ok, so I cleaned the PCV valve...

The valve "orifice" varies considerably according to the relative pressure
of manifold and crankcase

Go take a *good* look at the path the air, then air+blow-by gas takes in a
Honda system with breather box: air comes in from upstream of the throttle
butterfly through a tube to the valve cover, gets mixed with blow by gas,
sucked into the breather box mounted low on the block and up through the
PCV valve into the manifold. Later models just have the PCV valve mounted
on the valve cover.

Yes - it's a metered air leak according to the variable orifice of the PCV.

The air comes from upstream of the throttle - within the operating
parameters of the engine that's atmospheric pressure, minus a little for
the air filter restriction. I dunno what you mean by "new air" but that's
air that the ECU has not measured - it has bypassed the primary device for
signalling mass air flow to the ECU: the throttle and its position sensor.
I'd doubt it has any effect on the IAT or MAP sensors either.

The pressure at the air intake end of the breather is highly dependent on
the throttle opening - when closed it's essentially atmospheric as
mentioned above. That is also now a different path to the induction
system... past the throttle valve and there's no air mixed with the blow-by
gas. IOW you've lost the metered air-leak.

You will not be able to operate an engine in that condition for long before
you have serious problems with the intake manifold and its various
precision components, chambers and narrow tubes. Your throttle butterfly
is going to get in a helluva mess and the rubber intake tube is not oil
resistant rubber - it'll rot.

If the spring binds, the valve may not be able to float and jam in the
fully open or fully closed position... or anything in between. Most likely
it will float and stick intermittently initially.

Take a look at a used one - there's oil emulsion (acids), lacquering and
even fine particulates in there. Combustion gases mixed with oil mist and
water vapor takes more than "oil compatible" materials to stand up.

Absolutely not. You can put a new plug in, in say October, and it'll be
pushed out slightly by end of Winter. The rubber is not old or perished in
any way. I do not need some mechanic to explain to me how this works -
I've replaced the plug twice myself and seen the condition of the rubber on
two different cars. In fact I suspect that just cleaning up the old plug
would work as well as a new one - once the oil has worked its way past the
ribs on the plug, which is BTW only encouraged by pushing it back in, it
has established a path to escape.

Take your oil filler cap off with the engine running and feel the pulsing
of err, pressure. It is not a static even pressure. Note that pulsing is
there even if you had no blow-by at all and in all engines except for a
very few V configurations; add in blow-by and reduced capacity to extract
it and the pressure oscillation is increased as well as the max crankcase
pressure.

If both breather and PCV were blocked you'd blow out seals in a very short
space of time.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

Fer Pete's Sake, TeGGeR, let it die. Crossposting is classic
trolling.

 

I want misinformation (yours) spread as little as possible.

I'll try to respond to those who join us here at rec.autos.makers who seem to be
in the know, rational, open-minded, and who haven't had a response from others.

You can quote away in the other newsgroups. At some point, it's likely I'll
provide all the citations I provided earlier to help persuade people of the
truth that a malfunctioning PCV valve may (1) affect fuel economy and (2) may
cause excessive crankcase pressures, causing certain oil seals to fail
prematurely.
snip junk

This does not work, as it does nothing to explain what is happening in the
crankcase.

If one is going to consider in detail the relative pressures involved, yes it
does.

.... at low load on 91 Honda Civics (among other models) via the breather hose.
Agreed.

.... via a PCV valve which meters the flow according to intake manifold vacuum.

At low load and with a PCV valve that has failed shut (for whatever reasons),
you're assuming that the flow described above reverses. But you can't assume the
breather hose can accomodate the blow by the way the PCV valve can.

Just because the pressure in the crankcase is now higher than where the breather
hose connects to the air intake does not at all mean the crankcase will be
purged to the same extent it is with a correctly functioning PCV valve.

I still have no idea what denotes crankcase pressure in your system.

My analogy illustrates flow directions and relative pressures correctly for a
1991 Honda Civic (and no doubt other Hondas) at WOT (wide open throttle). Yours
doesn't illustrate it at any load, from what I can tell.

Again, my model assumes the level of the water denotes crankcase pressure.

Note: My model is only for the purposes of illustrating why crankcase pressure
may rise with a plugged PCV valve. It does not capture all the nuances of a PCV
system.

