Honda Power Unit Hardware & Software

[Bandit1216]

Hi

I'm reading this debate with a lot of interest. I don't claim to know what F1 is using, but I have experience tuning the injection and ignition of my own car. It looks like some mix up the point of injection and ignition. On road going cars a 60 - 2 trigger is used by most engine builders. The trigger wheel (mostly on the flywheel) has 60 teeth of with 2 are missing. I guess you know how this works? The missing tooth is mostly between 60 to 90 degrees BTDC. In my engine (PSA TU) it is even 138 deg BTDC. I can choose to inject between 138 deg and about 40 deg BTDC (could be even less, but not logical) by selecting what tooth triggers injection. You only need a camsinc sensor, to determine whether it's at compression or savaging top. This is independent from the ignition timing. I'm sure my system is stone-age compared to F1 software/hardware.

Why shouldn't they inject fuel way before 60 deg BTC? (Are we all talking cranck degrees here?). Direct injection doesn't mean it has to be injected on compression stoke,does it?

Just thinking out load. Flame speed also differs a lot with temperature, lambda value and pressure, right? Could they just as well inject some fuel say 300 deg before BTDC during intake, some more fuel later on, which concentrates closer to the nozzle of the TJI, and close to ignition point the last bit of fuel will be injected, which will be higher lambda within the TJI "pre-chamber". They could "play" with this for hours on the test bed.

In this way:

The rich fuel will ignite easily within the "pre-chamber" with low flame speed, but in very close quarters.
Ones the (flame) jets come out the TJI, the fuel will be leaner and flame speed will increase quite substantially.
The lean fuel mixture has a lot more time to develop (300 deg minus ignition advance)
During intake the fuel will be mixed well because of the turbulent the air entering with 3 bar pressure.
Ignition advance stays low, because the flame speed is high (overall).

Also, when all 4 use TJI, why keep it a secret to each other that they do?

[hurril]

How would you initiate combustion inside the pre chamber if the injector has to be outside of said chamber? There is a technical appendix to the regulations that specifies which types of injectors are allowed. What if in this appendix it is stated that the kind of multi stage injector needed to have a pre-chamber is not allowed?

We know the injectors are side mounted on the exhaust valve side on the Honda engine, and that the coil pack for the spark plug indicates it's mounted in a conventional way.
The most plausible method of having any kind of flame jet ignition with the current regulations is with the shrouded spark plug type.

When this has been discussed, it's been suggested that the mixture "compresses into" the pre-chamber. I, myself, struggle a bit with this and also with how it's cleared well enough post ignition. But I certainly want to believe/ understand it.

At the beginning of the compression stroke the density in the cylinder (including the pre-chamber) is about 3 times atmospheric. At the end of the compression stroke the density in the combustion chamber (including the pre-chamber) is about 15 times that i.e. 45 times atmospheric. (assuming 15:1 CR). This means that 14/15 ths (93%) of the contents of the pre-chamber have flowed into it during the compression stroke and almost all of this 93% is fresh charge. The very worst case scenario where all the exhaust remaining in the pre-chamber at the end of the exhaust stroke stays there throughout the intake stroke, would still only allow 7% residual exhaust in the pre-chamber.

Right and this is a very good explanation. Why or how, though, would the mixture become denser within the shroud/ pre-chamber than it is without it? I take it this has to entail some vert intricate injector spray pattern and timing combined with some very precise fluid dynamics; because for a denser mixture to end up inside, it has to first form outside the pre-chamber.

If the mixture is not denser within the pre-chamber, then what does it buy? Is there more to having a shrouding or containment than its effect on the mixture? Does containment of the primary ignition event improve its triggering effect in a way that makes it ignite a homogenous mixture better?

[saviour stivala]

Information on this page:-“Mahle-powertrain”. In particular the last sentence in the paragraph. “Inside the igniter assembly is a small ignition chamber with a direct injector (DI) that provides a small amount of auxiliary fuel (<5% of the total system fuel) and a spark-plug to ignite the charge. The ignition chamber is connected to the main chamber by a number of orifices which allow jets of partially combusted products to ignite the main charge. The smaller orifices size jets causes turbulence in the hot gas jets which then penetrates deeper into the main combustion chamber and causes an evenly distributed ignition effect. The main chamber is fueled through a conventional port or direct-injection injector”.

[henry]

When this has been discussed, it's been suggested that the mixture "compresses into" the pre-chamber. I, myself, struggle a bit with this and also with how it's cleared well enough post ignition. But I certainly want to believe/ understand it.

At the beginning of the compression stroke the density in the cylinder (including the pre-chamber) is about 3 times atmospheric. At the end of the compression stroke the density in the combustion chamber (including the pre-chamber) is about 15 times that i.e. 45 times atmospheric. (assuming 15:1 CR). This means that 14/15 ths (93%) of the contents of the pre-chamber have flowed into it during the compression stroke and almost all of this 93% is fresh charge. The very worst case scenario where all the exhaust remaining in the pre-chamber at the end of the exhaust stroke stays there throughout the intake stroke, would still only allow 7% residual exhaust in the pre-chamber.

