Compound expansion plus water injection to increase fuel efficiency of a Diesel cycle engine? 4

[TSLexi]

I was recently reading the research of Prof Dr-Ing Gerhard Schmitz into compound expansion ICEs, and Mr Bruce Crower into water injection. So I'd like your advice as to whether this is feasible, as I am merely a community college student planning on entering automotive or mechanical engineering.

Most ICEs waste a whole bunch of power by having the exhaust expand into the outside world. And they require massive radiators to keep the parts from getting heat-damaged. And they also require camshafts to open and close the valves, which adds a whole bunch of weight. And Otto cycle engines require spark plugs and have to use lower compression ratios to prevent detonation, which adds weight as well and decreases power.

We can solve this by starting with a basic four stroke Diesel cycle engine. We replace the camshaft with electronically-controlled solenoids to actuate the valves. Then we can dispense with the timing belt as well.

We then add a low pressure cylinder that the exhaust is directed to. This cylinder will be at TDC when the exhaust valves on the combustion cylinder are open, and when it's at BDC, it's contents will be directed into the air. Steam engine designers figured out the concept of compound expansion centuries ago, why has it taken ICE engineers this long?

Finally, the next time the combustion cylinder is at TDC, instead of injecting diesel, we inject distilled water. This will a) provide an extra power stroke, as water expands 1600x when it turns into steam, and b) cool the engine. We could also increase power-to-weight ratio, depending on the weight of water needed to replace the weight of the radiator. Once again, steam engines don't need radiators, as they produce power, they also dissipate heat. You'd just have to ensure ) that the water is very pure, and b) use steam engine lubricant.

And to ensure the engine will always be operating at peak efficiency, we couple it to a continuously-variable transmission.

So we have the benefits of a) the very efficient Diesel cycle which doesn't require spark plugs and produces more torque, b) use electronically-controlled solenoids to actuate the valves, which saves the weight of the camshaft, c) compound expansion to extract extra power from the exhaust, d) steam power, which keeps the engine cool and adds extra power, and e) a CVT, which eliminates the inefficiency involved in gear-shifting.

If we build the engine out of carbon fiber, this engine will be very light and strong, and able to survive the high temperatures involved. And a lighter engine means it can rev higher and have larger cylinders. It's better to produce torque at a higher rpm, because then you can take full advantage of the CVT's gearing to produce optimum power, and there's no replacement for displacement.

So thanks for advising me!

Lexi

 

[GregLocock]

"Most ICEs waste a whole bunch of power by having the exhaust expand into the outside world"
"adds a whole bunch of weight"

Engineers like numbers, not bunches.

"We replace the camshaft with electronically-controlled solenoids to actuate the valves"
Been around for 20 years, never made it into production. Why not?

"why has it taken ICE engineers this long? "
because there is a better way, within the constraints of the design at the time. Exactly how much is this bunch of power, and is it worth increasing engine weight and complexity by 50% to harvest it?

"the next time the combustion cylinder is at TDC, instead of injecting diesel, we inject distilled water."
and wash all the oil off the cylinder walls.


No problem with you asking questions, but try and do some homework first.










Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[tbuelna]

"....the next time the combustion cylinder is at TDC, instead of injecting diesel, we inject distilled water...."

That's a great question. Injecting distilled water into the compressed air charge at TDC will lower the temperature of the air charge through the latent heat effect of the water vaporizing. But how will this add energy to the working fluid that can be recovered to produce a positive output of work? Injecting diesel fuel into the hot air charge results in combustion and a release of thermal energy. Injecting distilled water does not.

 

[TSLexi]

Well, a camshaft weighs about 4 lbs. And unnecessary weight does reduce power-to-weight ratio. The goal is to make the engine as light as possible, and as powerful as possible, and as fuel-efficient as possible.

And you can prevent the oil from being washed off by using Mobil 600W Super Cylinder oil, which is what they use in steam cylinders, as it resists the washing action of the steam.

We get extra power from the steam stroke and compound expansion, but the main gain is an increase in fuel-efficiency and a decrease in operating temperature, which means we can run the engine at higher rpm, as it won't overheat as quickly. It's better to produce torque at higher rpm, as then you can take advantage of gearing.

Koenigsegg is developing an engine with electronically-actuated valves.

 

[TSLexi]

tbuelna, the heat of the compressed air will cause the water to turn to steam, which pushes the piston down.

