Lean burn engine (E85?)

[Juho]

No, it would have required bigger fuel injectors or flexfuel kit to run on ethanol.
Both options would have made the fuel consumption meter be off about 30%, which is something I don't like.. and cold start would have been a problem, unless buying a more expensive flexfuel kit that allows fine tuning the cold start injection maps (price tag is almost the same as the car!)

I sold the C5 and V40 2.0T automatic (which I had previously a longer time) on January and bought the BMW as it's easier to modify to run only on E85.

[cRiPpLe_rOoStEr]

As the EW10 engine was formerly available as a flexfuel in Brazil, it actually does surprise me the BMW engine being easier to tune for E85.

[Juho]

It's because there are no easily available or cheap tools or software that allows fairly easy ecu tuning like the BMW's Siemens ecu.

I only needed a k-can obd cable + 20pin adapter, and then just downloaded softwares required to download the ecu bin file to computer and upload it after changes made; Romraider and Siemens flash tool.

But as said earlier , the stock ecu (Siemens MS42) has it's limitations and the best solution is going aftermarket.
Which works fine, because the BMW is registered the first time on 09/1999.
OBD test is required on cars registered 01/2001 or later, and it's a problem if going for aftermarket ecu.. I could take the emission test with another car with a working obd port, but sometimes the inspector wants to test the emissions anyways and that is when it would fail MOT inspection.

[cRiPpLe_rOoStEr]

Originally Posted by Juho

It's because there are no easily available or cheap tools or software that allows fairly easy ecu tuning like the BMW's Siemens ecu.But as said earlier , the stock ecu (Siemens MS42) has it's limitations and the best solution is going aftermarket.

Odd enough, in my country the Siemens ECU and other hardware are often blamed for the unpopularity of an engine which was fitted to some Argentinian versions of the Ford Ranger until late-2011.

OBD test is required on cars registered 01/2001 or later, and it's a problem if going for aftermarket ecu..

Only in 01/2012 OBD-2 compliance became mandatory in my country, yet I'm not sure if the most popular aftermarket ECU supplier has any product suitable to this requirement.

[Juho]

A small update, even though I haven't had any great success.
But atleast I do find it quite interesting what I have found with very high compression and E85

- about 16,5-17:1 static compression ratio
- +25bar cranking pressure - I only have a cranking pressure gauge for gasoline engines, and maximum is 21bars, and the needle went to whatever stopped it which is estimated about 25bars
- E85 fuel
- Vanos not in use; very late intake cam timing and advanced exhaust -- no overlap at all. Because I had to use stock cylinder head because of my crappy tooling.. otherwise I would have already in use the better head for the engine and vanos in use for both cams.
- knock sensor input almost disabled - caused a lot of odd behaviour, and knock sensors "see" the proper self ignition as "knock" ; atleast based on their signal / voltage on log files.

I was able to make the engine compression ignite properly at part throttle at highway speed, when the engine got heated enough (water temp 98-99ºC)
But it stopped when the lambda feedback got randomly turned off and mixture went about 5% more lean.

The engine started suddenly accelerating itself and went slight uphill with about 8L/100km momentary fuel consumption.
Which is quite promising.

They keys for compression ignition on ethanol seem to be:
- heat, heat and more heat.
I need to figure out how to heat intake air without egr, before I can finish the other cyl.head so I can have internal egr in use
- very sensitive of fuel mixture. If it knocks -> more lean mixture. Knock seems to mean at this engine the mixture burns way too early and all the energy gets wasted because of that.
- load, to 'light off' or run it on compression ignition "mode" - not too much ignition advance because otherwise it will knock very bad (like a diesel with injector timing way off)
- currently spark ignition still in use, to help it light off. It's very difficult to find what it actually wants. I had to go leaner than lambda=1.15 before I got rid of the knock on part throttle -- full throttle no problems
- and better mixing of the air-to-fuel mixture; this is where it wants tumble flow, which I do not have because of bad intake port design. My other cyl.head does have port shape to cause tumble flow.
- injectors; bosch ev14 12-hole injectors -- currently using stock 3 liter injectors, which have only 2 holes and running them 5bars fuel pressure. They do have "ok" fuel spray at the distance but I am sure the ev14 12-hole injectors help a lot.
Related pdf here: https://www.speed-talk.com/forum/vie...hp?f=1&t=63866

Running estimated lambda=1.60 mixture at the leanest. Stock engine max lean was 1.40-1.45 and it started missfiring. Lambda=1.60, no missfiring.

https://www.youtube.com/watch?v=Y5nkuqGnScg

I had it on the max. (gas afr 22.4 / lambda=1.52), and went 10% leaner, which is why I guess it should be about 1.60
On the video I had lost a bit power, because going even leaner.
Earlier it could keep the speed with about 8L/100km momentary fuel consumption and sometimes required more throttle -> 9L/100km.
Only randomly it started igniting properly, assuming this because the mixture went slightly richer and injector pulse width was still the same and had more power.

