HCCI Engines - Benefits, Complexity, and Alternatives

[Rx7man]

I just don't quite get all the hubbub about HCCI engines... I understand the benefits of lean burns, high compression ratios, etc, but the requirement to use only compression to ignite just doesn't make sense to me considering the complications, unpredictability, etc required to get it to work... What is wrong with getting to the n'th degree away from compression ignition and still using a spark to ignite the mix at a much more predictable time? You can still have all the variable valve timing etc and all the benefits they provide. I just fail to see where there's a clear advantage.

 

[jack action]

I would say flame propagation. With a single ignition point, the flame must travel to the end of the combustion chamber. With high compression, the combustion chamber may be too small to permit proper flame travel (which is mostly done with the swirl of the mixture). Think of high-compression race engines with a grinded path into their dome pistons to solve that problem:

Having an instantaneous ignition within the entire combustion chamber seems to be a nice advantage.

 

[Rx7man]

multiple spark plugs seems like it would be less complex way to get very close to the same result..

 

[jack action]

But having more than one spark plug won't help igniting lean air-fuel mixture which is one of the advantage sought with HCCI.

 

[Rx7man]

But if you were SO CLOSE to self ignition, wouldn't the spark be enough to set it off reliably?

 

[jack action]

Maybe, I don't know. I'm not rooting for HCCI, I'm just being the devil's advocate.

 

[Rx7man]

Maybe, I don't know. I'm not rooting for HCCI, I'm just being the devil's advocate.

I'm not rooting against them, but being the devil's advocate as well and trying to figure out how it's really beneficial

 

[jack action]

I haven't heard about HCCI in a while. Maybe the researchers found that the advantages were not good enough for the disadvantages?

 

[NTL2009]

This guy explains it pretty well.

 

[Rx7man]

I don't find it a great explanation... it's an introduction for those who don't know what HCCI is, but nothing more... there isn't any explanation of how everything works together or how some of the challenges are dealt with

 

[NTL2009]

I don't find it a great explanation... it's an introduction for those who don't know what HCCI is, but nothing more... there isn't any explanation of how everything works together or how some of the challenges are dealt with

Well, it's a 5 minute video, he talks pretty fast, but you can only cover so much ground.

He actually did talk (briefly) about some means to control the temperature (valve timing, EGR) which sounds like one of the main challenges of HCCI. If you can find a more in-depth explanation, please post.

I don't think I've come across this - but would it make sense to add a spark plug or glow plug to aid in starting, or even as an operating mode for those points on the speed/power curve where HCCI conditions are difficult to maintain? Seems that getting the HCCI efficiency for most of the cycle would still be progress overall.

This may be a bit off topic, and I might start a new thread later, but wouldn't HCCI make more sense in an application where the engine runs at a more limited speed/power range - like an emergency or portable generator, or the power plant for a series hybrid?

 

[RogueOne]

The dual spark plug method works well with high compression engines that are not necessarily lean-burn. It decreases the amount of time needed for heat release and flame propagation. However, it does not protect against detonation. The HCCI scheme allows the engine to run more like a quality-regulated engine rather than quantity-regulated.

Compression ignition exposes the entire fluid in the combustion chamber to the igniting force, which is why it can work well with lean-burn or quality-regulated fuel systems. Spark ignition does not expose the entire fluid in the combustion chamber to the ignition force, because it is concentrated to a small area. There is instability when running lean burn with spark ignition. It leads to knocking or misfires. Dual spark plugs help mitigate that, but that will not outright solve the problem like HCCI will. HCCI has less cycle-to-cycle variability than HCSI when the mixture is lean.

 

[NTL2009]

Can someone explain the reduced pumping (throttle body) losses that he discusses int he video. I didn't understand how that played into it. Or with HCCI, can it run so lean (or even rich under full power?) that you can run with a full open throttle at all times? I guess that is what a diesel does, right?

So air/fuel mixture just isn't that big a deal with compression ignition? Why is air/fuel ratio so critical with a spark ignition, but not compression ignition?

Knocking/pre-ignition was just mentioned, but that would need to be controlled in a compression ignition engine as well, right? Even more so with the higher compression ratios?

 

[RogueOne]

Specifically as the throttling is concerned, I’ll start with a problem definition:

-The engine has to do work to decompress the density of the intake charge. Throttling the intake is how this is done. It takes a lot of work to decompress the air on the intake side, and then exhaust it into the atmosphere.

-Without throttling, the engine would fill the cylinders completely full of air and fuel charge. The engine would be generating way too much power most of the time as it would be unregulated.

-Need a way to reduce that amount of power. There are 2 options:

- Reduce the amount of fuel/air charge coming in (requires throttling) That engine is called “Quantity Regulated”

- Reduce the amount of fuel coming in, but don’t reduce the amount of air (doesn’t require throttling) That engine is called “Quality Regulated”

When you can use quality regulation, the engine does not have to do the work that is required to decompress the intake air. However, this means that the mass air:fuel ratio is going to be very lean at lower loads. You are reducing power by reducing fuel, so that makes a lean burn condition. The advantage to lean burn is that the engine does not have to decompress the intake charge as much, and it also makes for higher pressures in the cylinder during ignition. There is more air in the cylinder, so it has higher pressure when the piston is at TDC. Higher pressure means faster heat release. It also means more prone to auto-ignition.

As to why Compression ignition is used with Quality Regulated engines, see my last post (Post #12) for the basic idea behind it. Also, with most compression ignition engines, direct injection is used. There is not really any knocking or preignition because the fuel is not administered into the combustion chamber until the combustion event is desired.

