Honda Power Unit Hardware & Software

[dren]

Interesting Glenntws. That's close to what Mazda's new engines are doing except the rich mixture that's ignited is in the center with the rest of the swirling outer very lean.

[glenntws]

Interesting Glenntws. That's close to what Mazda's new engines are doing except the rich mixture that's ignited is in the center with the rest of the swirling outer very lean.

I think having the rich mixture on the outside allows for picking up oil at the cylinder walls. And of course you want to keep your heat source away from any cylinder walls to reduce heat transfer so having a small amount of rich mixture at the outside to ignite the lean main mixture would allow to greatly reduce heat transfer to the cylinder walls.

[GhostF1]

Interesting Glenntws. That's close to what Mazda's new engines are doing except the rich mixture that's ignited is in the center with the rest of the swirling outer very lean.

I think having the rich mixture on the outside allows for picking up oil at the cylinder walls. And of course you want to keep your heat source away from any cylinder walls to reduce heat transfer so having a small amount of rich mixture at the outside to ignite the lean main mixture would allow to greatly reduce heat transfer to the cylinder walls.

It's good to have you back sir! Congratulations on your endeavours so far over the past year.

As usual, your theory is an interesting one. More so as Honda are claiming to have a new engine by mid year with an advanced combustion step. So what do you think, based on that, they would be changing dramatically if your theory is based on their current design?

[glenntws]

Interesting Glenntws. That's close to what Mazda's new engines are doing except the rich mixture that's ignited is in the center with the rest of the swirling outer very lean.

I think having the rich mixture on the outside allows for picking up oil at the cylinder walls. And of course you want to keep your heat source away from any cylinder walls to reduce heat transfer so having a small amount of rich mixture at the outside to ignite the lean main mixture would allow to greatly reduce heat transfer to the cylinder walls.

It's good to have you back sir! Congratulations on your endeavours so far over the past year.

As usual, your theory is an interesting one. More so as Honda are claiming to have a new engine by mid year with an advanced combustion step. So what do you think, based on that, they would be changing dramatically if your theory is based on their current design?

Thank you very much

That's a good question you ask.. Wazari said their combustion process is the base and it gets refined so I guess the big update will be a reposition of valves, injector and spark plugs to refine the combustion process as well as designing the intake ports and runners as good as possible for promoting high amounts of swirl. Another thing is the exhaust side, which in my concept really has not that good flow characteristics so if they indeed use a similar concept this is a area where you can improve very much because the position of the exhaust valves is extremely concept air-flow wise.

Another part is of course friction reduction, there are all kinds of weird ideas you can do with the bearings for examples... Or even a improved turbine geometry. All these little can sum up to a huge improvement in power. And to me it seems likely that they are doing this right now since they not longer are being forced by McLaren to do everything right NOW but instead have their time to find good points for improvement and make these improvements reliable.

[godlameroso]

I mentioned a few times last year the engine was in egg phase at the beginning of last year, now it's mid adolescent phase. It just started having it's "growth spurt", the muscles are coming in, but it won't be until mid season until it can really compete with the big kids.

[loner]

well glenntws you could be onto something since Hasegawa said last year that Honda power unit don't use TJI

[63l8qrrfy6]

Honda have already said that they use a prechamber. It was debated here not long ago when that japanese magazine came out.

[63l8qrrfy6]

Engine radiator on the left?

I tend to believe MCL has set the position of the cooler first, and Honda will follow the direction. This year clearly opposite

The coolers are a collaboration between the teams and the company they outsource the coolers to.
Most of the teams use the same company. PWR.
I think Mercedes is the only one they do not supply.

That where you work then ?

[godlameroso]

Did he? I also recall him saying at this point everyone has some version of it. Flame jet ignition is probably a better term than TJI. There's two schools of thought when it comes to combustion, swirl, or tumble, and you can't just favor one or the other, both are present during the combustion cycle. Why not focus on the transition between the two, swirl on intake, tumble near TDC.

