlikely to be putting extra drives, graphics cards, PCI cards etc in your laptop, its internal power delivery capability should be of little interest to your average user, but the external wall-draw might be. Given that a laptop power adapter will likely be more efficient (as it is smaller, uses less power, will therefore run cooler, and require less active cooling in terms of fans etc etc), then it wouldn't surprise me to see it up around the 90% efficiency mark. Add to that the fact that your average laptop user should only really need to know the wattage rating when using the laptop somewhere unusual, say a plane, or some other mobile/portable power source, it may well be more useful to know the exterior power rating rather than the internal delivery rating. Therefore, 85W at the wall * 90% efficiency is 76.5W in the laptop, and likely explains the discrepancy.

The MagSafe power adapter, like nearly all power adapters for modern electronic equipment, is a switching (switched-mode) power supply.

Switching power supplies achieve their light weight and small size by using smaller, lighter transformers for voltage conversion, which require high-frequency AC input to operate. This high-frequency input is produced by a separate circuit which rapidly switches the low-frequency input on and off (hence “switching”), and in fact does so controlled by feedback from the DC output side in order to achieve a tightly-regulated DC output. This regulation itself is also something that varies over time as the input and the output load change.

So, there's lots of varying electrical currents, and therefore varying electric and magnetic fields, inside a switching power supply. These fields exert mechanical forces on the components within the power supply, which may cause them to vibrate and therefore produce sound!

In a well-designed power supply, the basic frequencies are chosen to be above the range of human hearing (and the components most subject to vibration may be held down especially firmly). However, for various reasons (waves hands, I don't actually know the details) you may get effects at fractions of those frequencies which you can actually hear.

In your particular case, if the power supply is running overloaded (or at the upper limit of its capability), then that might put some subsystem oscillating at the upper end of audibility, producing the sort of noise you're hearing.

You can often hear noise from cheap power supplies (e.g. USB chargers) when they are unloaded (nothing connected to the output), and hear it shift out of audible range when you connect something.

(Disclaimer: I Am Not An Electrical Engineer; this is just bits and pieces I've picked up. These folks would have the whole story; e.g. this answer which is a bit short on detail. But, I hope I've at least somewhat demystified why apparently solid lumps of electronics can make noise. (If anyone sees a factual error or can explain better, please feel free to edit this answer!))