Right and this is a very good explanation. Why or how, though, would the mixture become denser within the shroud/ pre-chamber than it is without it? I take it this has to entail some vert intricate injector spray pattern and timing combined with some very precise fluid dynamics; because for a denser mixture to end up inside, it has to first form outside the pre-chamber.

If the mixture is not denser within the pre-chamber, then what does it buy? Is there more to having a shrouding or containment than its effect on the mixture? Does containment of the primary ignition event improve its triggering effect in a way that makes it ignite a homogenous mixture better?

So maybe they use fluidic diodes? This brief description http://www.fdx.de/en/products/fluidic-diode/ is for a device for gas turbines. So similar environment. Laminar flow in turbulent out, seems to match well with the needs to Turbulent Jet Ignition. Maybe something as simple as a bell mouth in and sharp edge out on the orifices in the spark plug shroud.

[hurril]

So maybe they use fluidic diodes? This brief description http://www.fdx.de/en/products/fluidic-diode/ is for a device for gas turbines. So similar environment. Laminar flow in turbulent out, seems to match well with the needs to Turbulent Jet Ignition. Maybe something as simple as a bell mouth in and sharp edge out on the orifices in the spark plug shroud.

Interesting idea! The mixture "wants" to gather inside the shrouding.

[Dr. Acula]

Right and this is a very good explanation. Why or how, though, would the mixture become denser within the shroud/ pre-chamber than it is without it? I take it this has to entail some vert intricate injector spray pattern and timing combined with some very precise fluid dynamics; because for a denser mixture to end up inside, it has to first form outside the pre-chamber.

If the mixture is not denser within the pre-chamber, then what does it buy? Is there more to having a shrouding or containment than its effect on the mixture? Does containment of the primary ignition event improve its triggering effect in a way that makes it ignite a homogenous mixture better?

Well it think there is a very good reason for a shroud.
As i wrote already, flame propagation speed can be increased massivly by turbulence which can help to get more power out of the same amount of fuel because you end up with a higher MEP.
So the ideal thing to have in the main chamber with the very lean mixture is a rather turbulent flow. But the problem as you may know with a lean mixture is that spark plugs are notoriously bad at igniting it. You can only go so lean with a conventional spark ignition unless you use a stratified charge where you have a richer A/F ratio around the spark.
But with increased turbulence it will also be increasingly difficult to create the proper stratified charge needed because the turbulence disturbs the wanted inhomogenous distribution and you can't simply inject the fuel extremly late because the extra fuel around the spark plug also needs enough time to properly evaporate.
Now with a shroud you can artificialy create a far less turbulent area around the spark plug which should help to keep the richer mixture in this area.
At least that's what i think why they do it this way.

[hurril]

Well it think there is a very good reason for a shroud.
As i wrote already, flame propagation speed can be increased massivly by turbulence which can help to get more power out of the same amount of fuel because you end up with a higher MEP.
So the ideal thing to have in the main chamber with the very lean mixture is a rather turbulent flow. But the problem as you may know with a lean mixture is that spark plugs are notoriously bad at igniting it. You can only go so lean with a conventional spark ignition unless you use a stratified charge where you have a richer A/F ratio around the spark.
But with increased turbulence it will also be increasingly difficult to create the proper stratified charge needed because the turbulence disturbs the wanted inhomogenous distribution and you can't simply inject the fuel extremly late because the extra fuel around the spark plug also needs enough time to properly evaporate.
Now with a shroud you can artificialy create a far less turbulent area around the spark plug which should help to keep the richer mixture in this area.
At least that's what i think why they do it this way.

I like this explanation!

[dren]

So maybe they use fluidic diodes? This brief description http://www.fdx.de/en/products/fluidic-diode/ is for a device for gas turbines. So similar environment. Laminar flow in turbulent out, seems to match well with the needs to Turbulent Jet Ignition. Maybe something as simple as a bell mouth in and sharp edge out on the orifices in the spark plug shroud.

Nice find henry, certainly a possibility.

[godlameroso]

So maybe they use fluidic diodes? This brief description http://www.fdx.de/en/products/fluidic-diode/ is for a device for gas turbines. So similar environment. Laminar flow in turbulent out, seems to match well with the needs to Turbulent Jet Ignition. Maybe something as simple as a bell mouth in and sharp edge out on the orifices in the spark plug shroud.

Interesting idea! The mixture "wants" to gather inside the shrouding.

The shrouding itself has acoustic properties, that in conjunction with the available charge in the chamber, and propagation via spark creates very favorable conditions for combustion, more so than in the main chamber. Even if total CC mixture is roughly homogenously lean, the chamber could still be effective at initiating combustion for a variety of reasons.

