Max ignition advance

[NickRP]

Hello Guy,

I have a short question: could there be a situation where a reasonable automotive engine would require more than 44° of ignition advance?

I am developing some sort of ignition controller, so I need to decide on spark advance range.

Many thanks,
Nikola

[Guy Croft]

Yes Nick, allow for up to 46 deg.

GC

[NickRP]

Thanks Guy, your advice is of much help!

Best regards,
Nikola

[dfom]

Nick,

Just to make an extra to Guy info.

I´ve already advanced my ignition up to 45 degrees without any detonation signal (audible and detonation sensor).
My engine is 16v CR 9.5:1 2l Fiat - Gasoline was Brazilian one (25% ethanol + 75% Gas) (E25). Our gas burns slower than 100% pure gas due to the 25% ethanol, so more advance needed!
I didn,t dyno the engine, but at road i felt it a little "smarter".

100% ethanol engines (E100) (stock ones from factory) uses 12:1 CR OR 10:1 CR with an even higher advance (around 7 degrees over 25% ethanol (E25)) to cope with the "low" CR for an Ethanol engine.
Tuned ethanol engines goes over 14,5:1 CR and small advance.

Flex fuel engines (now almost 100% of our new flet) uses a CR of around 11:1 and an algorithm that (based on NARROW lambda signal and a table of A/FR for all the possible mixtures from E25 to E100) determines the amount of Ethanol on the mixture and adjust advance to deal with the correct timing for the mixture.

Hope to add useful information (I have a power point presentation about the flex fuel algorithm if you wish i may submit you by mail)

Bye

Daniel Melo - Brazil

[NickRP]

Hello Daniel,

My engine is 16v CR 9.5:1 2l Fiat - Gasoline was Brazilian one (25% ethanol + 75% Gas) (E25). Our gas burns slower than 100% pure gas due to the 25% ethanol, so more advance needed!

Which engine management is factory installed in Brazilian Tipo 16v?

Flex fuel engines (now almost 100% of our new flet) uses a CR of around 11:1 and an algorithm that (based on NARROW lambda signal and a table of A/FR for all the possible mixtures from E25 to E100) determines the amount of Ethanol on the mixture and adjust advance to deal with the correct timing for the mixture.

I must admit here that I didn't get this part of your post 100% clear. The following is not clear:
-How does the engine management know which fuel is in tank? (you said "table of A/FR for all the possible mixtures from E25 to E100", but how does the ECU know which table to use?)
-What exactly is the narrow band lambda signal used for? Does it play a role in determining ignition advance?

I would be interested to see the presentation on flex fuel algorithm. I'll send you a PM with my email address.

Best regards,
Nikola

[dfom]

nick,

Tipo 16V were all imported from italy. Usually imported cars are reprogramed to deal with the E25 fuel by correcting AFR and ignition, but this is not aways done.
So all the Tipos 16V we have here in Brazil use the same ECU from italy. I don´t exactly know the supplier or model since i´ve never worked on a tipo 16V (mine is tempra)

How does the engine management know which fuel is in tank? (you said "table of A/FR for all the possible mixtures from E25 to E100", but how does the ECU know which table to use?)

Here is were lies the "trick" used by the algorithm. It works this way:

After car is started, ECU mesures lambda. If rich or lean it start an learing algorithm to determine the correct AFR necessary to balance the mixture.

It´s done by progressive (step by step) enrichment or leaning of mixture by changing a VARIABLE that holds AFR for the fuel used.
So AFR is not fixed, but an variable that is calculated at engine first running moments.

Once a good variable is found, AFR is no loger a variable (until the next engine start) and it´s hold on memory and used as a common ECU would do.

Using a table that relates AFR to % composition of Ethanol do Gas it´s possible to find the % of ethanol to pure Gas. With this info ECU correct ignition advance.



What exactly is the narrow band lambda signal used for? Does it play a role in determining ignition advance?

