Smart questions
Smart answers
Smart people
Join Eng-Tips Forums

Member Login

Remember Me
Forgot Password?
Join Us!

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips now!
  • Talk With Other Members
  • Be Notified Of Responses
    To Your Posts
  • Keyword Search
  • One-Click Access To Your
    Favorite Forums
  • Automated Signatures
    On Your Posts
  • Best Of All, It's Free!

Join Eng-Tips
*Eng-Tips's functionality depends on members receiving e-mail. By joining you are opting in to receive e-mail.

Posting Guidelines

Promoting, selling, recruiting, coursework and thesis posting is forbidden.
Jobs from Indeed

Link To This Forum!

Partner Button
Add Stickiness To Your Site By Linking To This Professionally Managed Technical Forum.
Just copy and paste the
code below into your site.

DKJones96 (Automotive) (OP)
3 Aug 07 18:01
Does anybody here know if someone is working on an engine that has no manifold vacuum?

I have a few things i've gome up with that involve an IC engine with no manifold or chamber vacuum at part throttle. The engine basicly purges the charge from the cylinder again into the intake manifold(against a reed valve to create a bit of pressure to force exhaust out on overlap) until it reaches your 'throttle' point. My way of doing it is infinately variable and completely mechanical and to help with this I've developed a completely mechanical way to have variable compression ranging say from 8-12:1 and also thought of a low pressure direct injection system for it. I plan on building this for a Toyota 5M-GE I6 engine starting in a couple weeks.

I was just wondering what you guys thought about efficiency increases and if anyone is working on this kind of stuff. Sadly...I'm only 22 so most people don't take me seriously. Just wondering what you guy thought.
Fabrico (Automotive)
3 Aug 07 19:21

No manifold vacuum? So what makes fresh air want to rush into the cylinder?

BrianPetersen (Mechanical)
3 Aug 07 21:19
Diesel engines don't have (significant) intake manifold vacuum because they are unthrottled, and most of them these days are turbocharged.

BMW's Valvetronic system operates not entirely unlike what you are describing, except that they are shortening intake lift and duration at part load so that the air never goes into the cylinder in the first place, rather than sucking it in and then pushing it out again. It normally operates unthrottled (i.e. no significant intake vacuum), although there is a throttle only used as an emergency back-up in case the Valvetronic system fails.

Honda's upcoming A-VTEC system will also operate conceptually like this, albeit with a different mechanical arrangement.

These systems are reported to improve part-load fuel consumption about 10 - 15 percent.

Variable compression ratio is ordinarily extremely difficult to do in a way that doesn't create more problems than it solves (generally it screws up the shape of the combustion chamber when it is in low-compression mode) but if you have found some way around that, go nuts with it.

Direct injection systems ordinarily need to operate at a rather *high* pressure in order to get adequate atomization of the fuel. Gasoline-direct-injection systems use a higher pressure than normal outside-the-cylinder systems, but lower than common-rail diesel injection systems.
patprimmer (Publican)
3 Aug 07 23:40
BMWs Valvetronic effectively uses the valve as a throttle, so from a pumping losses point of view it is still throttled.

I think the OP is talking about no change in restriction to airflow into the cylinder at all.

It is not clear to me how power delivery will be controlled.

There will of course be some vacuum at times as the system will not be 100% VE at all times.


eng-tips, by professional engineers for professional engineers
Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips Fora.

Tmoose (Mechanical)
5 Aug 07 20:56
the A-F ratio needs to be between something like 8-1 and 18-1 to be combustible by normal means (spark plug).  At low and part throttle that means there must be a zone of richness near the spark lug to initiate combustion that can then progress to the leaner portions of the chamber.  Stratified charge.

These guys have been working on igniting lean mixtures for 30 years, and report they are now up (down) to 30-1 ratios.
DKJones96 (Automotive) (OP)
5 Aug 07 21:40
The throttle is controlled by closing the intake valve later and later. Like idle might be closing the intake valve 45 degrees BTDC on the compression stroke(never been done so i just have to throw a number out there.