Regarding exhausts, I've always been curious about where the collector for the manifold is, do you just place it as close to the turbine as possible, is it better to do a 3 into 1 style collector, or a 3 into 2 into 1, or do you just run 6 individual pipes to the turbine? As for the runners themselves what kind of diameter transitions are we talking about? Usually from the flange of the exhaust manifold after the first bend, there's usually a diameter step up, that step up diameter is critical. Also think about almost all exhaust manifolds, we know most exhaust ports on a head curve roughly 90 to 50 degrees depending on valve angle, why do exhaust manifolds always start with another roughly 90 degree bend? What other fluid pipe system has this U bend? Sink and bathroom drain perhaps?

[63l8qrrfy6]

The exhaust pulses reflect at the step rings (there's more than one per runner if you look carefully) and at the collector entry. The first step is tuned at the higher engine speed and the collector at the lower engine speed.
Normally you want to keep the runners separated for as long as possible and you want the pipes to be parallel just before they enter the collector.
Interestingly the step changes on the V6 appear tiny compared to the ones on the V8 (maybe 6+mm dia change) - I wonder why that is.

The ugly bends are mainly just for packaging reasons. The designer is constrained by minimum bend radii, runner lengths and bodywork.

[johnny comelately]

"Interestingly the step changes on the V6 appear tiny compared to the ones on the V8 (maybe 6+mm dia change) - I wonder why that is."

Possibly NA versus turbo which doesnt have the problems so much of the reverse wave

[Wazari]

Hey Guys,

after being a silent reader in this thread for a pretty long time, I decided to throw out my ideas and concepts about the new engine.

After starting by electric-technology study in last august, I have been busy all the time. Instead of having free time between the terms I work at a big automotive supplier in Neuss in Germany. Pretty cool but I also barely have any free time soo I even had to stop my own engine project because there's no time.

Well now to the Honda engine. First of all I think McLaren made a big mistake by leaving Honda. But let's put that aside.

Now, the sound of the new engine in idle is veeery interesting in my opinion because it actually gives out much information about what philosophy Honda uses for combustion (at least that's what I think). After analyzing the idle sound in Audacity for a bit, I came to the conclusion the the firing order has been changed, at least I'm somehow sure of that because the behaviour is very different, much smoother than before, the ignition pattern seems extremely even. I can't post a screenshot because masking all the parts would simply take to long (maybe I will do it later), but it seems to me that the engine is running on 3 cylinders in idle, keeping them firing for 4 turns and then shutting of the ignition for 4 turns. After a while, they seem to swap the active cylinders and do the exact same process (that's why there's a short pause in the sound).

This sounds very crazy but I actually think this is a extremely big hint to the combustion process. First, the swapping of the active cylinders is normal when doing cylinder shutoff because you want to keep the engine warm evenly so that really is no wonder in my opinion. What makes me going crazy is the actual shutoff of the complete engine for 4 turns, because it makes me think that the combustion process is highly dependent on the intake velocity and the overall air velocity in the cylinder.

Sounds weird first but it makes perfect sense. If Honda goes to low on the throttle, the velocity and pressure in the intake is lower than ok for the combustion process they use. That's why they play through 3 cylinders for 2 times with a relatively high amount of throttle (for idle) and then keep them turned of for 2 times to hold a steady rpm at idle.

Why would Honda need to do this?

Well a couple of pages ago I think roon made some interesing sketches promoting the usage of a swirl dependent combustion process. This idea hasn't left me alone. So here's comes my interpretation of a swirl dependent SPCCI-ICE.

- There is NO pre-chamber.
- One injector and one spark plug. The positions of both however is pretty variable in this concept so my arrangement is just for demo purposes.
- The air near the cylinder wall is the main ignition source. A big chunk of lean air in the center of the engine is the main power source.

https://i.imgur.com/NoKgyfi.jpg

Remember that this idea is not worked out very well, it's just a idea that in my opinion fits to the current things known about the engine and the fact that this year's combustion process is very different according to Wazari.