Nope. On a 1991 fuel injected Civic (as well as other cars):

At low throttle:
--breather hose flow is from air intake (near the air filter) to the crankcase.
-- PCV hose flow is from crankcase to air intake (near the cylinders' air
intake valves)

At wide open throttle:
-- breather hose flow is from crankcase to the air intake (near the air filter)
-- PCV hose flow is zero, because the PCV valve is shut.

http://tinyurl.com/yrq9s (Autozone repair guide) says this, too. It reproduces
exactly what is said in my Chilton's manual for 1991 Civics. Note: Figure 1 at
this site is also for 1991 1.5 Liter FI Civics.

Wrong.

The PCV valve does not employ the Venturi principle. Flow through the valve is
due merely to (1) the appropriate pressure differential that causes the valve's
control mechanism to open it; and (2) a pressure difference from crankcase to
PCV valve outlet (which would cause fluid flow through any hose/pipe/whatever).

Practical application of the Venturi principle generally involves two different
flows of fluids. Fluid 1 will pass through an orifice (a.k.a. Venturi channel)
of some kind. The orifice increases Fluid 1's velocity but reduces the pressure
(conservation of energy; Bernoulli principle). Near the discharge of the
orifice, Fluid 2 enters via a connected pipe (or hose, or whatever). Fluid 1
draws Fluid 2 into its flow, and they now mix together.

Parts of carburetors, for one, use the Venturi principle to draw gasoline flow
(Fluid 2) into the flow of fresh air (Fluid 1).

There is no such setup in the PCV system.

In this part of the thread, we were only discussing how crankcase pressure may
rise above normal (on average) with a plugged PCV valve.

I see you are now conceding this point.

That's another part of the thread. I have not joined this, as we did in fact
discuss this to some extent a few weeks ago. My citations and summary
explanations speak for themselves. But you want to re-engage, so...

You are convinced the ECU's proper functioning in open and closed loop has no
dependence on how the correct operation of the PCV valve.

I on the other hand am convinced that in either open loop, closed loop, or both,
the ECU's proper functioning depends on a certain design flow through the PCV
valve. For example, with a failed shut PCV valve, the ECU has planned for a
certain response to (intake) manifold air pressure changes *assuming* a certain
flow through the PCV valve is occurring.

You seem to be rebutting, "But in closed loop (O2 sensor in play, providing
feedback), if the response is 'wrong,' the O2 sensor will see the wrong level of
O2 in the exhaust and send a (too rich) feedback signal to the ECU. The ECU will
adjust the air/fuel ratio for optimal economy."

But how does the ECU do this? In my 1991 fuel injected Civic, the ECU uses
signals from the MAP (manifold air pressure) sensor and crank angle sensor to
control fuel injector "basic discharge duration." The ECU's memory has pre-set,
fixed fuel injector "basic discharge durations" depending on these two inputs.
The fact that these basic discharge durations are pre-set and fixed means they
assume a certain flow from the PCV valve is present.

The O2 sensor's signal (among other sensors' signals) will modify the basic
discharge duration somewhat, but the engine is still stuck with a particular,
basic discharge duration. The pre-set values are not appropriate for a no-flow
condition through a malfunctioning PCV valve.

http://tinyurl.com/22d46 describes the general setup of a 91 Civic LX Programmed
Fuel Injection system. The remarks above about "basic discharge duration," etc.
derive from it and my own engineering experience.

The above is one explanation of why in closed loop fuel economy may be bad with
a malfunctioning PCV valve.

 

I concur with this analysis of what may happen when a PCV valve gets gunk in it.
"Clogging" may result in either a fail open, fail shut, or fail "intermediate"
condition.

The rest of George's analysis looks fine to me as a general description. I do
note that, for one, the 1991 Civic does not, from what I can tell, have a mass
air flow sensor. The car (like many cars) has other minor idiosynchrasies that
distinguish it from George's description.

I hope we all agree that there is some variation from one car model's PCV system
to another and so are somewhat generous when reviewing descriptions of flow
directions, among other features.

 

Oops. Not.

At WOT, on the 91 Civic by design the PCV valve is shut, and crankcase gases by
design exit through the breather hose.

At non-WOT, on the 91 Civic by design the PCV valve is generally open at least
partly.

My model is simply illustrating that pressure will rise in the crankcase when an
ordinarily open escape path (the PCV valve or its hose) is blocked.

This is the same as blocking off one or more exit holes in the bottom of a
bucket of water previously being filled such that the rate of water entry = the
rate of water exit, and such that the system maintains a certain level of water
in the bucket. Goes ins = goes outs, as we used to say in the nuclear navy. But
block off an exit hole, and the level of water rises, just as average crankcase
pressure rises when blocking off the PCV line.

Tegger can argue that at WOT, the PCV valve is supposed to be fully shut, and
crankcase gases exit via the breather hose, by design. So a blocked PCV valve on
a 91 Civic has no effect at WOT.