 

[tbuelna]

tbuelna, the heat of the compressed air will cause the water to turn to steam, which pushes the piston down.

If the compressed air mass contains enough heat energy, it will indeed cause the injected water to vaporize. But how does this add energy to the working fluid?

 

[TSLexi]

The water itself is the working fluid.

 

[TSLexi]

1st stroke: intake of air.
2nd stroke: compression of air.
3rd stroke: injection of diesel and combustion, producing power
4th stroke: exhaust into LP cylinder when it's at TDC, producing power
5th stroke: injection of water (LP cylinder releases exhaust)
6th stroke: steam power stroke
7th stroke: exhaust steam into LP cylinder when it's at TDC, producing power
8th stroke: 1st stroke (LP cylinder releases steam)

So we have 8 strokes, 4 of which produce power.

 

[MikeHalloran]

For more on electric valves, see 'Lucas Helenoid'.
Given the amount of iron and copper that had to be present for magnetoelectrical reasons, an engine set had to weigh a bunch more than a mechanical valve train does, possibly without even including the high current power supply.

Practical steam cars, like the Doble Model E, did indeed have radiators; they were called condensers.

I think Bruce Crower reported that he couldn't limit cylinder corrosion in his SteamOline engine.
Come to think of it, it may have been spalling induced by thermal fatigue from the alternating hot and cold strokes. We may need another century of cylinder development to get that under control, longer if such a thing never goes into real production.

Steam cylinder oil was developed for a particular environment, generally hot and wet. Motor oil was developed for a different environment, generally much hotter and not as wet. SteamODiesel oil will have to perform in a different environment from either, damn hot, pretty cool, and fairly wet. ExxonMobil can't really make it happen until they can buy some crate engines to destroy in testing. ... and the crate engines will need special oil to survive, so there's kind of a chicken and egg problem to be wrestled.


I salute your enthusiasm. Try to hold onto that. Positive Mental Attitude is a wonderful thing; it can catalyze amazing progress, usually by loosening up a money flow.

The people who hang out here are generally responsible for turning that money flow into a bigger money flow. That takes hard science and hard math. ... both so called because they are, hard.

You are not yet well equipped to go bench racing with these guys. Do study hard.






Mike Halloran
Pembroke Pines, FL, USA

 

[TSLexi]

I know. I just wanted some advice on whether this is feasible at all, not advice on how to actually build one. Yet. I'll learn how to do that in MechE school.

I think silicone oil could be used, as it can't dissolve in water and resists heat.

 

[MikeHalloran]

Yeah, silicone oil >resists< heat, but when it decomposes from the heat it can't resist, one of the decomposition products is fine white sand, which doesn't do engines any good.

Hydrocarbons, OTOH, decompose to, among other things, elemental carbon, which is soft enough to not bother an engine, at least not as much as silica does.



Mike Halloran
Pembroke Pines, FL, USA

 

[TSLexi]

And I think spalling can be solved by using carbon fiber. If it works at high altitudes for jet engine blades, it can work here. Jets have freezing cold air that gets compressed into extremely hot air.

You could dispense with the engine oil entirely, and coat the moving parts in hexagonal boron nitride, which has a high melting point (2900C, well above the cylinder temps of 1000C), and is a very slick aerospace lubricant, and can survive oxidizing environments.

 

[SomptingGuy]

Governments throw large amounts of money at companies researching new and novel engine and power-gen concepts (or old ones, dusted down). Most (if any) don't make it into production, but the people working on them probably have a lot of fun and an education, taxpayer funded. The US government seems more keen on this than most, for example:

http://energy.gov/eere/vehicles/vehicle-technologies-office-advanced-combustion-engines

If you read the trade rags, there are always glossy one-pagers boasting of a brand new technology with awesome potential. Some have even built working prototypes.

Spending a few years of ones early career with one of the start-ups that are spending this money may be interesting.

- Steve

 

[dgallup]

I know people that have made careers out of it. Moving from one failed startup to another, rising up the company ladder and salary. The key is to jump just before each crash. Good money if you are good at whitewashing a pig.

----------------------------------------

The Help for this program was created in Windows Help format, which depends on a feature that isn't included in this version of Windows.

 

[MikeHalloran]

Or, you could work for a '___ Research Institute', and whitewash other people's pigs without having to move. You'll probably need at least a Masters degree to get in.

