Positive Crankcase Ventilation

Positive Crankcase Ventilation

as published in BritishV8 Magazine, Volume XVII Issue 1, July 2009

by: Jim Blackwood

The positive crankcase ventilation (PCV) system, and older "road tube" systems that preceded PCV, are for venting combustion products from the crankcase in the most unobtrusive manner consistent with the times. To understand these systems, it's necessary to realize that "blowby" inevitably exists in any internal combustion engine. For example, whenever a piston engine runs, there will be some passage of gasses and vapors past the piston rings, and into the crankcase. This will occur despite the finest components, the most precise machining and the most careful assembly. Blowby includes volatile and potentially explosive fuel fumes, as well as contaminants that degrade oil. (Solids, such as ash, are carried with the gasses.) Modern sealing methods are simply incapable of preventing blowby. It may not be much, but blowby is inevitable and it has to be dealt with.

Originally engines were built with an open crankcase and no special provision was needed. The closed crankcase came about because of a desire to re-use the lubricating oil which had previously been allowed to simply fall to the ground. As this evolved, manufacturers began sealing the crankshaft from the outside environment to both contain the oil and exclude contaminants. Engine speeds were increasing as well and pressurized lubricating systems were needed to deal with the increased bearing loads, meaning there was a great deal more oil being flung about. At this point it became clear that venting of some sort was required to avoid pressurizing the crankcase to a level which seals and gaskets weren't capable of withstanding, and various types of vents were developed. Due to the possibility of crankcase explosions almost all of these venting methods also included a flame arrestor of some type, but they were in general as simple and direct as engine manufacturers could make them.

The problem with this type of vent of course is that it was nearly impossible to remove all of the oil from the vapors that exited the vent, and as a result an oily residue was deposited nearby. Also under heavy acceleration when cylinder pressures are highest and the most blowby is produced, enough fumes exited the vents to create objectionable fumes in the passenger compartment of the vehicle. These problems led to development of the road tube which was very effective, resulted in much cleaner engine compartments and cleaner air for the driver and passengers. It diverted the objectionable fumes and oil spray down under the car and was a significant advance. The downside was that the oil vapors were now deposited on the road surface, leaving a wide, dark oily swath between the two tire tracks of public roads.

1961 Buick 215cid: road draft tube and breather

It was soon discovered that by cutting the end of the road draft tube at an angle and extending it into the airstream below the car an actual flow of fresh air could be created through the engine by providing an inlet breather, often in the form of an oil filler cap with a coarse mesh filter material, which acted as a flame arrestor and kept bugs out of the crankcase. A similar mesh in the road tube (sometimes placed inside the engine) served the same purpose on the other end. This new development was highly touted as an advancement which increased both engine life and oil change intervals by removing combustion byproducts quickly before they could combine with the crankcase oil and form acids, sludge, and other harmful and non-lubricating contaminants.

This insight ultimately led to the development of the modern PCV system as we know it today. Various schemes were tried in order to find new ways of evacuating gasses from the crankcase, but what most had in common was that they relied on some means of suction to do the job. This generally meant the use of manifold vacuum as the most practical method. For us this means two things: firstly that most PCV systems are pretty consistent, and secondly that other alternative methods are possible.

1974 MGB GT V8 - flame trap, breather/filter


 Please support the sponsoring companies who make BritishV8 possible, including:

The usual PCV systems come in two basic flavors, and I'll distinguish them here by referring to them as American and British respectively, due to the prevalence of their use by manufacturers in each country, and in particular in distinguishing LBC's from Detroit iron.

In the British system, which is the more straightforward of the two, manifold vacuum is plumbed directly to the crankcase using a 0.5" or 0.75" diameter line. An orifice of typically 0.030" or less is provided on a line feeding fresh air into the crankcase. Sometimes this line draws vapors from a vapor recovery canister as well, thereby purging the canister and feeding those vapors indirectly into the engine intake, and sometimes this orifice is contained in an oil filler cap, but for the purposes of PCV it functions the same either way. It does however have the potential to either enrich or lean the idle mixture to a limited degree. The manifold vacuum purges the crankcase of blowby fumes. By placing the crankcase under vacuum, a metered quantity of fresh air is drawn into the crankcase through the orifice... at least at idle and part throttle. As the throttle opens and engine load increase, blowby also increases proportionally until at some point the crankcase transitions from vacuum to positive pressure. This may not occur with a new engine in good condition using a larger vacuum line, but it certainly will with an engine having significant mileage. When the crankcase is at positive pressure, nothing changes on the vacuum side other than the volume of gasses going into the intake, but on the orifice side the flow reverses. This is usually not particularly significant due to the small size, but can contaminate the carbon in the vacuum canister over the course of time.

Enjoying this article? Our magazine is funded through the generous support of readers like you!
To contribute to our operating budget,  please click here and follow the instructions.
(Suggested contribution is twenty bucks per year. Feel free to give more!)

American systems use a large diameter vent line to the atmosphere, usually running from a valve cover to a location in the air filter housing, and a smaller diameter line to the intake manifold, usually 0.3125" to 0.375" in diameter. This system restricts the amount of combustible air which enters the engine with a PCV valve rather than an orifice, and it is placed in this intake line. The PCV valve is pretty unique in that it allows full flow at low pressure differentials across the valve and a metered restriction above that level. It also shuts off flow in the reverse direction, thereby eliminating the need for a flame trap. This is done by using a shuttle inside the valve which I'll return to shortly, as this is the most misunderstood part of the system.

