Passive Mahle Jet Ignition - potential retrofit opportunity?

[joshuaho96]

This isn't really an RB or Nissan thing but back in early April Mahle published a paper for the SAE WCX digital summit on their project with passive turbulent jet ignition. Active TJI has been used in F1 for a while now due to the fuel flow/refueling limits in those races. Passive TJI is kind of like active TJI, except instead of the complication of having a direct injection mechanism inside of a prechamber with a spark plug the passive system relies on the compression stroke to force fuel/air mixture into the prechamber, either via a main chamber direct injector or a port fuel injector. The disadvantage of this setup is that it doesn't allow for any lean burn tricks like the active version, but the advantage is that Mahle is claiming they've been able to get this passive variant to fit in the same footprint as existing M12 spark plugs:

 

They include a nice photo of what a production variant might look like:

HKS' prototype for comparison, which may or may not actually be functional:

The primary driver for this research right now is the ever-tightening noose of RDE conformance (step 1,2,3, etc) which is making all the ICE manufacturers scramble to come up with solutions for how they're going to get a gasoline engine to never enrich under high load which is going to cause them to blow past the PM/PN and HC/CO limits set by Euro6d. Introductions aside, the real question is what Mahle achieved. This research was done against their 1.5L I3 testbed engine which achieved a peak of ~37% thermal efficiency and has all kinds of tricks up its sleeve like GDI, integrated exhaust manifold for fast warm-up, dual wide angle (60 crank degree of traverse) VVT, DOHC 4V head, 83mm bore, 92.2mm stroke, electronic wastegate on the turbo, etc. This is broadly kind of similar to the Honda L15B7 which also achieves 37-38% peak brake thermal efficiency in the sweet spot but tails off to pretty mediocre levels at WOT in the 5000-6000 RPM range, maybe 25% due to the boost enrichment: https://www.epa.gov/sites/production/files/2018-10/documents/sae-paper-2018-01-0319.pdf

To convert this engine to MJI in a kind of drop-in application they changed the CR from 9.25 to 9.1, changed the engine to port injection only, added a low pressure cooled EGR system, and changed the turbo to a higher flow turbine due to the higher mass flow rate through the turbine from EGR. Cams remained pretty standard stuff, 246 degree duration for both intake and exhaust. One problem they noted here was that their first iteration of the spark plug had too much pre-chamber volume which led to extreme combustion rates, over 6 bar per crank degree. They ran a few different "shoot-out" experiments comparing this 9.1 CR engine with their passive TJI system to the same engine with a central spark plug to try and eliminate other variables.

These tests had limits set like stoichiometric AFR for the full sweep, 98 RON pump gas, <950C EGT pre-turbine, compressor outlet IAT <180C, and <6 bar/degree pressure increase. The main takeaway here seems to be that within those constrains they were able to get more timing advance in the sense that the burn rate of the air/fuel mix is fast enough that even the same ignition timing results in better combustion phasing, with the 50% mean fraction burned point getting closer to 10 degrees after TDC which seems to be what they settled on as MBT for this engine. If you look at the the ~80 kW/L load point at 5000 RPM it's almost 5 extra degrees of timing. The big caveat here is that maximum cylinder pressure is at least 20% higher than the central spark plug which the bottom-end has to be designed for. The engine also needs a pretty tremendous amount of EGR to not run boost enrichment:

I think this chart sums it up pretty well, the engine used in this fuel vs timing/EGT/etc chart is a different one they made for a different paper (miller cycle, 172 degree intake cam, 14.7 CR) shows that ~90 RON with passive TJI gives you better timing than 95 RON with a normal plug, and 95 RON gives you better timing than 99 RON with a normal spark plug. The big question in my mind is just how relevant this is to an old engine like an RB26. It sounds like cooled external EGR is a pretty central part of being able to realize full stoichiometric operation from these results, what kind of results would result from VVT-based internal EGR?

[joshuaho96]

Some kind of coincidence always seems to occur when I ask questions, it turns out that Mahle did study the delta between MJI vs MJI + LP-EGR:

 

Maybe this is a lot more doable than it seems at first glance? The problem as mentioned before is that EGR is necessary at high load to control the rate at which the mixture burns to keep things within what a reasonable bottom-end can handle. 