Mike Halloran
Pembroke Pines, FL, USA

 

[TSLexi]

Hmmm...I wonder if using an 20:1:1 diesel:water:methanol emulsion would be better than using a separate water injection stroke, as it will provide the same advantages with less complexity. Lower cylinder temperatures, less diesel use, and more torque, as have three expansion processes going on: diesel combustion, methanol combustion, and water vaporization.

And according to this article, water-in-fuel emulsions decrease emissions, especially NOx and PM:

http://www.hindawi.com/journals/tswj/2014/527472/

So we should set the dynamic CR so it will be high enough to autoignite the methanol, as it has a higher autoignition temperature then diesel, around double. And higher CRs increase efficiency.

 

[IRstuff]

I'm not going to try much to burst your bubble; but there have been gobs of people pondering ICE issues, and if something doesn't exist, there are probably plenty of sound, rational, and practical reasons for not doing it.

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529

 

[BrianPetersen]

Injecting water together with the fuel will just cool down the charge as a whole - and at the worst possible time, the time when you want to be RAISING the temperature of the charge. The heat that it takes to vaporize the water doesn't come from nowhere. Boiling water makes it expand (into steam), but cooling off hot air makes it shrink, and because of the latent heat of vaporization, the shrinking is more than the boiling ... this is not what you want to be doing. The air and the water/steam are in the same chamber. You can't tell the expanding steam to push the piston down while simultaneously telling the shrinking (cooling) air to act like it hadn't cooled off and shrunk (or lost pressure) ...

Water injection CAN be effective in some circumstances but I strongly urge the original poster to understand the thermodynamics. In a normal application of water injection, the water goes in BEFORE the air is compressed, i.e. it goes in during the intake stroke (typically, upstream of the intake valve ... or upstream of a turbocharger that is in front of said intake valve.) In that situation, the latent heat of vaporization comes from the air as it is compressed ... thus reducing the work of compression. By reducing the temperature of the charge at the end of the compression stroke, this also reduces the work of expansion on the following expansion stroke - there is no free lunch. But with a spark-ignition engine, it allows a higher compression ratio to be used without detonation - or higher boost pressure without detonation - and *that* is where the efficiency and/or power boost can come from.

Addressing various other random points that have been made ...

Regarding electronic solenoids for valve actuation ... Know how I know that you have never built an engine? When the cam lobe compresses the valve spring (taking mechanical energy out of the camshaft) that energy is stored up, and when the valve spring expands on the other side of the cam lobe, it sends a good chunk of that energy right back into the camshaft, so there is not as much loss as you might think there would be. If you've built an engine - particularly a single cylinder engine! - you would know this! If you open a valve using a solenoid, it took electrical energy to accelerate that valve. How do you get that energy back?

One of the reasons that electric/electronic valve actuation has never gone into production is that when you build such a system in the real world, the energy that it takes to actuate the valves is excessive.

You CAN, on the other hand, devise a mechanism to vary the timing of valve opening and closing within certain rather wide bounds, and you can even devise a mechanism to vary the amount of valve lift ... and such systems are in production on engines that you can buy today.

You criticize the weight of a camshaft and valve buckets or rocker arms. I can assure you that a set of solenoids to do the same thing, plus the increased size of the alternator that would be required to supply power to them, will weigh more and cost more.

I know about the Koenigsigg system and find it interesting; it may yet change things. It is not relying fully on electric actuation - as I recall, it appears more like pneumatic actuation under solenoid control. Still, "there is no free lunch". They've demonstrated the system; it remains to be seen if the benefits outweigh the downsides - and you can rest assured that there will be downsides that the inventors perhaps are not telling the world about.

The secondary "low pressure cylinder" - the secondary expansion - can be achieved in a more efficient manner - by simply changing the valve timing events so that the effective power stroke is longer than the effective compression stroke. And, this is in production today - a good many of the gasoline hybrid-electric cars already do this. A good many gasoline engines with variable valve timing play with the cam timing to achieve this effect at part throttle. It's better to do it this way than to use a secondary expansion cylinder ... because it avoids the very substantial fluid flow losses that would be associated with transferring the fluid from one cylinder to the other. Current mass-production engines known to use simulation of the Atkinson cycle at part throttle: Honda Civic (base engine, not SI), Fiat 500 (and all the other Fiats and Chryslers that use MultiAir), BMW (all that use the current turbocharged and Valvetronic-equipped 4 cylinder 2.0 litre engine), Toyota Corolla with the "eco" optional engine, and I'm quite sure there are others.

