It does pretty much what it says. Releases pressure from the engine. Engine runs, pistons open easier especially on admission cycle if there is no pressure on the other side of combustion chamber. Here's an illustration showing how this occurs. A small percentage of the energized gas during combustion cycle pass the piston's rings and pressurizes the main engine compartment or the area beyond pistons. Up until 70s they had this relief opening on top of the engine that was ending with a small pipe under. Blowby gas is a mixture of unburnt fuel, raw exhaust (not passed through catalytic converter), water vapor and oil vapor. Due to pollution complaints, that was not acceptable anymore and they came with this idea. Returning those gases in the intake manifold, mix them with air and gas and burn them again. It was called PCV system.
It was working and nobody complained about for a long time. However, if uncontrolled through design, it may produce a variety of problems, including corrosion and mainly condensation of oil on the cold intake runners' walls and pouring in the engine, affecting valves, spark plugs and creating carbon deposits on the top of the cylinder, etc..
The biggest problem of course is carbon deposit on valves which start to interfere with their functioning. Valves become "sticky" that is they can't open anymore at certain engine stages (low RPM, cold engine) by they hydraulic lifters wichi are designed to fill with engine oil to eliminate any play. Oil gets squished out of the lifters and starts foaming, which is a bad condition for the bearings that are lubricated with that oil.
At non GDI (Gasoline Direct Injection engines) has injectors spraying fuel in the intake runners, or before the valves. This help with cleaning of the valves. However, there are new generation high efficiency engines that use little gasoline at cruise speeds and that is not enough to clean those.
Many gas sellers advertise their gas as having cleaning features. This is what they're talking about and it's a real problem.
There are ways to minimize this phenomenon however not used by any manufacturer i know about. You can install an empty can on the return PCV line and have the oil condensate and fall in it. However, it may not work if the pipe before the can is close to the engine and gets hot. Blowby will still pass and condensates inside the intake. Last version of the catch can i installed on the Hyundai uses a corrugated copper pipe that is passed through an larger pipe (empty can) that allow air to flow and prevent heat from engine to reach for it and indeed it minimized the sticky valves phenomenon, becoming marginal and intermittent.
However, there is another problem. Fuel pump filter is not or hardly serviceable at this car, being in the tank and in time fuel pressure start do decrease. This is aggravated by a bad battery and connectors at the battery, lowering the voltage for the fuel pump, bad gasoline or gasoline without alcohol which usually cleans and especially dissolves water.
Last time i filled the car with gasoline at Spirit Mountain Shell i noticed the noise condition improved dramatically. For the first gallon of fuel used or that 50 miles trip back home. The next day it came back to "normal". I figured, maybe it has something to do with vacuum being created inside the tank by the purging system. Another newer anti-pollution feature of cars which insure that the fuel tank does not have and/or releases fuel vapors in the air, with a charcoal canister and a system of valves that allow the computer to "purge" or literally scuk back in the again intake the extra vapors. However, if too much vacuum is being applied to the fuel tank, it starts to interfere with the fuel pump and ultimately the fuel pressure at the injectors. (Some of) the injectors start to drip and can't clean the valves anymore. On this car computer has the possibility to allow atmospheric air to go into the tank to release the vacuum however through a, you guessed, title of the post, a small air filter. Which is on the service list of 30k and is located under the car next to the charcoal canister.
So i opened and closed the tank cap every time when i noticed the noise and mostly at the beginning of every trip and voia! Noise was gone last night for most of the 150 mile round trip to the beaches.
The ironic part is, i had the filter since 2 years ago when i almost figured the problem. But finding oil and manually purging it from the intake and fiddling with the catch can made me forget about it. Also three years ago i almost went for an appointment to replace fuel filter, which doesn't make much sense to replace at this car without the fuel pump.
So today i went and jacked the car and removed the small air filter. I was expecting to find it clogged (and or the air pipe) with pine needles that were built in that area up to half inch when i got the car (from the road she was doing daily probably). That was not the case. But what i found was a solenoid next to filter again used by computer, that could have been intermittently stuck. I poured some 91% isopropyl alcohol in it and it came dirty. I tested it with a meter and a battery and it works and on the bench it doesn't get stuck. I put it back together with the new filter.
In the image representing the whole purge system, solenoid is nr.6, filter is 7.
I figured it would have been useless to go to a dealer and have it done at service interval, if they didn't check the solenoid as well.
Too many times i noticed all these optimizing devices interfere with the function of the vehicle, in the end making the pollution worse because the conditions they create are hard to diagnose, unsignaled by "check engine light" and sometimes expensive to fix not because of parts involved but because of high quality labor and time needed for diagnose. Most mechanics ignore them anyways.