Can an HCCI Engine Be Achieved with Pre-Mixed Fuel and Air?

[chhitiz]

if i introduced a mixture of fuel and air which has been circulating in the intake manifold for sometime into the cylinder, let it rest for a brief period and then let it be compressed to a considerably high ratio, would it behave as an HCCI engine?

 

[brewnog]

No. Even if the compression ratio were high enough to provoke auto ignition, you'd have no control over ignition timing. You'd probably destroy the engine in a matter of seconds.

 

[chhitiz]

how could i control the ignition timings?

 

[Topher925]

how could i control the ignition timings?

There is no simple answer to this question. It's actually the main obstacle holding back HCCI technology. In a nutshell you control it with proper valve timing, air/fuel ratio, and all other parameters that are required for typical optimal engine performance (except the spark). In order to accomplish this you need some serious hardware, hydraulic valves, sensors up the wazoo, and a lot of computer hardware to run it. You actually need a control system that can monitor and modify engine function in real time down to less than microsecond intervals, that is in the middle of a stroke at 5,000 rpm. Not an easy task to accomplish.

 

[brewnog]

Indeed.

 

[chhitiz]

There is no simple answer to this question. It's actually the main obstacle holding back HCCI technology. In a nutshell you control it with proper valve timing, air/fuel ratio, and all other parameters that are required for typical optimal engine performance (except the spark). In order to accomplish this you need some serious hardware, hydraulic valves, sensors up the wazoo, and a lot of computer hardware to run it. You actually need a control system that can monitor and modify engine function in real time down to less than microsecond intervals, that is in the middle of a stroke at 5,000 rpm. Not an easy task to accomplish.

what if all the other parameters are constant? now if we go through the process, will it undergo homogeneous charge compression ignition? the process that i suggested keeps repeating over and over again i n the same manner.

 

[brewnog]

How do you propose to keep air fuel ratio absolutely constant over the engine's operating range of load and speed? How about ignition timing? How about valve timing? How about the control of whatever system you've devised to achieve HCCI?

If you could control all the factors affecting combustion, and make it affordable, then you might as well go and show all the automotive OEMs how to do it, because they've been puzzling over it for years.

 

[chhitiz]

How do you propose to keep air fuel ratio absolutely constant over the engine's operating range of load and speed? How about ignition timing? How about valve timing? How about the control of whatever system you've devised to achieve HCCI?

If you could control all the factors affecting combustion, and make it affordable, then you might as well go and show all the automotive OEMs how to do it, because they've been puzzling over it for years.

the air fuel ratio and valve timings are constant because they don't depend on the engine speed. it doesn't have a camshaft. as for ignition timing i guess it might self ignite. that's what i am trying to ask in the first place. as for varying load, i guess my engine would best function as a hybrid.

 

[brewnog]

the air fuel ratio and valve timings are constant because they don't depend on the engine speed. it doesn't have a camshaft. as for ignition timing i guess it might self ignite. that's what i am trying to ask in the first place. as for varying load, i guess my engine would best function as a hybrid.

So if you don't add any clever AFR control, how do you propose to keep air fuel ratio constant over the range of engine speed and load? Since your engine doesn't have a camshaft, how do you propose to control the opening and closing of inlet and exhaust valves at precisely the right time to let precisely the right amount of mixture into the cylinders, and to let precisely the right amount of exhaust gas out of the cylinders at precisely the right time (over the range 0-8,000rpm and 0-100kW, say)?

The key here is that you don't know whether your mixture will self ignite or not. Too low a compression ratio, and it might not self ignite (and your engine won't produce power). Too high a compression ratio, and it might well ignite, but it's bloody difficult trying to get it to ignite at precisely the right moment (too early and you may blow a big chunk out of your combustion chamber). Even if you get your compression ratio perfect, your engine will only be able to operate at one speed, at one load, and even then it won't be tolerant to changes in fuel quality, or air temperature. That's where there's a gap between the engine operating on paper, and being able to build a workable engine using this principle.

The best proposals so far are to use a combination of HCCI (at low speeds and loads) to improve the areas which traditionally lose efficiency in a spark ignition engine, and switch to a stratified cycle for higher speeds and loads which are traditionally much more efficient, and where control becomes more difficult (here, the stratified charge may be ignited by spark or by a pilot injection). Get it to work and you'll be a rich man.

Have you read the Wikipedia article on HCCI? It's a good introduction.

