Preventing abnormal combustion

[riff_raff]

I’m not questioning empirical results; but there’ a bit of an anomaly in the energy balance. The same amount of fuel provided PI or DI should provide the same latent heat cooling if both evaporate to the same degree. Both should be very near adiabatic – PI a bit less so. The PI fuel might see somewhat lower pre-TDC temps since it would have already cooled the charge somewhat, but the temp at TDC should be more than adequate to complete the evaporation.
Or the DI air could be cooled 30 degrees more than the PI charge since it arrives hotter yet end up at about the same temp.
A small point but sometimes that’s where the good stuff lives.

That's the crux of the issue. You are absolutely correct any given liquid (or fuel) has an established theoretical value for latent heat of evaporation. This value is the amount of heat energy input required to produce a phase change from liquid to gas. But this total amount of energy required varies based on factors such as the initial temperature of the liquid, the mass of liquid being considered, etc. The reason there would be any discrepancy between the latent heat charge cooling effect of a DI and a PI injection system has to do with how close to the theoretical limit each process gets. In reality, the lower injection pressures, lower ambient air temperatures, and larger fuel droplet sizes with PI means that it does not come as close as DI to achieving the max theoretical latent heat effect.

[WhiteBlue]

In reality, the lower injection pressures, lower ambient air temperatures, and larger fuel droplet sizes with PI means that it does not come as close as DI to achieving the max theoretical latent heat effect.

So we can now agree for F1 purposes that direct injection produces a larger latent heat effect than port injection would do. Just to simplify the outcome of the discussion. The consequence would also be that all things being equal less inter cooling is required for the same power generation which benefits the design of the inter coolers.

[olefud]

It’s really a bit too complicated to ruin a good discussion by stipulating. But I can contribute another theory as to why DI works even if it doesn’t provide more average cooling than PI.

Properly positioned and timed, DI could function in a modified stratified charge mode by injecting a fuel charge cloud in the ignition locus. Thus the initial ignition would be of a fuel-rich and locally cooled mixture even if the average combustion chamber temperature is rather higher. With much of the combustion complete under favorable conditions, the hotter, leaner remaining mixture would also combust in the usual stratified charge manner.

A given hot fuel charge actually yields more work than if cooler. The problem with the hot charge is pre-ignition and/or detonation. The above allows for the best of a hot mixture and a good burn.

[WhiteBlue]

That makes perfect sense to me. The trick is to provide the charge in a way that you can exploit the strategy. I guess the solution will something like spray guided combustion with more geometric tweaks to the injectors, higher pressure, faster actuators and some tweaking of the ignition systems.

[riff_raff]

It’s really a bit too complicated to ruin a good discussion by stipulating......A given hot fuel charge actually yields more work than if cooler. ........

olefud- Indeed! Why ruin a good discussion with stipulations.

Specifically, a hot fuel charge may not yield more work than if cooler......It all depends upon where the heat energy of the "hot" fuel came from. The energy absorbed from the intake air charge by evaporation of a DI fuel spray ends up contributing to the output of the combustion cycle. But if the energy input used to heat the injected fuel comes from some external source, outside of the combustion cycle, then it may not result in a greater yield of work from the available fuel energy.

The same is true with charge air cooling. More effective turbo intercooling can result in more power, but the higher heat rejection will also result in lower thermal efficiency.

[olefud]

It’s really a bit too complicated to ruin a good discussion by stipulating......A given hot fuel charge actually yields more work than if cooler. ........

olefud- Indeed! Why ruin a good discussion with stipulations.

Specifically, a hot fuel charge may not yield more work than if cooler......It all depends upon where the heat energy of the "hot" fuel came from. The energy absorbed from the intake air charge by evaporation of a DI fuel spray ends up contributing to the output of the combustion cycle. But if the energy input used to heat the injected fuel comes from some external source, outside of the combustion cycle, then it may not result in a greater yield of work from the available fuel energy.

The same is true with charge air cooling. More effective turbo intercooling can result in more power, but the higher heat rejection will also result in lower thermal efficiency.

By “given” it was intended to stipulate that the fuel charge ready to light off was identical except for temperature. In practice a hotter charge is less desirable because it requires a higher octane rating to avoid preignition or detonation. But with the right fuel the charge will burn smoothly at a higher temperature and produce a greater power rating. Sans detonation, a hotter charge yields hotter combustion products and greater pressure on the piston.
Of course a cooler charge is usually denser and produces more power, but would not be an otherwise “given” charge. My point was the mostly theoretical one that a hotter charge has more energy and will do more work.

[riff_raff]

........Sans detonation, a hotter charge yields hotter combustion products and greater pressure on the piston.
Of course a cooler charge is usually denser and produces more power, but would not be an otherwise “given” charge. My point was the mostly theoretical one that a hotter charge has more energy and will do more work.......

At the risk of annoying other posters, I will disagree somewhat with your specific comments.

First, higher initial charge temperatures do not necessarily result in higher peak combustion pressures. There are also other factors to consider, such as combustion heat release rate or ignition timing.

Second, a hotter charge may or may not produce more net work output. The most important thing to consider with work output is the amount of net work produced at the crank versus the total amount of potential heat energy added to the process by combustion of the fuel. For example, all other things being equal, a recip engine with lower average cycle temps might be more efficient than one with higher average cycle temps, since the heat transfer losses would be lower.

Regardless, what ultimately matters is how efficient the engine is at moving an air mass into the cylinder, compressing that air mass, mixing fuel with that air mass, combusting that air/fuel mixture, and then extracting the available energy from the combustion gas.