So instead of working against a strong vacuum at idle it pulls a full charge then purges most of it again til it gets to the 'displacement' it needs. It should not only save fuel but smooth out idle greatly on smaller 4 cylinder engines.
Fabrico (Automotive)
6 Aug 07 16:35

Varying engine breathing with valve timing is not new. Using reed valves on 4-strokes is not new. So, what's new?

turbocohen (Automotive)
10 Aug 07 23:59
If a plasma torch ignition system with enough btu's available to light lean gasoline mixtures then you can unthrottle an SI engine.  It has been donie in the lab with over 600MJ and that is enough to fry your weiner if current reaches your body.

Yes it is doable, no it is not practical yet.
hydrae (Mechanical)
18 Aug 07 12:15
it is called the atkin(sp) or miller cycle, they are used in the toyota hybrids (maybe others).  but they do not run well at low rpm's.  hence the use in hydbrids.
GregLocock (Automotive)
18 Aug 07 19:29
There is no reaosn why an Atkinson cycle would not run at low speeds, and to be honest I can't see any particular reason why a Miller wouldn't.

Both over-expanded engines have idle speeds of around the same as that used in Otto engines of the same capacity.


Greg Locock

Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips.

dicer (Automotive)
19 Aug 07 17:09
Like Brian says diesel, though some older ones did throttle the air intake.

And like also mentioned there is always going to be a vaccum, as long as you have a piston in a cylinder drawing in the air. Then add an airfilter etc. Of course unless you are boosting with positive inlet pressure.
DKJones96 (Automotive) (OP)
19 Aug 07 19:59
I understand that there will be a little bit of a vacuum. I guess i was too vauge in either my description and title or i had just assumed that everyone would understand that the only vacuum that won't be there is the one created in the intake manifold for engine throttling. I guess the closest thing to the engine itself would be a miller cycle engine except this one can keep the intake valve open way past what miller cycle does for 'throttling' and those usually have the cam set to the close angle via cam grind. This valvetrain i'm going to build will give you the same cam profile at WOT that any engine nowadays does.
Andreb (Mechanical)
20 Aug 07 18:16
Why pull the extra air into the cylinder if you are just going to push it back out again before closing the intake valve.  Close the intake valve on the intake stroke as soon as the required amount of air is in the cylinder.

I think I read something on some really early gas burning engines that did just that, I think they had separate valves for the air and gas.  They used valve gear something like the Corliss steam engines and avoided the throttle loss in the same way as the Corliss, it takes work to push/pull a gas thru a restricted opening.

Of coarse at 50 RPM a short valve open close event is not all that short compared to what you need at 5000 RPM.
DKJones96 (Automotive) (OP)
20 Aug 07 20:00
Because it takes less power to pump the air twice than it does to work against the vacuum. Especially the larger the engines displacement is.
hemi (Automotive)
21 Aug 07 18:51
Intuitively that seems wrong.  The vacuum will act like a spring, assuming the cylinder sealing is good.  Pumping air back & forth through the open valve is wasted work.
turbocohen (Automotive)
21 Aug 07 23:24
One of the reasons a diesel is more efficient it the low pumping loss.  Pumping a vacuum takes work.  Lean burn is not efficient combustion but the reason it yeilds efficiency gains is the reduced pumping against atmosphere..  Modern electronic throttled/evt engines with 6speed transmisions and better calibrations lug the engine stochiometric at a lower speed so manifold pressures are closer to atmospheric and trapped volume has more tumble introduced to reduce burn time..  less delta P across the piston = less work and lower rpms = less moting friction.
Nowadays some truck calibrations use tricky fuel cutoff or decel fuel and throttle to enhance engine braking while keeping the cat lit but thats another story...
patprimmer (Publican)
21 Aug 07 23:47
Diesels also have about double the expansion ratio to a petrol engine.


eng-tips, by professional engineers for professional engineers
Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips Fora.