So think about it that way. The intake valves and ports are arranged and designed in a way that promotes a extremely huge amount of swirl on the outer boundaries of the chamber near the walls. The charge in the center of the cylinder of course also achieves a pretty big amount of swirl but the velocity is nowhere near to the velocity of the air at the outer boundaries. Now during intake and compression, the air inside the complete cylinder gets enriched with fuel but to a pretty lean level. This charge would barely ignite on it's own, especially not under high rpm's.

The trick is with a second injection phase very shortly before the ignition point. A very pointed spray of fuel gets' sprayed in the direction of the outside chamber. The amount off added fuel to the center is very low in comparison and with good pre calculations the AF-ratio can still be kept very low in the center. The little amount of fuel raises the charge at the walls above the neccessary AF-ratio for ignition. A spark plug near the wall now ignites this charge which in turn (because of the violent swirl) reaches the complete wall in a very short amount of time. The charge pressure in the center of the cylinder now get's extremely hot and dense because of the fire ring around it.

This extremely high pressure is more than enough to let the charge in the center of the cylinder ignite in a way very similar to HCCI.

With good design and calculation, the amount of richer fuel on the walls can be kept very low, allowing for a overall very lean charge in the cylinder.

Now, why is this so velocity dependent? The problem with this concept is the huge dependency on the swirl. If the velocity and density of charge in the cylinder is not high enough at the cylinder walls, the fire ring will not be established fast enough and many not wanted this could happen. The charge in the center could ignite in a pattern which is not going well with some parts of the head or piston (pressure-wise) or the charge may not even ignite at all because the pressure in the cylinder is simply to low to keep this combustion process alive. Running the engine under very low load would require a very rich charge near the wall and a at least stochiometric charge in the center and the whole efficiency thing is gone.

Also with a too high amount of fuel near the walls you promote the washing out of oil which of course is not good for the engine. A slightly rich fuel mixture near the cylinder wall however would not do so much damage and the huge amount of swirl would even promote the pick-up of oil particles at the wall which could help with the ignition of the outer charge (there's your possible source of oil consumption).


So, these were my ideas for the current engine. Not a very refined sketch from me and surely not thought out to the end but definetly "doable". And: I try to keep up with the conversation a bit more in the next time. I just love this forum and this thread!

Hi Glenn, good to hear from you and I am impressed. You must have spent quite a bit of time think about this. I bolded out a portion of your post which is very astute and critical to achieve the desired combustion characteristics utilizing extremely lean mixtures under load. Are you sure you are the age you say you are?

[glenntws]

Hey Guys,

after being a silent reader in this thread for a pretty long time, I decided to throw out my ideas and concepts about the new engine.

After starting by electric-technology study in last august, I have been busy all the time. Instead of having free time between the terms I work at a big automotive supplier in Neuss in Germany. Pretty cool but I also barely have any free time soo I even had to stop my own engine project because there's no time.

Well now to the Honda engine. First of all I think McLaren made a big mistake by leaving Honda. But let's put that aside.

Now, the sound of the new engine in idle is veeery interesting in my opinion because it actually gives out much information about what philosophy Honda uses for combustion (at least that's what I think). After analyzing the idle sound in Audacity for a bit, I came to the conclusion the the firing order has been changed, at least I'm somehow sure of that because the behaviour is very different, much smoother than before, the ignition pattern seems extremely even. I can't post a screenshot because masking all the parts would simply take to long (maybe I will do it later), but it seems to me that the engine is running on 3 cylinders in idle, keeping them firing for 4 turns and then shutting of the ignition for 4 turns. After a while, they seem to swap the active cylinders and do the exact same process (that's why there's a short pause in the sound).

This sounds very crazy but I actually think this is a extremely big hint to the combustion process. First, the swapping of the active cylinders is normal when doing cylinder shutoff because you want to keep the engine warm evenly so that really is no wonder in my opinion. What makes me going crazy is the actual shutoff of the complete engine for 4 turns, because it makes me think that the combustion process is highly dependent on the intake velocity and the overall air velocity in the cylinder.

