Canceling out pumping losses of throttle governed engines
 

Tests have revealed that a vacuum on the crankcase is of negligible benefit of an otherwise stock engine.

At 7 in/hg of vacuum the seals were screaming bloody murder while drawing in air.

This thread can be considered null

They do this with vacuum pumps for race cars... yes, vacuum in the block is a good thing, and it actually provides more than "marginally greater" ring seals.

I'm not entirely sure how much vacuum they use, but you can get belt-driven vacuum pumps from Summit Racing and Jegs for the purpose.

It is illegal in the US.

Source?

www.supersixmotorsports.com/pdf/PCVBypass.pdf

What exactly is a catch can and how does it benefit? [Archive] - The Australian 300zx Owners Association

Oops ignore Australian one lol

Try again - that article had to do with bypassing the PCV system. What the OP is talking about involves using the OEM PCV system with a higher vacuum - i.e. still feeding the crankcase vent into the intake system to be properly recirculated and burned.

Ah ok. I thought he was bypassing it. My mistake.

Honda's R18 has a system to help with that, right?

Pulse and glide solves the problem.

...and then some (no internal engine friction for large chunks of your trip)

Use this in conjunction with your pulse and glide and you've got something.

The vacuum pump increases internal friction by forcing the rings to seat under non-combustion conditions... It reduces pumping losses by evacuating pressure from the crankcase, creating a positive pressure differential between the top of the piston and the bottom of the piston, which makes the power stroke "more effective", since the piston doesn't have to displace a mass of fluid (air) in order to reach BDC, nor does it have to fight against the same to reach TDC.

Also, increased ring seal prevents blow-by, so marginally better efficiency is produced due to less pressure loss into the crankcase.

The power loss due to the increase in friction is inevitably over-ruled by the decrease in crankcase pressure, which aides in more efficient production of power.

pumping losses
 

The 2-stroke was the great hope of the auto industry,as with a power stroke for each rev,and same pressure above and below pistons,engineers hoped to exploit that tech.,and its high power/weight,and volumetric efficiency.They just couldn't lick emissions issues and everyone has pulled the plug.Honda has a "sparkless" gasoline engine with same BSFC as a diesel and is investigating computer controls to allow operation at low loads where an ignition system is required.------- I've been looking at the Navy's diesel-powered torpedo engine which has only half the "pumping",as a peroxide oxidizer is introduced to the combustion-chamber.There is no intake tract as we understand it in atmo engines.

We have talked about pumping losses. Yer pretty much stuck with P&G, smaller engine swap, cylinder deactivation, and high re-gearing.

I kinda wondered about using a system kind of like that, where you spray the liquid oxidizer into the cylinder just before the fuel, so that the oxidizer heats up enough to create combustion diesel style. You could precisely meter the amount of fuel/oxidizer that goes into the engine, so you could control apparent "displacement" (even thought the physical displacement would be the same, you could control how much oxidizer is present, and how much fuel is used in turn.)

Couldn't find anything, hate to cover old ground again for you guys. Any links off the top of your head? Either way around, I'll attempt it when this thing is in a testable state. Thanks

It's OK to re-cover the same information, because there are always newbies and new information might come out. People who complain that "it's already been covered X-long ago" aren't usually interested in actually advancing technology so much as just talking about it, so don't worry if you bring up something that's already been covered awhile ago, especially if you don't understand. Better to explain something 20 times to people than have 20 people mis-explain it to someone else.

Run 10"-12" Hg. At levels significantly higher than that you start getting into issues with seals and oil spray patterns, not to mention that you do need some atmosphere to allow your standard oil pump to work.

You can run lighter tension rings if you can consistently pull 5" or greater. Don't expect an exhaust extraction valve to accomplish this. Your engine may already have light tension rings. Best to check that out on an engine-specific forum with some very sharp techs. Fairly esoteric knowledge.

There are some common electric vacuum pumps that will draw down the engine like this. I think some of the large GMC SUVs used them for brake boosters; large BMW sedans as well. Be sure to use a well designed catch can inline to reduce the presence of oil vapors or neat oil in the pump -- internals probably not designed for that.

I have also been told that some Toyota diesel engines have an alternator with a vacuum pump built in. That would be trick to adapt over.

I would not expect something like this to survive, say, California EPA inspection -- it would fail the visual test immediately if the technician was familiar with the engine or your setup looked too "Home Depot Engineered." Note that the visual inspection does not "care" if the vehicle has the same or better emissions. There is a formal procedure for getting something like this approved.