 

[chhitiz]

the air fuel ratio is constant because the amt of air mixing with the fuel is constant at all times.
it can't function at variable speeds anyway, not all by itself anyway. one would have to determine the best possible compression ratio. i don't know what a stratified cycle is. i have read the wiki article on hcci but it still doesn't answer my questions.

 

[brewnog]

The amount of air required by the engine depends upon load and speed, as does the amount of fuel. To say that the AFR is constant because the amount of mixing stays constant is misguided and meaningless. My question remains; how would you contro; the AFR; just ignore it and hope it stays constant? How do you propose to control load?

If we've not answered your questions, then I'm not sure what your questions are. Let's summarise:

if i introduced a mixture of fuel and air which has been circulating in the intake manifold for sometime into the cylinder, let it rest for a brief period and then let it be compressed to a considerably high ratio, would it behave as an HCCI engine?

No, absolutely not.

how could i control the ignition timings?

This is the subject of much current work. Efforts have been made to do this by clever valve timing, variable compression ratio, variable charge temperature, and the use of other cycles (spark ignition or direct injection as well as HCCI).

what if all the other parameters are constant? now if we go through the process, will it undergo homogeneous charge compression ignition?

Possibly, but controlling all those factors (even to keep them 'constant' as you say, which would not work) is incredibly difficult.

Think of it like this. You're compressing a quantity of mixture thousands of times per minute, to a pressure, and hoping it will autoignite at exactly the same part of the cycle every single time. However, the amount of mixture varies from cycle to cycle. The pressure you achieve varies from cycle to cycle. The amount of time for each event varies from cycle to cycle. The temperature of the charge varies from cycle to cycle. The mixture strength varies from cycle to cycle. And despite all this, you're just hoping the ignition timing will remain constant for every cycle? It's really, really difficult.

 

[chhitiz]

Think of it like this. You're compressing a quantity of mixture thousands of times per minute, to a pressure, and hoping it will autoignite at exactly the same part of the cycle every single time. However, the amount of mixture varies from cycle to cycle. The pressure you achieve varies from cycle to cycle. The amount of time for each event varies from cycle to cycle. The temperature of the charge varies from cycle to cycle. The mixture strength varies from cycle to cycle. And despite all this, you're just hoping the ignition timing will remain constant for every cycle? It's really, really difficult.

i don't think that the amount of mixture, mixture strength, or temperature should vary much from cycle to cycle. so can it now autoignite at the same part of the cycle?

 

[Topher925]

i don't think that the amount of mixture, mixture strength, or temperature should vary much from cycle to cycle. so can it now autoignite at the same part of the cycle?

Of course it can. If you have a car driving quickly down a twisting round which will affect engine cooling and intake velocity. Or when a car is driving up and down a hill greatly changing the BEMF. Theres a lot of variation that all needs to be precisely controlled. In engines, nothing ever stays constant without some type of control input.

 

[chhitiz]

If you have a car driving quickly down a twisting round which will affect engine cooling and intake velocity. Or when a car is driving up and down a hill greatly changing the BEMF. Theres a lot of variation that all needs to be precisely controlled. In engines, nothing ever stays constant without some type of control input.

it can't affect the intake velocity, if a camshaft is not used for intake. i don't think cooling would change much. what's BEMF?

 

[brewnog]

i don't think that the amount of mixture, mixture strength, or temperature should vary much from cycle to cycle. so can it now autoignite at the same part of the cycle?

You "don't think" these factors vary much? Then please could you explain why the cylinder pressure COV for a typical conventional SI/CI engine is as high as it is?

it can't affect the intake velocity, if a camshaft is not used for intake. i don't think cooling would change much. what's BEMF?

Why will the cycle to cycle repeatability of intake conditions be vastly improved by replacing a camshaft with a hydraulic/pneumatic/electrical actuator?

 

[mheslep]

...

The best proposals so far are to use a combination of HCCI (at low speeds and loads) to improve the areas which traditionally lose efficiency in a spark ignition engine, and switch to a stratified cycle for higher speeds and loads which are traditionally much more efficient, and where control becomes more difficult (here, the stratified charge may be ignited by spark or by a pilot injection). ...

Seems like a lot of trouble. This variable load problem was already solved with diesel electrics. In this case, run the HCCI at constant load and speed charging a battery, either on/off, period. Then let the battery and electric motor handle the variability in speed/load.

 

[brewnog]

Seems like a lot of trouble. This variable load problem was already solved with diesel electrics. In this case, run the HCCI at constant load and speed charging a battery, either on/off, period. Then let the battery and electric motor handle the variability in speed/load.