GregLocock (Automotive)
22 Aug 07 0:47
Heywood has a nice plot of the efficiency gains from increasing the expansion ratio (eg via Atkinson). That last litle bit doesn't really do you much good, much better to use it to drive an expansion turbine.


Greg Locock

Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips.

globi5 (Mechanical)
22 Aug 07 14:45 expansion turbine driving a compressor, allowing to increase the power output. Therefore decreasing the size of the engine and thus reducing the pumping losses (less displacement = smaller pumping losses). And of course a smaller engine also needs to be driven at higher throttle settings in order to move the same vehicle (= less 'pumping loss time').
Besides weight reduction possibly another reason why turbocharged engines are becoming increasingly popular.

Then again Toyota does use the atkinson/miller cycle engine on its Hybrids - so there's some benefit to it.

One could question why there is currently no miller cycle engine or very few engines with compressors on the market?
(I guess a turbo is small, few moving parts uses an inherently efficient compressor, where its rpm is not fixed to the rpm of the engine its actually not that surprising even considering its lag).

Regarding Diesel: Keep in mind Diesel-fuel is also 15% more dense than gasoline (also adding to the fuel economy of a diesel engine - a factor often forgotten when a Hybrid is compared to a Diesel).

Back to the topic, to what level does EGR and valve overlap reduce pumping losses on a modern engine?
Fabrico (Automotive)
23 Aug 07 17:50

Have to agree with hemi on this one.
As mentioned before...block a fan or vacuum cleaner, and hear it immediatley speed up.

DKJones96 (Automotive) (OP)
23 Aug 07 18:29
Most testing i've done to date involved single cylinder OHV/OHC engines. We tested alot at 50% throttle because that's where most poeple spend time accelerating at in most cars before they let off to 5-25% depending on speed. We drove the engine by electric motor to measure power required to maintain a given rpm we ran 3000.

From our testing the pumping losses from the engine taking in and expelling to 50% charge was less than the losses required to pull against a vacuum for an entire intake stroke and 'spring' back to full atmospheric at 90deg BTDC on the compression stroke (like where we closed the valve fully) before it started compressing the charge. In addition to that, when we added the reed valve in the intake for the engine to partially pump against it created a pressure on the back of the intake valve that worked great to expell any extra combustion gasses out the exhaust valve durring overlap which increased our part-throttle torque numbers alot. The reed valve's benefit, however, is only seen at mid/low rpm part-throttle engine use.

Two engines compared at full throttle there was a bit less power output on the engine with the new valvetrain but the losses are on par with the fact it's running an extra cam with extra valvespring.
turbocohen (Automotive)
23 Aug 07 21:56
Globi5, EGR dilutes the Nitrogen and oxygen from atomsphere with carbon dioxide resulting in more inert gas diluting o2, aka charge dilution.  With less O2 per unit volume the manifold pressure is closer to atmospheric while maintaining a more advantageous air/fuel ratio.  When EGR and lean burn work together very significant efficiency gains are possible, 10-20% in the lab is not unusual and 5-10% is typical on the street.

EGR, whether via external valve or internal, helps reduce pumping losses, reduces peak cylinder temps for Nox control, Helps increase available charge motion in cylinder and in the case of some higher compression engines helps reduce octane requirements.
globi5 (Mechanical)
24 Aug 07 13:29
Thanks turbocohen for the explanation.

Fabrico, no offense, but you can't compare a centrifugal pump (vacuum cleaner) with a piston pump.
A piston driven vacuum cleaner would slow done and not speed up since it has to work against atmospheric pressure and a centrifugal pump simply has to move less air molecules.
wwest (Computer)
27 Sep 07 12:40

What you seem to be doing is trying to apply the Atkinson cycle technique to an engine with less than 4 cylinders or multiples thereof.

With a 4 cylinder engine when you have one piston on a compression stroke and with the intake valve remaining open to "exhaust" some of the charge back into the intake manifold you also have an "opposite" piston on an intake stroke that will, in effect, absorb the charge being "exhausted".