Yes, they were pretty cool but it would be hard to think of a more dangerous propellant system in the hands of the general public. Maybe some types of rocket fuel. Might significantly help with any overpopulation problems, though. Much Darwin Awards fodder, for sure.
;)

umm....

wouldn't the same vacuum "helping" the down-traveling pistons be a hinderance on the up-traveling pistons?

Vacuum and pressure in the block really doesn't affect the pistons' travel at all, actually...

I say this because the engine is designed to run in a balanced configuration, so that while one piston is going down, another matches it's speed on the upstroke. The result is that all the air in the block gets moved around alot, but the pressure doesn't change (other than in minute variations) inside the block as a result of piston travel.

The primary benefit of this is that when you force something into a vacuum, it just goes easier than if you force it into 2 atmospheres of pressure. I.E. The pistons are affected by the density of the air inside the block, not the pressure. Vacuum has much less density... in fact, if you pull true vacuum, there is no density... that's the definition of vacuum in this case. A complete lack of atmospheric density.

So instead of the piston fighting to move air around (not compress it), it's just going up and down in what is essentially a lack of atmosphere, which is much easier, and accounts for a reduction in pumping losses.

ah, makes perfect sense there - you're just not churning the fluid (air) back and forth between up piston and down piston if there's no air to churn.

Precisely.

Also, the vacuum is trying to pull air down past the compression rings, so they seal better against the cylinder walls, meaning less opportunity for "blow-by" in either direction. Less combustion gasses leaking into the oil, less oil in the combustion chamber, less compresssion loss before combustion, etc.

So basically, it's a combination of small boosts that end up being a decent boost in HP, and, at least in theory, economy.

What about a single? ;)

Actually there has been quite a bit of work done on trying to reduce pumping losses on a single in the motorcycle world.

Also, there is a fairly marked variance in pressure in the crankcase at higher rpms. Enough so that one or even two quarts of oil are actively drawn into the vortex created by the pressure differential. There are also serious differences in the paths the pumped air needs to take. Compare the pattern of a 180 degree straight four to a 120 degree straight six. Probably the boxer configuration is best to minimize pumping losses. Look at some of the designs with knife and fork connecting rods and disk bearings (fairly low rpm engines).

Single - drill a hole in the block and put a bypass filter on it.

The engine in my minivan has been known to pull so much vacuum under load that it "misplaces" the spark plug tube seals at high RPM extended periods. In this case, the PCV system of my engine is a "performance enhancer" in that it's large enough to allow the intake's vacuum to affect internal pressures.

It sucks, literally, because I get oil spray if I start revving the hell out of it though, which is obvious b/c of the smoke in the exhaust.

My old Honda also used a vacuum PCV system, employing a venturi in the intake stream to draw gasses out. Not enough vacuum to pull oil, other than vapor, though.

I can't remember what my racing tractor's engine has, but it will probably end up getting changed, since it's a splash oiling system, and I plan on running it at nearly full throttle most of the time.

But what size hole?

Lots of thought has been devoted to that. You have to avoid having the system do mork work than necessary.

The danger with intake PCV (and exhaust extractor systems) is the variability of the vacuum.

The crankcase of a single can be regarded as an air spring for the piston. It is pretty efficient at any pressure, but with lower windage losses when run at low pressure. To recover throttle plate losses, replace it with a turbine and bypass arrangement, or a variable pitch turbine.

I was joking about the single. I don't work with those too much. Apparently, my joke struck on something that's actually workable, which wasn't my intent.

The one time I worked with a single, it wasn't imperative that it run at high power, and it was a diesel cycle engine. The intake air was drawn through the block, into a small supercharger, then into the cylinder. The exhaust was also routed through the block, to maintain increased combustion temps. It's a project I was working on with a shop teacher (not mine) to demonstrate heat effects on diesel combustion. The engine was only about 30 CC, something he'd made in his days as a casting mill operator, I guess.

What we found out was basically that while it was a good idea to keep the fuel hot and keep the block as hot as possible, the thing DID NOT like hot intake air, at all. The hotter the air got, the less power the engine was capable of producing, and the more fuel it required to produce the same power.

Since RPM remained relatively constant, the shop teacher could test the torque output at that speed by just measuring how much force in opposition it took to make the engine speed vary to the point where it wouldn't pick up speed again. Fairly inaccurate, but it worked for what we were playing with.

We actually found out that to some given extent, it is more efficient to have the coldest air possible going into the engine, and running extremely lean, than to run stoich with hot air. It's been almost 15 years since I was working on this, but if I recall correctly, we made it all the way down to something like 45:1 average AFR before it grenaded, with intake temps as low as 60*F.

The above is true about less windage and better ringseal, but the primary benefits are canceling the forces acting on the pistons themselves.