With HCCI, different method of controlling autoignition is used. To control autoignition, you need to control how much heat and pressure the charge is exposed to. HCCI engines will do this by varying the valve timing to change the effective compression ratio. That is different than the Geometric compression ratio.

 

[jack action]

Can someone explain the reduced pumping (throttle body) losses

If you reduce the power by reducing the amount of fuel (lean) instead of closing the throttle, you reduce the pumping losses.

Why is air/fuel ratio so critical with a spark ignition, but not compression ignition?

With lean or rich mixtures, there is too much chance that a spark will encounter only air molecules or fuel molecules, therefore combustion will not happen. With CI, the heat is the same across the entire fuel mixture, so there bound to be an air-fuel molecule set that will ignite.

 

[Rx7man]

I see it as having may of the same issues as an SI engine... perhaps if you can make it run on a VERY lean mixture like a diesel, you won't have enough heat to cause metalurgic problems, but at some point you're going to want to increase the output and get somewhere near stoichiometric or even richer.. as you go from some very lean 22:1 mixture or so, down to 13:1 to make power, you're going to have to pass through a zone where you're both making power, and have a very hot, lean burn that has a lot of energy which will damage things in short order... A lot of the reason current engines can't run leaner than they do is because of this.

 

[NTL2009]

... as you go from some very lean 22:1 mixture or so, down to 13:1 to make power, you're going to have to pass through a zone where you're both making power, and have a very hot, lean burn that has a lot of energy which will damage things in short order... A lot of the reason current engines can't run leaner than they do is because of this.

But doesn't a diesel have this same issue? It seems to me, any automobile engine would be designed to work within the heat/power constraints to avoid damage. Of course, after applying those constraints, maybe the advantages of HCCI would not be so great, but I don't know if that's the case.

 

[Rx7man]

But doesn't a diesel have this same issue? It seems to me, any automobile engine would be designed to work within the heat/power constraints to avoid damage. Of course, after applying those constraints, maybe the advantages of HCCI would not be so great, but I don't know if that's the case.

A diesel actually pretty much always stays lean of stoich... by the time you get to stoichiometric A/F ratio it smokes too much, and yes, heat is a problem, but everything in a diesel is so big and heavy it can absorb and dissipate the heat a lot better... It's also why they're usually quite rattly because of the increased tolerances needed for heat expansion

 

[RogueOne]

There is a difference in the mixtures between HCCI and DICI. Diesels, as we know them, use Direct Injection. That means that they directly inject the fuel into the combustion chamber while the piston is at/near TDC on the compression stroke. The fuel is immediately auto-ignited when it contacts the oxygen. The fuel is being introduced to the combustion chamber at the same point, which means that it is locally rich. The charge is not Homogenous. It is concentrated near the injection site in the combustion chamber. So what you end up with is a very rich mixture near the injection point. Obviously, the behavior later in the combustion event is different, but it is a different combustion than HCCI. The charge is stratified at the beginning.

 

[NTL2009]

Just getting back to this... Thanks for the previous posts, I have a much better understanding now. Part of my problem was I took the analogy to Diesel engines too far. Now I see that both Diesel and spark have that flame propagation property, and HCCI is different from both of them in that regard.

I'm still a little lost on the lower pumping losses for HCCI versus SI. It was discussed that the intake valve would be timed to adjust the air intake rather than a throttle body, but wouldn't choking it off with the valve be much the same as choking it off with a throttle body? Or is it mostly due to that much leaner mixture, which means you are letting more air in on average, so lower air pumping losses overall?

 

[Rx7man]

I really wondered about that (pumping losses) as well... Perhaps using a Freevalve setup and opening the intake valve briefly, but fully, early in the intake stroke could reduce pumping losses... minimize the vacuum at the beginning of the stroke, close the valve, and then the energy used to decompress the gas in the rest of the intake stroke would largely be recovered on the compression stroke (there'd be a significant vacuum 'sucking' the piston back up)... I still don't quite see it as being a real source of efficiency gain.

 

[RogueOne]

Just getting back to this... Thanks for the previous posts, I have a much better understanding now. Part of my problem was I took the analogy to Diesel engines too far. Now I see that both Diesel and spark have that flame propagation property, and HCCI is different from both of them in that regard.

I'm still a little lost on the lower pumping losses for HCCI versus SI. It was discussed that the intake valve would be timed to adjust the air intake rather than a throttle body, but wouldn't choking it off with the valve be much the same as choking it off with a throttle body? Or is it mostly due to that much leaner mixture, which means you are letting more air in on average, so lower air pumping losses overall?

Taking in more air for a given amount of fuel as two main benefits. 1- The cylinder pressure will be higher leading up to ignition than it would be with an equivalent stoich charge. 2- Reduced pumping losses from throttling at low engine loads.

I don't see the valve timing being used as a primary method for engine speed/load regulation as it plays into HCCI. It will have a similar effect by changing the volumetric efficiency of the engine, but is primarily done to control the point at which the pressure will break the auto-ignition threshold.

And yes, some of the pumping losses are regenerative, especially in engines that modify intake valve closure timing to change load control. That allows more of the decompression work to be localized to the cylinder.

Here is a good read on the subject of pumping losses and how valve closure can be used to minimize the pumping losses. Keep in mind that there are other purposes for different valve closure strategies, such as modifying the effective/dynamic compression ratio. https://www.mechadyne-int.com/refer...ion/part-load-pumping-losses-in-an-si-engine/ They did a nice job with this statement "The work output of an IC engine is indicated by the difference in area contained within the power loop and the pumping loop. As engine output is reduced therefore, the area of the pumping loop becomes closer to that of the power loop and hence the majority of the positive work produced by the engine is being utilised to overcome pumping and frictional losses."