Results being a relatively constant vacuum within the LARGE VOLUME of the intake manifold.

So, with a 4-cylinder Atkinson cycle engine you still have intake throttling, vacuum in the intake manifold, and pumping losses (lowered) associated with same.

The problem you undoubtedly encountered in trying to apply this technique to a single cylinder engine is pulses of reverse airflow, A/F mixture, "spitting" out of the carburator. Made worse on a single cylinder engine by having virtually NO intake manifold VOLUME to act as a pressure vessel, accumulator.

Same as would happen with anything other than a 4 cylinder engine or multiples thereof, no synchronization of airflow into and out of the cylinders resulting in reverse flow "pulses" at the throttle valve.

Which is why Mazda went to the Miller Cycle for their V^ in the Mazda Millenia "S".

You arrived at the obvious solution, use a 2-cycle engine's reed valve at the carburator throat.

As you point out, pressure pulsing is then contained within the intake manifold itself and will actually help "fill" the combustion chamber on the next intake stroke....SuperCharging...??

The one thing you seemed to have missed is raising the mechanical compression ratio to ~13:1 but keeping the delayed closing of the intake valve so the effective compression ratio remains at 10:1.

Then at WOT the two engines would have equal HP/torque output capability.
mishar (Automotive)
28 Sep 07 0:56


Because it takes less power to pump the air twice than it does to work against the vacuum. Especially the larger the engines displacement is.

Not really. Pumping air twice is all lost work and “working against vacuum” is all conservative because you are getting it back in the very next cycle when that vacuum is “pulling” the piston up. Lost work comes just from vacuum decrease because of heating the gas.
wwest (Computer)
28 Sep 07 12:23
An easy example, test, even if only done mentally.

Take an old style bicycle tire air pump and cover up the inlet opening and now try to "prime" the pump by pulling the handle upward.

Pretty hard "travel" after about 50% distance, huh?

That's what it's like, "pumping loses" for a piston on an intake stroke with the throttle butterfly valve virtually closed.

patprimmer (Publican)
28 Sep 07 18:48

When you pull the pump up, try letting the handle go and see what happens on the compression stroke with the butterfly valve closed.


eng-tips, by professional engineers for professional engineers
Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips Fora.

wwest (Computer)
29 Sep 07 9:24
Yes, I agree, the "piston" will spring back.

But just like the upcoming downhill "run" doesn't fully "cover" the energy you used to get to the top in the first place the "spring" doesn't fully return the input energy.
GregLocock (Automotive)
29 Sep 07 20:21
True, but which causes the bigger loss - pumping the air twice, or whatever you lose in the springback process.

Pumping losses in a typical engine are around 4% of the power output from memory, doubling that would be a bad thing.


Greg Locock

Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips.

CCycle (Automotive)
3 Oct 07 10:56
Another take on this is to say, well ok, si engines will have pumping loss.
Lets recover the loss.

Look at US Pat 3180079.  Freeman in '61 was doing this mechanically with an air motor and roots type blowers.
Kawamura et al of Isuzu have several modern patents to this effect.

The general idea is to put a turbine in the intake.  The part load pressure drop is made across the turbine which is coupled to a generator to recover the pumping loss as electrical power.  
When extra power is desired from the engine, the generator can be switched to become a motor which turns the turbine wheel to supercharge the engine.
turbocohen (Automotive)
3 Oct 07 11:21
KKK, Garret or IHI produced a prototype turbo motor-generator a while ago that was shown to aviation buffs years ago.  It produced about 0.3bar for brief bursts and a variable vane throttled the nozzle area.  The concensus was that it was too risky and expensive and a few of the proto's wound up on a car project.  I saw one well over a decade ago on a 1.9 liter benz 4 banger.
DKJones96 (Automotive) (OP)
3 Oct 07 13:13
But you see. Pumping losses are about 4% at what throttle?