ON your typical 4 cylinder you'll have one piston continuously exposed to the atmosphere (or lack of), give or take given the intake valve duration, but for simplicity sake suppose the intake valve duration is 180* and omit scavenging, resonances, ect for a moment since they are of a minor influence to the subject.

Imagine each piston in the dead middle of its stroke, and firing order of 1,3,4,2



cyl# ->: ------- 1-----2-----3-----4



90* ----------- INT - COM - EXH - PWR

270*---------- COM - PWR - INT - EXH

450*---------- PWR - EXH - COM - INT

630*---------- EXH - INT - PWR - COM


This is what I sketched up quick. 2 cylinders cancel out the other 2 in the sealed crankcase and the absolute pressure remains the same down there (omitting blowby), BUT the force acting on the top side of the piston is a different story.

Put it to the test: Balance the ambient pressure in the crankcase (if it's vented to an ambient pressure source) by opening the throttle to 100%. Drive at certain speed and at specific landmark, cut the ignition then floor it (this is with a manual trans btw) and at another landmark a few seconds later check your speed. Repeat this again at the same landmarks and same speed, but with the throttle closed. I'll bet your final speed is quite a few mph slower.

So the same thing can be achieved by balancing the absolute pressure in the crankcase to the Intake manifold. The limit is going to be a safety limit at a debatable 10-15 in/hg. Thus, if you cruise at a 15 in/hg of intake manifold pressure, you can have 15 in/hg crankcase pressure.

There is an old but very good discussion on this link:

http://www.eurospares.com/sucker.htm

The problem (with race engines) is the consistency of the vacuum. The genesis of the difficulty will vary depending on whether you have intake induced vacuum or exhaust induced vacuum. Either way will yield an expensive blown engine under the right circumstances or possibly just banned from the track for fogging everyone with oil. Driving at a constant speed and engine rpm is another matter.

Were you holding the fuel injection timing as a constant parameter?

It's not really a problem, that I've ever seen... this is why we have catch cans, oil separators, and vacuum valves (they're like BOV's, but in reverse).

Catch can keeps fluid oil from venting to atmosphere, oil separator keeps vaporized oil from venting to atmosphere, and vacuum valves will automatically close when vacuum reaches a specified level, e.g. 5", if that's what you want.

Simple A/C clutch would stop the pump (if belt driven, as on NHRA engines) from removing atmosphere until the vacuum valve opens back up. Poppet closes, vacuum pump shuts down, poppet opens, vacuum pump clutches up and turns back on. This would keep you closer to your required vacuum level in the block, easily.

Engine speed was constant, and the engine was just a bench motor, so I'm going to have to say yes, we were. The engine wasn't designed to rev up and down, it was designed to hold a constant speed for test purposes.

The only time load was applied was when we were checking for off-load torque, and it was never intended to account for load, since the engine's performance wasn't the issue, other than to test for apparent efficiency gains from hotter or colder air, changing AFRs, etc.

I get to hear about these problems all the time. Catch cans are not all created equal, that's for sure. Oil separators are easily overwhelmed by foamed oil -- this includes dry sump tanks if the system is not designed with enough reserve. Vacuum relief valves obviate the need for the Mad Max clutch -- though that would be cool. Make sure the relief valves draw in filtered air. Try to draw the vacuum from a properly baffled port in the block versus the valve covers. This is not always practical, of course.

My guess is that, with the hot air, the fuel was being injected too early and the engine was working against itself.

I had an idea that a vacuum regulator would do the trick of keeping vacuum at a steady level. The simple addition of a pair of pcv valves should allow blowby gasses to vent and act as a check valve, a variable restrictor will be necessary as well.

Contrary - due to lack of control, the fuel timing was set non-optimal from the beginning, so that at it's normal operating speed, it was being injected at TDC or as close to it as possible, to prevent this very occurrence.

That's part of the reason I keep saying that power wasn't the main point of the experiment, and was only measured as a metric of efficiency per air temp/fuel consumption.

I don't like the idea of drawing vacuum from the valve covers, because if you get enough air moving, it could work against the oil flow dropping back into the block, causing oiling issues and changing the center of gravity of the vehicle (allbeit not by much) as a whole, in upright applications.

I'd rather draw the crankcase vacuum from two vents, one at the back end, one at the front end, both at the highest possible point that would provide the largest distance between the vents and the oil. The vacuum ports in the block should have either wet filters or screens on them to help prevent liquid oil flow, or at least help to vaporize whatever does get out, so the catch cans and separators have a better chance of doing their job effectively.

Anyone know how to get/make dual sided seals? So that you can hold oil in and still hold air out?