But how often is a car at WOT?

Certainly not all the time.

 

Some mfrs, notably those with Bosch FI systems, use a mass airflow sensor -
often a hot-wire thermistor AFAIK. Honda derives mass flow, I assume in
the ECU programming, from throttle position, intake air temp and manifold
absolute pressure. Note that those actual measurements can also be used
for other programmable parameters of the ECU.

I think there are some which just have to be wrong though - possibly
outdated? IMO the Toyota doc (h63.pdf) looks the best explanation.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

They're not *all* bad - good ones who are honest with the customer are
rare... IME.

No it has *not*. As already mentioned umpteen times, the breather tube
intake is *upstream* of the throttle plate; geez you told me to look at my
car but you haven't even done as much yourself! BTW there is no air flow
meter in the Honda systems that I've seen.

I guess we just know better. See above.

Not exactly but in the h63.pdf, a Toyota doc, it states quite plainly: "The
installation of an incorrect valve may cause engine stalling, rough idle
and other driveability complaints". It's not a big hop to conclude that a
malfunctioning PCV is an err, incorrect valve.

Just RTFM... the service manual and apply a bit of scientifically informed
interpretation.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

Yes. Perhaps you're also getting at the fact that fuel/air ratio is not the only
closed loop controlled engine parameter. A certain Ford service manual I was
reading the other day noted that engine RPM control is also closed loop: The
actual RPM is fed back to the computer and compared to the desired RPM, and the
computer responds to adjust the RPM.

The overlap of these closed loop engine control systems is inevitable of course.
The modeling is sophisticated and probably proprietary. The latter comment is
based in part on having known a few PhD'd Ford control systems engineers who
work on emissions control design and implementation of the same. Not trying to
name drop; just trying to say there are of course people who have specific
model's full ECU control system model on paper (er, computer) and could give a
very precise and practical answer as to how the ECU responds to different
changes for that model. But access to this info by the public is difficult or
impossible. Plus, I for one no longer have access to a computer or software
capable of handling simulation of such complex models that might offer more
insight into the interactions of the systems.

So I think analyses here of how the ECU controls the engine can only be at a
very basic level.

For Toyotas, I agree. But one item that I think is impeding analysis here is
whether the PCV valve is wide open or fully shut at wide open throttle (WOT). I
am convinced (for now) that this does vary from one car make, model, and year to
another. All the designs still accomplish the same thing--removing crankcase
gases. But they are engineered differently. (I expect you know all this, but for
the sake of the written record... )

In support of this, at the library the other day I pulled down several car
makers' Chilton manuals. It's easy to turn to the first few pages of Section 4
"Emissions" to see specifically whether the PCV valve was wide open or fully
shut at WOT. Some of the manuals are a little vague on this point. Others are
not.

All or at least several models of Hondas and the early 1990s Acuras have the
PCV valve fully shut at WOT. The manuals explicitly add that the crankcase gas
flow at WOT is then exclusively out the breather hose to the air intake
(upstream of the throttle, like you noted in your other, recent post). Because
of this explicit statement, it does not seem like this is a mistake. It could
be. As you say, the manuals do have mistakes. But I'm not yet convinced.

By contrast, the reports for Toyotas (and some other makes) I've seen have the
PCV valve fully open at WOT.

Pressure differential and flow-wise, both the Toyota explanations and the Honda
explanations appear to me to be rational. Truly intelligent commentary on why in
one the PCV valve is shut and in the other it is open might demand more inside
knowledge of the systems than is available.

IIRC Mitsubishi engineers its PCV system setup quite differently from the other
manufacturers I read.

As your Toyota document http://www.autoshop101.com/forms/h63.pdf shows, some PCV
systems use a fixed orifice instead of a PCV valve. The orifice of course is not
as precise at metering in accord with engine load.

SHO Ford Tauruses from 1986-1995 do not even have a PCV valve nor apparently
orifice. (I was recently helping a relative with fuel mileage concerns on his
Taurus, thus I've been reading about its PCV system design too.)

I suspect the extent to which fuel mileage may be affected by a malfunctioning
PCV system will vary, and possibly quite a lot, from one car make to another.

Nonetheless, for the readers here, my understanding is that a properly
functioning PCV system first and foremost ensures the crankcase vapors are kept
to a minimum. I'd say at a minimum for all cars the engine oil may become
dirtier sooner with a malfunctioning PCV system. This by itself warrants its
inspection and/or replacement as a routine maintenance item. If fuel mileage
drops off, it's also certainly worth considering replacement/cleaning of the PCV
valve.