[WhiteBlue]

Regardless, what ultimately matters is how efficient the engine is at moving an air mass into the cylinder, compressing that air mass, mixing fuel with that air mass, combusting that air/fuel mixture, and then extracting the available energy from the combustion gas.

That was nicely defined. But what we ultimately want to know is which particular solutions are going to end up on top under for the next generation of F1 rules of fuel restriction. It is a fundamental game changer and we need to reconsider many things we are used to from basically air limited engines.

What are the winners when it comes to extracting the highest power? How are we going to prevent abnorml combustion under the new rules? So far the discussions seems to be building some consensus that direct injection will have a positive contribution.

How about a specific inter cooling strategy, about the type of injectors, about the ignition system, about the location for spark plugs and injectors and control strategies for the engine and turbo management. Is there anything we can say already now what will be winning design aspects and strategies? That would be interesting.

[olefud]

........Sans detonation, a hotter charge yields hotter combustion products and greater pressure on the piston.
Of course a cooler charge is usually denser and produces more power, but would not be an otherwise “given” charge. My point was the mostly theoretical one that a hotter charge has more energy and will do more work.......

At the risk of annoying other posters, I will disagree somewhat with your specific comments.

First, higher initial charge temperatures do not necessarily result in higher peak combustion pressures. There are also other factors to consider, such as combustion heat release rate or ignition timing.

Second, a hotter charge may or may not produce more net work output. The most important thing to consider with work output is the amount of net work produced at the crank versus the total amount of potential heat energy added to the process by combustion of the fuel. For example, all other things being equal, a recip engine with lower average cycle temps might be more efficient than one with higher average cycle temps, since the heat transfer losses would be lower.

Regardless, what ultimately matters is how efficient the engine is at moving an air mass into the cylinder, compressing that air mass, mixing fuel with that air mass, combusting that air/fuel mixture, and then extracting the available energy from the combustion gas.

Generally I would agree that belaboring a fine point doesn’t gain much. But our issue is one that could be informative as to the new rules and the appropriate amount of intercooling.

My point concerns using DI to provide a localized volume of rich and cool mixture at the point of ignition to start combustion such that the hotter, leaner residual volume also burns. This is in contrast to an otherwise identical fuel charge with a homogeneous temperature that will detonate when ignited. For the latter charge to be workable it would have to be cooled to a lower temp allowing a proper burn.

Under these circumstances the hotter DI mixture will yield a bit more power in that the identical fuel mixtures will provide the same energy but the hotter DI mixture will reach a higher combustion temperature and thus more power since it was hotter preignition.

As I understand it, there is a concern that intercooling may be limited for aero reasons under the new rules. It is only in this context that my suggestion may be worth consideration.

[Tommy Cookers]

the Audi DI (spark ignition) that was endurance raced used a stratified injection with mean mixture strength near stoichiometric
so the core mixture burned faster (than would stoichiomtric) because it was rich
and there was some injection relatively early to reduce (by cooling) the work needed for compression
(this mixture being presumably too lean to detonate when compressed)
then the main injection late

this is the 2014 route, surely ?

[riff_raff]

With an SI engine, I believe you will always produce best power with a stoichiometric homogeneous mixture. So I don't know if a DI stratified charge would be desirable, unless SFC was an issue.

[ringo]

As for intercooling, detonation prevention isn't the reason it's there.
Intercooling is more of a benefit to pre compression. That's my rule of thumb anyway. If there's no compression next in a cycle feel free to heat things up.
If there's a compression stage next, then inter cooling is the way to go. Things need to be as cool as possible between compression stages.

Theoretically if fuel is injected after compression of the air, then it should be injected hot; peak temps should increase after combustion (but as riffraff said, it's all empirical anyway.
Port inject and some DI is before compression is complete, so this favours cool injection i suppose.

[jamsbong]

With regards to DI, it will provide benefits in intake charge density and intake charge temperatures through fuel latent heat effects.

I’m not disagreeing; just wondering why incompressible fuel cooling by evaporation differs when it occurs in the cylinder rather than partially in the intake port –dry flow aside.

Because the heat transfer rate is higher when temperature differences is high. DI compresses the air thus increases the temperature. The injected fuel can then cool the hot air more effectively.
in Port injection, the fuel injects to a stream of (comparatively) cool air which is then compressed and thus temperature rises.

You can think of it as cooling at the wrong place and time.

[jamsbong]

Interesting discussion. Had to catch up and read all the 4 pages.

Regarding the intercooling vs packaging and aero.
Having a tight radiator packaging does not mean it is less efficient. force convection used in removing heat from radiators (i know the name says radia-tor but it is convection heat transfer technically) is a function of mass flow rate, area and temperature difference.
Since proper aero shape is a priority in F1, you won't want a hideous radiator blocking the airflow.
So you can use channels to speed up the airflow going towards the radiator and maximise the mass flow through the radiator.

Think of it as air jets shooting at radiator to cool it.

[olefud]

Interesting discussion. Had to catch up and read all the 4 pages.

Regarding the intercooling vs packaging and aero.
Having a tight radiator packaging does not mean it is less efficient. force convection used in removing heat from radiators (i know the name says radia-tor but it is convection heat transfer technically) is a function of mass flow rate, area and temperature difference.
Since proper aero shape is a priority in F1, you won't want a hideous radiator blocking the airflow.
So you can use channels to speed up the airflow going towards the radiator and maximise the mass flow through the radiator.

Think of it as air jets shooting at radiator to cool it.

There’s also the consideration of laminar vs. turbulent flow through the radiator. Since the fins tend to act as flow straighteners and heated air forms a thicker boundary layer, heat transfer can be diminished. High flow rates and vortex generators into the fin openings help.