[Ben C34]

No one is going to do that mate. Not for at least 20 years

[GTSBoy]

Yeah, I rubbished the idea last time the HKS project for this was mentioned (a couple of months ago). Not so much because it is a bad idea. Just because there seems to be far too much other crap beyond just a "retrofit sparkplug" required to make an old engine** work this way.

 

**Where "old engine" was specifically an RB26. But really this has to be true for almost everything.

[joshuaho96]

8 hours ago, Ben C34 said:

No one is going to do that mate. Not for at least 20 years

It's been in development for like 10 years, half the paper is discussing their efforts to get this thing ready for mass production, there's a lot of "boring" stuff in there like warm idle control authority via spark timing, cold start catalyst heating, cold start combustion stability down to -8C, etc. They're also claiming that the impact of main chamber GDI vs PFI with this ignition method is reduced because there's a lot less time for spark knock to develop. I think OEMs are probably pretty interested in the idea if only because it reduces their costs/complication by eliminating GDI in the passive variants.

6 hours ago, GTSBoy said:

Yeah, I rubbished the idea last time the HKS project for this was mentioned (a couple of months ago). Not so much because it is a bad idea. Just because there seems to be far too much other crap beyond just a "retrofit sparkplug" required to make an old engine** work this way.

 

**Where "old engine" was specifically an RB26. But really this has to be true for almost everything.

I get the distinct impression that Mahle is trying to sell this to OEMs as a cheap retrofit, there's also mention of controlling the pre-chamber volume to get to the desired target pressure rise rate which sounds like it may help in cases where you can't just retrofit EGR or an EGR cooler. I think the real deal-killer is likely to be working around an extra 30% in peak cylinder pressure, it'd be interesting to know what kinds of bmep the RBs run these days and what kinds of peak cylinder pressures. This kind of improvement is pretty substantial IMO:

The question I really have is more about VVT than anything else. When I look at the packaging of this engine in these cars I really don't see a viable method of recirculating exhaust gases externally. Is there some method of using VVT at high load to achieve internal EGR without adverse knock-on effects? I guess you could crank the boost?

[GTSBoy]

4 hours ago, joshuaho96 said:

 Is there some method of using VVT at high load to achieve internal EGR without adverse knock-on effects? I guess you could crank the boost?

I think you'd need to go camless Koenigsegg style valve actuation. So you can pop the valve open when you want it, rather than just move the whole normal valve event forward and backward wrt the normal lobe centre.

It really is last gasp stuff though, because the petrol engine is going to stop making sense sometime real soon.

[joshuaho96]

12 minutes ago, GTSBoy said:

I think you'd need to go camless Koenigsegg style valve actuation. So you can pop the valve open when you want it, rather than just move the whole normal valve event forward and backward wrt the normal lobe centre.

It really is last gasp stuff though, because the petrol engine is going to stop making sense sometime real soon.

Yeah, I'm pretty sure this is the last gasp. I'm really only interested in looking into this because the compromises involved in E85 (only a few stations, major rework of the fueling system end to end, corrosive, significant range reductions) and water injection (reliability, oil contamination/piston wall scuffing concerns, finding space for a water tank + water lines, maintenance requirements, actually sourcing injectors that work well, etc). Passive TJI seems to be a bit simpler in that the fueling doesn't need to adapt, it seems like maybe it won't be a guaranteed silver bullet that allows for lambda 1 everywhere all the time in retrofit applications but it would allow for better timing + more lambda 1 operation.

[Ben C34]

36 minutes ago, joshuaho96 said:

Yeah, I'm pretty sure this is the last gasp. I'm really only interested in looking into this because the compromises involved in E85 (only a few stations, major rework of the fueling system end to end, corrosive, significant range reductions) and water injection (reliability, oil contamination/piston wall scuffing concerns, finding space for a water tank + water lines, maintenance requirements, actually sourcing injectors that work well, etc). Passive TJI seems to be a bit simpler in that the fueling doesn't need to adapt, it seems like maybe it won't be a guaranteed silver bullet that allows for lambda 1 everywhere all the time in retrofit applications but it would allow for better timing + more lambda 1 operation.

How much do you drive for any of those concerns to outweigh the hassle?

 

Or is this a theoretical discussion?

[joshuaho96]

1 hour ago, Ben C34 said:

How much do you drive for any of those concerns to outweigh the hassle?

 

Or is this a theoretical discussion?