[godlameroso]

Hi

I'm reading this debate with a lot of interest. I don't claim to know what F1 is using, but I have experience tuning the injection and ignition of my own car. It looks like some mix up the point of injection and ignition. On road going cars a 60 - 2 trigger is used by most engine builders. The trigger wheel (mostly on the flywheel) has 60 teeth of with 2 are missing. I guess you know how this works? The missing tooth is mostly between 60 to 90 degrees BTDC. In my engine (PSA TU) it is even 138 deg BTDC. I can choose to inject between 138 deg and about 40 deg BTDC (could be even less, but not logical) by selecting what tooth triggers injection. You only need a camsinc sensor, to determine whether it's at compression or savaging top. This is independent from the ignition timing. I'm sure my system is stone-age compared to F1 software/hardware.

Why shouldn't they inject fuel way before 60 deg BTC? (Are we all talking cranck degrees here?). Direct injection doesn't mean it has to be injected on compression stoke,does it?

Just thinking out load. Flame speed also differs a lot with temperature, lambda value and pressure, right? Could they just as well inject some fuel say 300 deg before BTDC during intake, some more fuel later on, which concentrates closer to the nozzle of the TJI, and close to ignition point the last bit of fuel will be injected, which will be higher lambda within the TJI "pre-chamber". They could "play" with this for hours on the test bed.

In this way:

The rich fuel will ignite easily within the "pre-chamber" with low flame speed, but in very close quarters.
Ones the (flame) jets come out the TJI, the fuel will be leaner and flame speed will increase quite substantially.
The lean fuel mixture has a lot more time to develop (300 deg minus ignition advance)
During intake the fuel will be mixed well because of the turbulent the air entering with 3 bar pressure.
Ignition advance stays low, because the flame speed is high (overall).

Also, when all 4 use TJI, why keep it a secret to each other that they do?

I apologize to me this is significant from a valve timing perspective, but in my dyslexia I didn't make that obvious in any way.

[hurril]

So maybe they use fluidic diodes? This brief description http://www.fdx.de/en/products/fluidic-diode/ is for a device for gas turbines. So similar environment. Laminar flow in turbulent out, seems to match well with the needs to Turbulent Jet Ignition. Maybe something as simple as a bell mouth in and sharp edge out on the orifices in the spark plug shroud.

Interesting idea! The mixture "wants" to gather inside the shrouding.

The shrouding itself has acoustic properties, that in conjunction with the available charge in the chamber, and propagation via spark creates very favorable conditions for combustion, more so than in the main chamber. Even if total CC mixture is roughly homogenously lean, the chamber could still be effective at initiating combustion for a variety of reasons.

This is awesome. I think I'm getting this now and I also (now) think that this is highly likely to be what they're doing. It seems awefully sensitive (or delicate, perhaps?) though.

[tcooper27]

My theory= a supposition of a system of ideas based on general principles, cannot be any more and cannot be any less. as long as what is being used by the respective four is such a highly guarded secret. maybe when time passes and this present combustion development is Superseded by a more advanced development we will get not only to know but also see with our own eyes like what happened with the “write-up” permission of the FERRARI 2000 winning car publication, the first of its kind in such a secretive technical matters.

So you have no theory that you think is good enough to contribute but you're 100% sure they're not using TJI because a third party translator doesn't know anything about a specific engineer at Honda?

Do you have any experience in prototype engine development or advanced combustion technologies to back up your claims? Any real evidence that they're using something other than what they're publishing? Any reason to believe Honda is lying to all of us through these publications? Anything at all?

TJI as used by Mahle with the injector and spark plug in one housing isn't being used, HCCI isn't being used. Those are his only claims, which don't require much to verify. Now whether his claim that a diesel style pre-chamber isn't being used, probably not too far out there. There's plenty of ways to skin a cat(ie improve the speed and power of combustion), fuel chemistry being a biggie.

Yes, Mahle's commercial TJI system and HCCI were ruled out 300 pages ago. There are more ways to implement TJI than Mahle's published approach. Even Mahle itself has developed another way to achieve it in a way that complies with F1 rules and have been running it in the Ferrari since 2015.

Straight from the mouth of Mahle's VP:
More than five years of development preceded that crucial phone call to Maranello with the proposal: “We’ve found an interesting new solution for you.” A Mahle press release explains. “Within a few weeks, in spring 2015, we adapted our solution to the Formula 1 requirements, allowing Ferrari to compete in Canada with this solution for the first time,” recollects Fred Türk. Vice President MAHLE Motorsport.

Not only is TJI getting widespread use in F1, it's also tickling down to other series. In GT500, Honda, Nissan, and Lexus all use TJI in their engines.

To say that TJI is not in use in F1 or not applicable to race engines is ridiculous.

[1158]

The regs allow up to 5 injection pulses per combustion cycle, right?

Way out there idea, but would it be possible to have different nozzles that allowed for different spray patterns or even different spray patterns from the same nozzle. One of the spray patterns directs the fuel at an opening to the prechamber?

That would be pretty complicated and sounded better in my head than it did once I typed it out...

[godlameroso]

5 sparks per cycle, unlimited injection pulses.

[hurril]

5 sparks per cycle, unlimited injection pulses.

What's a good fantasy or idea about why there's a limit like that?