At full throttle I can see a 4% loss because it doesn't work against much as there is no vacuum or anything. At part-throttle you have a big vacuum, and while there might be a spring back it's not near as much pull as when the valve was open because the cylinder now has air/fuel in it. You lose usable springback at about 45-60 deg crank angle.
GregLocock (Automotive)
4 Oct 07 3:34

Figure is from memory, it is in Heywood


Greg Locock

Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips.

CCycle (Automotive)
4 Oct 07 9:31
Turbochoen:  Yes, there have been several suggestions of a turbocharger with a motor/generator added to the shaft, US Pat 6449950, 4827170 and many more.  These are interesting since the device can also turbocompound.  Probably the reason they introduced it to the aircraft industry it that turbocompounding is most effective at high engine outputs.  Aircraft engines run near full output most the time.  Automotive engines however, run at low output most the time and are not good candidates for turbocompounding.

Recovery of pumping loss is most effective at low engine outputs hence automotive applications are good candidates.  This same device can also recover pumping loss but there is less pumping loss power than there is exhaust power to be recovered.

DKJones and GregL:
It is hard to determine what they are talking about when pumping loss is expressed as a percent.  I assume that the stated percentage is percent of full power.  However pumping loss is zero at full power.  So the stated 4%  would be 4% of full power but this would be available only at some significantly reduced power level.

If we describe pumping loss as the percentage of power output at the condition producing maximum pumping loss, i.e.    Pumping Loss/Output Power   
This can go over 100% since operating conditions can be chosen where there is more pumping loss than power out of the engine.
wwest (Computer)
4 Oct 07 15:26
Don't forget that one of the advantages of super or turbo charging is having an intercooler downstream to reduce the heat gained from pressurization. If the piston does ALL of the compressing no such advantage is available.
turbocohen (Automotive)
4 Oct 07 18:07
CCycle  I beg to differ that aircraft engines run at near full output most of the time.  They are hammering out thrust at takeoff and moderate load up to altitude then cruise at moderate loads.. some needs to be there in case of aborted landing.

Autos run at variable loads most of the time.
GregLocock (Automotive)
4 Oct 07 20:06

if you think an engine at WOT has no pumping losses then you are sadly mistaken.

And the concept of a pumping loss as an important parameter in an engine that is not at WOT is a bit of a puzzle.


Greg Locock

Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips.

wwest (Computer)
5 Oct 07 9:33
Aircraft engines...

We always ran our T210 at ~75% power level, hour after hour, in level cruise.

300HP x .75 = 225HP

My 911/996 C4 in "level cruise" at 70MPH uses about 40-50HP. 20-25% of WOT.

Pumping losses...

WOT...As the piston moves downward from TDC on an intake cycle it creates a vacuum at the top of the piston. As we all know, are certainly should know, mother nature HATES a vacuum so air immediately begins to rush into the cylinder bore to fill the area just previously displaced by the piston.

Ideally the cylinder area vacated by the piston would be filled fully, all the way up to local atmospheric pressure.

BUT, and this is where K&N makes their money.

If the outside of the intake valve were truly open, fully open, NO RESTRICTIONS, to the atmosphere then the filling of the cylinder bore would be much easier and quicker.

Just like everything else, air molecules have inertia, so it takes a "moment" to get them moving, and at 5000RPM there are not many "moments" available.

So even at WOT "Otto" engines are not exactly "free-breathing"

Now, fully close that WOT and only rely on the IACV for intake airflow.

The pumping losses quickly become huemongus.

The reason an airplane wing "wants" to move upward, WILL move upward, is to fill the vacuum above. For that same reason the piston "wants" to move upward against the "demands" of the connecting rod/crankshaft.
CCycle (Automotive)
5 Oct 07 13:12
Wwest  Good point if your concern is max power out of a certain displacement.  External supercharging with intercooling does have the potential of lowering the work needed to compress the charge.  
But look at this from energy efficiency.  Any heat rejected by the intercooler is a dead loss.  No way in most practical situations of any recovery.  However, if you compress it in one step in the cylinder, it does take more work but the work, or most of it at hopefully, is recovered in the subsequent expansion.