I think the problem is more like because I don't drive the car very much E85 becomes a problem. It's about to do a ~550 km road trip down to the CA grey market emissions lab though, and I plan on doing some road trips like that out to the local tracks for some driving instruction in the future. Water injection seems far too fraught with landmines for me, far too many stories of injector problems and other headaches to really want to deal with it. As far as emissions and fuel economy goes I understand that a car that doesn't get driven very much doesn't really need to get good fuel economy or good emissions but it would be nice to improve those as well as achieving more power output than I otherwise would on ~96 RON CA gasoline. It pains me to think about how easily the engine gets into boost enrichment right now, obviously with a tune it'll do better but it's a long way off from modern GDI turbo engines that can go as lean as 0.9 lambda at ~3000 RPM WOT.

Maybe retrofitting TJI becomes hilariously, wildly impractical but if it's really "just" a built bottom end, intake + exhaust VVT, a major revamp of the timing/fueling map to account for the different ignition delay/knock limit, and a weird-looking spark plug then maybe it's viable when I inevitably blow up my engine doing something dumb.

[GTSBoy]

I would suggest it's a US$5000000 R&D exercise to come up with a workable design for retro to the RB26. If HKS make it work, it will cost about the same to buy it from them!

[joshuaho96]

12 hours ago, GTSBoy said:

I would suggest it's a US$5000000 R&D exercise to come up with a workable design for retro to the RB26. If HKS make it work, it will cost about the same to buy it from them!

They'll give a good price, finger in the air estimate probably 80-100k USD. 

[DanGreen006]

The combustion engine isnt going anywhere, we'll transition to synthetic and or bio fuels / hybrids 

Honda tested at 40:1 afr and settled on 34:1 with a 16:1cr in a video online when F1 switched to E10 with TJI

TJI is the biggest thing to happen in ignition since the spark plug was invented 100yrs ago aside from high frequency ignition which is banned and expensive for large scale oems.

You wont have to change anything it'll be a retrofit and you'll be able to tune it on a crappy old fcon v like HKS have done with your existing bottom end

I dont think it will be expensive either - Mahle, Federal Mogul, Beru all seem to getting ready to cater to oem applications in anticipation of Euro7 emmission laws -

$5m ?

I found some for sale for a gas turbine engine from Champion for $350 USD others for $500 for a big German Gas turbine unit

I think it will def solve a lot of your issues running Cali's crap fuel esp in regards to knock not to mention the insane 47mpg HKS are quoting for the heritage RB 

The secret will be in the tune imo

 

[TurboTapin]

On 3/17/2024 at 9:55 AM, DanGreen006 said:

The combustion engine isnt going anywhere, we'll transition to synthetic and or bio fuels / hybrids 

Honda tested at 40:1 afr and settled on 34:1 with a 16:1cr in a video online when F1 switched to E10 with TJI

TJI is the biggest thing to happen in ignition since the spark plug was invented 100yrs ago aside from high frequency ignition which is banned and expensive for large scale oems.

You wont have to change anything it'll be a retrofit and you'll be able to tune it on a crappy old fcon v like HKS have done with your existing bottom end

I dont think it will be expensive either - Mahle, Federal Mogul, Beru all seem to getting ready to cater to oem applications in anticipation of Euro7 emmission laws -

$5m ?

I found some for sale for a gas turbine engine from Champion for $350 USD others for $500 for a big German Gas turbine unit

I think it will def solve a lot of your issues running Cali's crap fuel esp in regards to knock not to mention the insane 47mpg HKS are quoting for the heritage RB 

The secret will be in the tune imo

 

What gas turbines have you seen this on? I work on a variety of turbines (GE, Siemens, Rolls Royce, Solar, Etc.) and our ignitors do not cost 350$ a pop nor do I understand how TJI would apply to a turbine. 

Edited March 20 by TurboTapin

[DanGreen006]

On 3/20/2024 at 11:58 AM, TurboTapin said:

What gas turbines have you seen this on? I work on a variety of turbines (GE, Siemens, Rolls Royce, Solar, Etc.) and our ignitors do not cost 350$ a pop nor do I understand how TJI would apply to a turbine. 

Sorry not industrial gas turbines but large gas industrial generators MTU4000.   When i was looking up the pre chamber plugs i incorretly assumed this was a industrial turbine not generator