It's an easy way around the problem, but overall efficiencies drop significantly.

 

[mheslep]

It's an easy way around the problem, but overall efficiencies drop significantly.

Where? The electric motor is ~90% efficient. If the HCCI is 40% efficient, then the overall is 36% or so. Perhaps another 1-2% for hybrid/battery weight? Still far better than a spark & throttle engine alone (17%).

 

[Topher925]

Where? The electric motor is ~90% efficient. If the HCCI is 40% efficient, then the overall is 36% or so. Perhaps another 1-2% for hybrid/battery weight? Still far better than a spark & throttle engine alone (17%).

No, you have more losses than that. You have generator losses, VSI + other electronics losses, secondary battery losses, etc. Series hybrids (Volt) usually are less efficient than parallel hybrids (Prius) except for short distances. That is the distance available from a fully charged battery. A well designed ICE and electromechanical power train in a series hybrid configuration can be more efficient than just a stand alone ICE. I'm not sure how it compares to a well designed HCCI engine though.

 

[mheslep]

No, you have more losses than that. You have generator losses, VSI + other electronics losses, secondary battery losses, etc. Series hybrids (Volt) usually are less efficient than parallel hybrids (Prius) except for short distances. That is the distance available from a fully charged battery. A well designed ICE and electromechanical power train in a series hybrid configuration can be more efficient than just a stand alone ICE. I'm not sure how it compares to a well designed HCCI engine though.

Well at the moment there is no HCCI that works over a practical range of loads and speeds. As I understand it, the experimental HCCIs fall back to spark and throttle the moment a significant demand change is placed on them. So, until that situation improves it appears HCCI series hybrid is an immediately available solution, that also has the benefit of running all electric (zero fuel) for short commuting distances.

 

[Topher925]

Well at the moment there is no HCCI that works over a practical range of loads and speeds. As I understand it, the experimental HCCIs fall back to spark and throttle the moment a significant demand change is placed on them. So, until that situation improves it appears HCCI series hybrid is an immediately available solution, that also has the benefit of running all electric (zero fuel) for short commuting distances.

I concur, but I wouldn't say immediately available. I believe this is the whole idea behind GM's Volt concept. However, I don't believe it will have a HCCI engine due to high costs and/or lack of technological development. I'm surprised no one has brought up CVTs in this thread yet.

 

[mheslep]

I concur, but I wouldn't say immediately available. I believe this is the whole idea behind GM's Volt concept. However, I don't believe it will have a HCCI engine due to high costs and/or lack of technological development.

That's right, nor is the Volt going to use a diesel, which surprised me initially. GM's Lutz stated they were staying with gasoline because of the difficulty of meeting the new EPA standards (almost zero particulates), cost of diesel, etc.
http://fastlane.gmblogs.com/archives/2007/07/checking_back_w.html
It appears HCCI is a way around most of the diesel drawbacks- lean mix, low combustion temperature, thus low emissions - while keeping the retaining greater efficiency of high compression ratios.

 

[brewnog]

Where? The electric motor is ~90% efficient. If the HCCI is 40% efficient, then the overall is 36% or so. Perhaps another 1-2% for hybrid/battery weight? Still far better than a spark & throttle engine alone (17%).

There are far more sources of inefficiency than that, but I wasn't comparing your quoted solution with a conventional SI engine, I was comparing it with the "HCCI/Stratified Charge" system I described in Post 11.

 

[mheslep]

There are far more sources of inefficiency than that,

Yes, alright, generator, converter, motor ~75%. Anything better than 50% in the electric system and the HCCI+serial electric is still better off than a conventional ICE.

but I wasn't comparing your quoted solution with a conventional SI engine, I was comparing it with the "HCCI/Stratified Charge" system I described in Post 11.

Ok, so where do 'overall efficiencies drop significantly' with HCCI+serial electric?

 

[brewnog]

Yes, alright, generator, converter, motor ~75%. Anything better than 50% in the electric system and the HCCI+serial electric is still better off than a conventional ICE.

Indeed. Again, I wasn't comparing a HCCI/electric hybrid with a conventional SI engine.

Ok, so where do 'overall efficiencies drop significantly' with HCCI+serial electric?

You listed most of them. It's in the electric drive/storage as you know.

 

[AWOC]

how could i control the ignition timings?

Have you guys ever thought of using old diesel engine parts. Their mechanical injector timing methods could be helpful