The crankcase is never under vacuum but can become slightly pressurized. The large diameter vent line allows fresh air into the crankcase and also allows excess blowby to vent into the air cleaner housing through a flame arrestor where they are ingested by the engine, whenever engine loads and throttle openings are great. Under part throttle the PCV valve is open due to the lower vacuum applied across it and the lower level of blowby so most if not all of the blowby is sucked into the intake, drawing fresh air into the engine in the process and also tending to lean out the intake mixture during cruise. During idle there is enough vacuum to shuttle the PCV valve to the metered position and there is usually very little blowby so most of that is ingested along with a small amount of fresh air which is accounted for by adjustment of the idle mixture screws. This is one reason the idle mixture has to be adjusted as the rings wear. Under heavy throttle, although the level of vacuum drops to near zero, should the level of blowby become great enough the shuttle will shut down the flow to the metered level, diverting most of the blowby to the air cleaner and ingesting only a metered amount of blowby through the PCV valve.

The effect of routing these gasses through the air cleaner is to enrich the intake mixture proportionally because the carb doesn't know the difference between fresh air and recycled combustion by-products. So it is interesting that the PCV system helps to give us a lean cruise and a rich WOT, but there is little or no correlation between PCV, carburetor, and volume of blowby other than the initial calibrations of the carb and PCV valve and no mechanism to account for engine wear other than the shuttle valve and idle mixture screws. The interesting thing about this is that the same PCV system is still in use with very little modification on our newer fuel injected engines, although they do have a feedback mechanism in the form of an oxygen sensor.

These systems work quite well normally, but things tend to get interesting once performance modifications are made. Often the PCV systems are unintentionally modified to the point that they can no longer function properly, and this is particularly common with aftermarket intakes, air filters, valve covers, forced induction and the like. It is still possible on almost any performance engine to design and tune for a PCV system that works properly and this is especially important in a street driven car. For all out performance it is less of a consideration and in fact the blowby fumes do dilute the intake charge somewhat, so in these cases a simple crankcase vent is often used, harking back to the early days, with all their attendant inconveniences. Often modifications are made to the system unintentionally, in the quest for more performance and a better appearance, and this can result in problems. One of the most bothersome is pressurization of the crankcase, with symptoms of excess oil sprayed about the engine compartment in various places as it is forced past gaskets and seals. Another is the chance of a crankcase explosion should the need for a flame trap be overlooked. Then of course, any improperly functioning system will have a need for more frequent oil changes, as the combustion byproducts contaminate the oil more rapidly.



For your street driven car there are answers. Sometimes the British system is more appropriate, and sometimes American, but it's best to keep in mind how each one operates and not try to mix the two. When fitting an open element air cleaner onto a typical American 4-barrel carbureted V8, for instance, it's quite easy to install the vent line to the base of the air cleaner inside the element, thereby keeping the system intact. Things aren't quite so neat and clean when fitting a similar air cleaner on a Rover V8, because Rover engines were originally set up with an orifice-type system. Some of these are easily changed over and some are not, depending on the fittings on the rocker covers. Bear in mind that the large vent line of 0.625" or possibly 0.75" in some cases cannot be replaced with a 0.375" hose, or even two of them. Flow increases by the square of the diameter, meaning that a 2" pipe flows four times as much as a 1" pipe. So you would need four 0.375" lines to match one 0.75" line. If your valvecovers do not have the proper fittings and you are not willing to add larger ones to them then you may need a larger line to the crankcase in an alternate location. Early SBC's had a vent line that went through the block web behind the intake manifold. A large diameter tube rising from the pan may be another option. You might use the mechanical fuel pump mounting boss. Or you may be able to use the orifice system, bearing in mind that using smaller lines will cause it to operate under pressure more than it might otherwise. This may require recalibration of your carburetor due to the differences mentioned above, or it may be possible to find a suitable carburetor calibrated for the orifice system, since some British cars were available with a four barrel carburetor. There are other cases where an orifice system is a good choice, such as any induction system where the throttle body is at the inlet of the system. This might be the case with an IR setup or where an engine is supercharged. In these situations there is no practical way to locate the vent tube, and therefore no way to ingest the excess blowby fumes. The solution is to port the crankcase to manifold vacuum (or inlet vacuum in the case of a blower) and use the orifice to restrict flow into the crankcase and subsequent leaning of the intake mixture. It is worth noting here that often EFI systems take control of the PCV system, such as by including a purge valve for the emissions canister, allowing options such as pulling fresh air in at idle to give more precise control of the mixture. Some of these systems may be able to divert PCV intake on WOT for maximum power output. WARNING: All lines from crankcase to intake system must have some form of flame arrester! To overlook this is to invite a crankcase explosion, which in the best possible scenario will have you replacing your lifter valley pan.

Finally, we have the alternative systems, the most familiar being the collector scavenger tube. These are really not suitable for a street driven application because of two things: mufflers, and the fact that they do not work well at idle. For racing it's a good idea, but once you add the restriction of a muffler you have created backpressure which will reverse the flow in the scavenger tube and make the system ineffective. Another option is an external scavenger pump. These tend to be a more complicated solution, but are another possibility that may have merit in special situations. Theoretically it would be possible to evacuate the crankcase using positive pressure and this would tend to lead to seal problems, but if those woes were overcome one might even find a way to route forced induction through the crankcase on its way to the cylinder, provided an adequate air/oil separator were designed. Two stroke engines inherently use this principle, simply burning the oil as they go.

So that's the basic lowdown. No doubt there are details that I've left out but it's enough to give the average person the picture. Hopefully it's enough to help sort through the tangled mess and maze of confusion that typically surrounds these systems.



Disclaimer: This page was researched and written by Jim Blackwood. Views expressed are those of the author, and are provided without warrantee or guarantee. Apply at your own risk.

British V8 Home:        Read the Magazine        Photo Gallery        Web Forum        Annual Meets        Contact Us        Site Map