Intake air temp only 11-12ºC. I'll see what improvements I could do.

And full throttle test on 3rd gear:

https://www.youtube.com/watch?v=jVoL0BuqLtQ

Some random pictures of engine:






I'm very surprised I haven't been able to cause any engine damage, that I could notice with compression pressure test, from spark plugs or looking through the spark plug holes what the top of pistons look like.
Even tho the engine has knocked very loud. But doesn't seem to be that dangerous to engine if it lasts only a short time and doesn't run on high load; e.g full throttle.

I'm also running on platinum spark plugs; NGK BKR6EQUP.
But it doesn't have enough platinum to cause catalyst ignition, which I would like to try.
Platinum glow plugs would be a good solution, but only available at too small sizes for the r/c car/boat/plane etc. engines which run on nitromethane or methanol.

It also seems like ethanol is actually more difficult to light off with a catalyst (platinum) than methanol.
Not 100% sure tho, but that's what I understood from what I found out.

I tried searching for glow plugs with platinum tip, similar to diesel glow plugs but it seems to be very hard to find information of materials used on glow plugs.
I did find out, that atleast Bosch glow plugs have inconel tip, which doesn't work as catalyst for ethanol.
But it is an option, to use a glow plug as hot spot to cause self ignition.
It could cause major problems at full throttle tho, if it ignites way too early.

And I am planning on installing a switch for ignition coils, to disable spark ignition when compression ignition is happening (well enough) to try it I could get the engine running only on compression ignition.
First I need to figure out how to get the compression ignition work better / more stable.

[Piotrsko]

You could modify the gap width a lot and wind platinum thermocouple wire around the electrode, or even weld on more material to wind the platinum around

[cRiPpLe_rOoStEr]

Originally Posted by Juho

about 16,5-17:1 static compression ratio

Most modern turbodiesels run at this range of static compresion ratio. Scania's compression-ignition dedicated-ethanol engines resorted to an even higher compression, yet they also require an ignition improver additive.

[Juho]

Yes, that is correct.

That is why it does require 'some' help to ignite E85 with compression.
It still should be possible, in a narrow window to work only on compression ignition, even with port injection, naturally aspirated engine.

The help it may require to work, might be just egr - or maybe it needs a catalyst; platinum.
Or richer mixture, but that is not an option.

I was able to raise intake air temp about 10C, and got it more stable on highway.
But 10C more heat is not enough.
Needs an undertray, so the engine bay won't get cooled so well.
It's quite cold outside, so it isn't very easy to reach +50C intake air temp.
--
@Piotrsko - That is an interesting idea.
Platinum wire seems to be quite difficult to find in my country - from China, atleast it is well available.
I'll keep that idea in my mind, thanks.

[cRiPpLe_rOoStEr]

Originally Posted by Juho

It still should be possible, in a narrow window to work only on compression ignition, even with port injection, naturally aspirated engine.

The help it may require to work, might be just egr - or maybe it needs a catalyst; platinum.
Or richer mixture, but that is not an option.

IIRC most often a richer AFR decreases the chance of self-ignition.

[Juho]

That is what I had thought before.
If I go closer to lambda=1 mixture, it will start knocking very loud even tho I have quite retarted ignition timing.
And at higher load (around 25-35% throttle position) it knocks even if going as rich as lambda=0.80 instead of 1.00.
But if I adjust it to lambda=1.25-1.30, it works fine.

So going leaner, e.g lambda=1.20 or leaner, the knock stops and it starts running fine.
That probably sounds a bit weird, and I found it actually quite funny how it works.

The knock is, like I earlier explained, is very similar to diesel knocking noise caused by bad injectors or something like that.
It ignites the mixture too early and burns quickly before TDC - or that is what I am assuming, when it really gets bad.

It is possible, that the behaviour is explained by fuel injector spray, which is not even close to "good" compared to EV14 12-hole injectors or direct injection injectors.
And air and fuel is not mixing so well.
Making it much more difficult to ignite the very lean mixture.