Treatise over.

 

True and I don't have inside info on those particular aspects of auto
design for any mfr. I'm pretty sure though that after all the theoretical
behaviors have been modeled and optimized, the production model programming
of the ECU is subject to fine adjustments based on empirical testing for
low emissions.

Chilton's manual says this? They are notorious for being full of mostly
generic info/pages supplemented by a few token pages for the specific model
at hand. The one I got for my VW Beetle had a section on replacing
coolant.<shrug>

I have Helm manuals for '88 Integra, '90 Civic, '92 Integra, '99 Integra
and 2K Accord and none of them go into details on exact behavior of the PCV
valve - all have just the simple hose pinch check for verification of
functionality. I'd like to know where Chiltons got their info from and how
they arrive at their conclusion - frankly I have my doubts about their
accuracy.

The illustrations of PCVs in the Honda Helm manuals, though small, look
very similar to the one in the Toyota h63.df doc, with a pintle being
sucked at high manifold vacuum into a restricted neck, floating somewhere
in the middle of its travel at low load cruising and being very close to
closed at high load. That last condition where the spring is *almost*
fully extended provides the least restriction at the pintle end of the
valve and therefore the highest flow of blow-by gas.

Toyota notes that in that last condition, if blow-by exceeds the PCV's
ability to flow the gases, they will take the route of the breather hose.
This all makes sense to me.

I'll add that a PCV which is not allowing sufficient flow is going to cause
accelerated deterioration of the components of the induction system: the
rubber intake hose, throttle valve, EACV, fast idle valve and any other
vacuum driven items. Besides the usual hose pinch check, it's also worth
checking the rubber intake hose where the breather pipe enters it for oily
emulsified mess as a guide to whether the PCV needs replaced.

Chiltons notwithstanding, I think we're in agreement.

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

Absolutely.

I recall the academic folks I mention above giving Christmas presents to the
techs in the Ford motors emission controls labs, where new "on paper/computer"
designs and theory were being implemented. They all worked that closely. (Not
sure how they got along, but they definitely had verbal communications.)

snip

Yes. The Autozone online repair guides repeat it.

I've never had a Helm manual in my hands. Someone here recently said the manual
pages accessible via
http://www.honda.co.uk/owner/workshopmanuals2.html were identical to Helm. While
I realize manuals in general are full of errors (inaccuracies?), I don't see
anymore in my Chilton's than I do at the link above.

One matter over which I'm a little anal right now is how neither Chilton's nor
the UK site manuals above talk about removing the cruise control actuator for
their 1991 Hondas in the timing belt instructions. The UK site manuals only seem
to have caught onto the necessity of this step in the 1995 or so model year
Hondas. (Haven't checked the 1996 Chilton manuals yet.)

OTOH, you have been reading these various manuals longer than I and have the
Helm manual in your hands, so of course I recommend readers defer to your
experience on this.

They don't say that the opening of the PCV valve is proportional to intake
manifold vacuum?

E.g. from http://www.honda.co.uk/owner/CivicManual/pdf/11-109.pdf for 1995-1997
Civics:
"When the engine starts, the plunger in the PCV Valve is lifted in proportion to
intake manifold vacuum... "

Not to drag this out too much, but I'm figuring at the moment that at WOT,

-- blowby gas production is at its maximum.
-- average crankcase pressure is likely at its maximum

Where things become confusing is when assessing the pressures at

-- the breather hose's connection to the air intake, just upstream of the
throttle. This absolute pressure should decrease as the throttle is opened.

-- the PCV valve connection to the air intake, downstream of the throttle
(should be a varying vacuum; lower vacuum = higher abs. pressure at WOT, at
least for Honda's systems)

I figure on Hondas, at WOT, the differential pressure from crankcase to breather
hose connection meters (or just accomodates) the flow well, by design. The Honda
PCV valve shuts.

On Toyotas at WOT, the differential pressure operating the PCV valve and the
valve design meters the flow well. The Toyota PCV valve stays open at WOT(??)
but is assisted by the breather hose?

To be very clear, I recognize two valve "seats" in the PCV valve. The Toyota
drawing's bottom seat is pretty much an "all or nothing" seat design. The Toyota
drawing's upper seat is designed to regulate in accord with pintle position.

So above I hope you mean the bottom of the pintle in the drawings at high load
is almost touching its seat. At the same time, the top portion of the pintle's
location is such that the second, upper valve passage is as wide open as
possible.

As load increases and the bottom of the pintle nears the bottom in the drawings,
the flow through the valve may be followed by tracing the valve flow
characteristic curve (page 2) leftwards, until the peak valve flow is reached.
Continuing left there is a quick dropoff on the curve.