Sounds weird first but it makes perfect sense. If Honda goes to low on the throttle, the velocity and pressure in the intake is lower than ok for the combustion process they use. That's why they play through 3 cylinders for 2 times with a relatively high amount of throttle (for idle) and then keep them turned of for 2 times to hold a steady rpm at idle.

Why would Honda need to do this?

Well a couple of pages ago I think roon made some interesing sketches promoting the usage of a swirl dependent combustion process. This idea hasn't left me alone. So here's comes my interpretation of a swirl dependent SPCCI-ICE.

- There is NO pre-chamber.
- One injector and one spark plug. The positions of both however is pretty variable in this concept so my arrangement is just for demo purposes.
- The air near the cylinder wall is the main ignition source. A big chunk of lean air in the center of the engine is the main power source.

https://i.imgur.com/NoKgyfi.jpg

Remember that this idea is not worked out very well, it's just a idea that in my opinion fits to the current things known about the engine and the fact that this year's combustion process is very different according to Wazari.

So think about it that way. The intake valves and ports are arranged and designed in a way that promotes a extremely huge amount of swirl on the outer boundaries of the chamber near the walls. The charge in the center of the cylinder of course also achieves a pretty big amount of swirl but the velocity is nowhere near to the velocity of the air at the outer boundaries. Now during intake and compression, the air inside the complete cylinder gets enriched with fuel but to a pretty lean level. This charge would barely ignite on it's own, especially not under high rpm's.

The trick is with a second injection phase very shortly before the ignition point. A very pointed spray of fuel gets' sprayed in the direction of the outside chamber. The amount off added fuel to the center is very low in comparison and with good pre calculations the AF-ratio can still be kept very low in the center. The little amount of fuel raises the charge at the walls above the neccessary AF-ratio for ignition. A spark plug near the wall now ignites this charge which in turn (because of the violent swirl) reaches the complete wall in a very short amount of time. The charge pressure in the center of the cylinder now get's extremely hot and dense because of the fire ring around it.

This extremely high pressure is more than enough to let the charge in the center of the cylinder ignite in a way very similar to HCCI.

With good design and calculation, the amount of richer fuel on the walls can be kept very low, allowing for a overall very lean charge in the cylinder.

Now, why is this so velocity dependent? The problem with this concept is the huge dependency on the swirl. If the velocity and density of charge in the cylinder is not high enough at the cylinder walls, the fire ring will not be established fast enough and many not wanted this could happen. The charge in the center could ignite in a pattern which is not going well with some parts of the head or piston (pressure-wise) or the charge may not even ignite at all because the pressure in the cylinder is simply to low to keep this combustion process alive. Running the engine under very low load would require a very rich charge near the wall and a at least stochiometric charge in the center and the whole efficiency thing is gone.

Also with a too high amount of fuel near the walls you promote the washing out of oil which of course is not good for the engine. A slightly rich fuel mixture near the cylinder wall however would not do so much damage and the huge amount of swirl would even promote the pick-up of oil particles at the wall which could help with the ignition of the outer charge (there's your possible source of oil consumption).


So, these were my ideas for the current engine. Not a very refined sketch from me and surely not thought out to the end but definetly "doable". And: I try to keep up with the conversation a bit more in the next time. I just love this forum and this thread!

Hi Glenn, good to hear from you and I am impressed. You must have spent quite a bit of time think about this. I bolded out a portion of your post which is very astute and critical to achieve the desired combustion characteristics utilizing extremely lean mixtures under load. Are you sure you are the age you say you are?

Yes I am sir! I'm even writing a test in software development today

[MrPotatoHead]

I tend to believe MCL has set the position of the cooler first, and Honda will follow the direction. This year clearly opposite

The coolers are a collaboration between the teams and the company they outsource the coolers to.
Most of the teams use the same company. PWR.
I think Mercedes is the only one they do not supply.

That where you work then ?

No.
It's common knowledge. They post it on their website.

[ncassi22]

Seems like Glens' idea would be the practical application of fire whirls!



https://phys.org/news/2016-08-newly-blu ... sions.html