I won't address the materials-compatibility and lubrication issues associated with the water-injection-stroke system for the simple reason that I don't know enough about it, aside from noting that others have touched on this issue. Suffice it to say that "there is no such thing as a free lunch".

I hate continuously-variable transmissions.

Carbon-fiber engine block??

It's great to be enthusiastic ... but it is also very important to be realistic, and to understand that it's often better to make one small advancement at a time (and hopefully succeeding, and in any event, limiting the amount of damage in case it does not succeed) as opposed to trying for a technological moon-shot with an exponentially greater risk of failure. In today's market all it takes is for ONE small aspect of your design to be wrong.

What happens to your fancy distilled-water-injection system when someone leaves the car overnight in -30 C and starts the engine in the morning? (Some diesels are having issues with the water-based DEF NOx control systems because of this)

What happens to your electronic-valve-actuation system if the system voltage is out of spec? What happens if a fuse blows or a wire breaks with the engine running? If you use the Koenigsigg compressed-air-assisted system, what happens if the compressor fails? In any of these situations, does the piston come up and smash into the valve? Oops, that's an expensive warranty claim - or worse, if the sudden loss of power causes an accident ... a number of auto manufacturers are currently in a wee bit of a pickle because of this right now.

Can your carbon-fiber engine block catch fire?

If your engine relies on some super exotic chemical concoction in place of normal engine oil, what happens when your mother dutifully takes the car to the garage on the corner for its 3000 mile oil change as she has always done for the last 30 years (because that's the only thing she knows about car maintenance), and the flunkie with the toolbox fails to read and pay heed to the massive warning sticker that you applied to the engine in the hope that they will use the correct exotic fluid instead of Quaker State 5w30?

 

[TSLexi]

Thanks Brian! Also, I now realize injecting water along with the fuel would be a bad idea. But I was originally talking about injecting water after the exhaust stroke when the cylinder is at TDC, letting it expand, get exhausted, and THEN have the next air intake stroke.

And Atkinson cycles engines sacrifice power density, as it can't take in the air needed to fully combust the fuel. But that can be remedied by adding a supercharger and/or turbocharger (or NOS if you really want it), making it a Miller cycle engine. Cold air is more oxygen-dense then warm air, and the steam strokes will cool the cylinders, cooling the air charge. You'd probably still need a radiator, but it could be much smaller than usual.

And carbon fiber can survive temperatures of up to 1200C. They use it in clutches and brake pads for racing cars.

So thanks for giving me constructive criticism, instead of calling me stupid like a friend of mine did.

 

[BrianPetersen]

"Atkinson cycles sacrifice power density"

Indeed. What do you think the displacement of your secondary expansion cylinder needs to be, relative to the size of the normal conventional power production cylinder? How much contribution to power output do you think that cylinder would add, relative to the increase in size of the engine as a whole? How would this compare to the additional friction? What about the fluid (pumping) losses of that extra-big cylinder? How big would the valves have to be - how much friction would they add?

The hybrid cars (various Toyota and Ford models) that use Atkinson cycle, use the hybrid system to top-up the torque and power output of the engine. There's nothing wrong with that, and you can't argue with the results.

The non-hybrid cars that emulate the Atkinson cycle do so by using VVT so that it uses Atkinson during periods of moderate torque output - which, for most drivers in most cars, accounts for most of the time. They shift the valve timing back to normal when high torque output is demanded. OK so it loses a bit of efficiency during full-torque operation ... but perhaps that's better than dragging along the friction and heat loss of a much-bigger-displacement secondary expansion cylinder ALL the time.

And speaking of which, I suspect that the amount of secondary expansion for maximum benefit might be smaller than you think it is. At a certain point, it becomes not worthwhile because of the extra friction. At part-load operation, ideally, less extra expansion is needed than at full torque. But if the engine is spending most of its life at part load, it's better to optimize for that. It is a trade-off.

Turbocharging and supercharging are no free lunch, either, as I am sure you are aware. A number of manufacturers (*cough* Ford!) seem to be experiencing a rather large discrepancy in fuel consumption between the ideal conditions associated with the various EPA, Transport Canada, NEDC, etc government testing cycles, and the way people actually drive in the real world. My sister has a Ford Escape 1.6 Ecoboost, so this is a situation that we have rather personal awareness of ...