The Toyota valve's flow always stays above the blowby curve?

The Honda valve's flow drops below the blowby curve.

Unless of course Honda Chilton's is all screwed up.

Same here, except the Toyota site isn't clear about whether the PCV valve
actually closes (as its characteristic curve implies, somewhat) at WOT.

snip

Due to oily acidic(?) exhaust compounds accumulating on parts?

Yes.

 

Yes, those pages resemble the Helm manuals... because the Helm manuals are
actually Honda manuals. Helm just handles printing/binding in some cases
and distribution.

It's not in my Helm manuals for '90 and '92 either.

Yes, but I mean more details: lift does not necessarily translate into more
flow, as we know.

I don't think either of the above are that important at WOT - with the
throttle valve open there's no pressure drop across it.

In both cases, the only criteria I see here are the spring tension when
it's fully extended and the restriction against flow through the breather
tube due to its diameter - the crankcase pressure has to rise sufficiently
to lift the spring off its seat for the gas to make its way through the PCV
valve... remembering that with the engine off the PCV has to be closed.

I haven't seen the Chilton curve but I'll project it through my prejudice
of Chilton's manuals.

It seems pretty clear from the text and the illustration that it is nearly
closed and you get "maximum flow of blowby."

Rgds, George Macdonald

"Just because they're paranoid doesn't mean you're not psychotic" - Who, me??

 

If water does not change it's density, why does ice float???

Probably you meant that water can not be compressed.

 

6/14/2004 3:39:37 PM


<opr9lnkbawfwcok5@martin>

Well, it ain't easy, but that's wrong too, take some work, but
liquids can be compressed too<grin>

http://www.newton.dep.anl.gov/askasci/chem00/chem00101.htm

 

I guess if we are talking 'black holes' then you are 100% right (grin)

 

6/16/2004 4:17:16 PM


<opr9pem2lvfwcok5@martin>

(grin)



Absolutely! Alt fuel source and I'm sure I carry one in my wallet!

 

Or a person can take a peek at the inlet and outlet hoses on a car's power
steering pump.

Guess why the outlet hose's diameter is smaller.

Liquids can be compressed. For many applications, it's not a trivial design
consideration.

 

Wow, compressible fluid flow at the power steering pump, so compressible
that the outlet is smaller. I never thought of that.

Well, next time I buy brake fluid, I'll make sure it's the expensive stuff
because the cheap fluid can be compressed and make my brakes feel spongy.

 

Sorry, but I don't think so in this case. When the pump is sucking from
the resevoir (open to atmosphere) the most suction possible is 1 bar (1
atmosphere), however on the output side, the pressure is as high as the
pump is capable of (less any pressure release valve etc), hence the
smaller diameter. In everyday applications, one can normally assume that
liquids don't compress enough to make much difference. However
temperature can have huge effects with both viscosity and themal expansion
changes, and working fluids (power steering fluid, A/T & engine oil) will
increase their temperature.

 

It's the pressure change across the pump that's important.

Note: 1 bar does not equal 1 atmosphere. It's close, but no cigar.

Less any pressure relief valve?

If the pump has a pressure relief valve and it's open, the pump is not operating
in a normal condition.

You do know that the density of the liquid and its pressure are directly
correlated, right? Given that

-- mass flow = density * pipe cross-sectional area * velocity
-- mass flow in = mass flow out
-- velocity is fixed by pipe and pump material considerations
-- pressure is higher at the outlet (otherwise it's not a pump), so the liquid's
density at the outlet is higher

what conclusion can one rationally draw? The cross-sectional area must be
smaller at the pump outlet.

If by "everyday applications" you mean those pumps that cause small pressure
rises, I agree it's fair to assume the density is fixed.

But for pumps that cause large pressure rises, and depending on the particular
fluid, the compressibility of liquids is taken into account when determining
needed pump inlet and outlet pipe sizes.

See for example, http://www.koken-boring.co.jp/e/e_product/pump/mg/mgpump.htm ,
where a number of pumps and their specifications are shown. Note that, for the
higher pressure pumps, the outlet diameter is always smaller than the inlet
diameter.

On my Honda's power steering pump, I just "miked" the diameters (= measured with
a micrometer), and the outlet is about 0.075 inch smaller in diameter than the
inlet. The difference is visible to the naked eye.

 

Are you funnin' me? ;-)

I have no idea if the expensive brake fluid is superior from a compressibility
standpoint and whether that even has a detectable effect when operating the
brakes.