Turbocohen:  Way back when I was flying, cruising speed was considered 67% of full power.  This is over three times the 20% or so that the average auto engine runs at cruising.  The point is that aircraft engines work harder hence are better candidates for turbocompounding.

GregL:  OK  We all realize that there is pumping loss at WOT.  Wwest makes a good point (prev. Post), thanks wwest.  
Si engines are controlled by making them inefficient.
At WOT any attempt to recover pumping loss with a turbine wheel mounted in the intake will result in reduced power.  Yet at low power levels there is significant pumping loss available and it is ok to recover this power since the reduced power level is desired.
turbocohen (Automotive)
5 Oct 07 14:18
CCycle, In response to your statement "Any heat rejected by the intercooler is a dead loss.  No way in most practical situations of any recovery.  However, if you compress it in one step in the cylinder, it does take more work but the work, or most of it at hopefully, is recovered in the subsequent expansion."
Got data to back that up?  Using an intercooler reduces the net exhaust backpressure for a given mass of compressed air. More dense airat a lower pressure is easier to pump into a manifold than hot lower density air.  i.e, If boosting to 10 psi without an intercooler vs boosting to 7 psi with an intercooler (aiming for the same mass flow) then the 7psi system will require less work from the turbine and less backpressure will be working against the piston during the exhaust stroke.  Additionally there will be less residual spent gas in the combustion chamber.  There are other factors as well.

Just my 0.02 dinars, Turbo
patprimmer (Publican)
5 Oct 07 19:46

Why are pumping losses all that different at closed throttle or WOT.

At closed throttle, you have more suction against the piston on the induction stroke, but you have less pressure against it on the compression stroke. +1-1=0

At WOT you have less suction against the piston on the induction stroke, but more pressure against it on the compression stroke. -1+1=0

I agree with turbocohen, there is energy lost to the intercooler, but less manifold pressure is necessary to move the same mass of air into the engine.

If people think that the deceleration felt when lifting off the throttle is proof of increased pumping losses, they certainly need to think through all the other things that are happening, like inertia, friction etc

This engine braking goes away at light throttle not because of a change in pumping losses, but because the engine starts making power by burning enough fuel to overcome pumping and inertia and friction losses in the engine, drive train and chassis.


eng-tips, by professional engineers for professional engineers
Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips Fora.

GregLocock (Automotive)
5 Oct 07 22:31
Sorry, my point was that at part throttle you /want/ a pumping loss, unless you think that the loss past the butterfly is somehow different in quality to the loss around the intake valve.


Greg Locock

Please see FAQ731-376: Forum Policies for tips on how to make the best use of Eng-Tips.

mishar (Automotive)
5 Oct 07 23:34
Something I proved to myself many times: every second step in any evolution or analyses has to be big picture and major lows of physics. Otherwise, in a very few steps everybody is about to reinvent perpetuumobile.

Regarding pumping loses, I think that there is no point to evaluate it on part throttle because that’s only way Otto engine is regulated. If, by any miracle, we succeed to lower them, we would need to introduce more of them immediately in order to lower the power output.

Way to lower them under WOT is to reduce resistance and utilize resonance.

Supercharging just increase those losses because there is more flow trough same configuration.
wwest (Computer)
6 Oct 07 20:50
An "infinitly" variable displacement engine (swashplate? rotating piston group??) would be the ideal method for doing away with engine throttling via limiting intake airflow.
turbocohen (Automotive)
6 Oct 07 21:05
WWest, vario setups galore have one thing in common.  Cost/Risk to manufacture.  The crankshaft will be with us for a while.
A mechatronic piston driving a linear generator seems like a good idea but a fuel cell is even better until you have to deploy it under all of the conditions that the ole crankshaft does everyday.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members!

Back To Forum

Close Box

Join Eng-Tips® Today!

Join your peers on the Internet's largest technical engineering professional community.
It's easy to join and it's free.

Here's Why Members Love Eng-Tips Forums:

Register now while it's still free!

Already a member? Close this window and log in.

Join Us             Close