It is used for the step-by-step procedure for finding the best AFR for the current fuel used.
With AFR, as said, advance is corrected based on the mixture in tank.
I am not sure on how it is done. May be a fixed value added to a table value or an interpolation between two tables (one for E25 another fo E100)

Any possible mixture may be used (E25 to E100) . Cars run nice without problems.
When Ethanol is cheaper you fuel the tank with it, otherwise use gas, or any mixture.

I would be interested to see the presentation on flex fuel algorithm. I'll send you a PM with my email address.

Ok, i will find it here (on my messy PC) and send you..

Bye

Daniel Melo - Brazil

[NickRP]

Daniel,

How big is really the difference in AFR between those fuels?

I must say that I am a bit suspicious of whether this type of control (at least how I understood it) can work at all.

To list some of the issues I see here:
-O2 (lambda) sensor output cannot be used until sensor AND engine are warmed up. So, regulation and fuel detection in "first running moments" are hard to be made (a bit of understatement here).
-Adaptive fueling correction factor is VERY variable thing even on single fuel engines, let alone flex fuel ones. Let's say, it is quite common to have values of plus/minus 15% on stock, factory mapped, fully heated, single fuel engine under different working conditions. Such variations would guide an "automatic fuel recognition algorithm" into recognizing fuel incorrectly.
-System starts with incorrect timing. I have no idea how big is the difference in optimal ignition advance between different fuel mixes, but it could be significant (otherwise there would be no point in making complex ignition adjustments we are discussing here). So, detection algorithm has to be able to recognize correct fuel in situation where ignition timing is off... which is making the thing even more complicated (O2 sensor readings are temperature dependent to some extent)!

Best regards,
Nikola

[dfom]

Daniel,

How big is really the difference in AFR between those fuels?

I must say that I am a bit suspicious of whether this type of control (at least how I understood it) can work at all.

To list some of the issues I see here:

-O2 (lambda) sensor output cannot be used until sensor AND engine are warmed up.
So, regulation and fuel detection in "first running moments" are hard to be made (a bit of understatement here).

-Adaptive fueling correction factor is VERY variable thing even on single fuel engines, let alone flex fuel ones. Let's say, it is quite common to have values of plus/minus 15% on stock, factory mapped, fully heated, single fuel engine under different working conditions. Such variations would guide an "automatic fuel recognition algorithm" into recognizing fuel incorrectly.
-System starts with incorrect timing. I have no idea how big is the difference in optimal ignition advance between different fuel mixes, but it could be significant (otherwise there would be no point in making complex ignition adjustments we are discussing here). So, detection algorithm has to be able to recognize correct fuel in situation where ignition timing is off... which is making the thing even more complicated (O2 sensor readings are temperature dependent to some extent)!

Best regards,
Nikola

Nikola,

I quite don´t understand the subtbles of the algorithm. The lines I depicted were knowledge I´ve from simple drafts (those solutions are patented, so they don´t explain how it´s done).
But I assure the described algorithm works OK since 100% of new cars here in Brazil use it. Even Honda, Toyota and other use it.
I´have no idea when does the ECU start to recompute AFR, based on what i´ve read it´s after start up.
This step-by-step algorithm for AFR may be set to run after engine warm up. Since fueling do not happens at home, but after car is already a little hot at the gas station nearby.

The AFR are 13.8:1 for E25 and 8,9:1 for E100 (100% pure ethanol). This means 44% more fuel volume for the same air mass, taking into account the different densities.

The advance difference is about 7 degrees. Ethanol needs 7 degress more advance, not sure on how advance is set when ECU is at an "not defined" fuel state. Probably Gas Map beeing used as it´s the more conservative one.

This is everything i know about it. Tecnical papers are avaliable from SAE and IEEE. This system was evolved by Bosch at mid 90´s.
With the rise of petrol prices in the recent years ethanol became a fuel with an atractive, and flex cars becomes an market success (99,999 % of new cars are flex fuel)

Ethanol fueled cars are more powerfull than gasoline ones due to the fact that ethanol takes more heat to vaporize than gas and have a really high equivalente "octane rate".
From memorie i guess Ethanol has 90 MON and 120 RON. Deals ok with 14,5:1 CR and really cools the intake, reducing detonation tendency a lot. Almost all charged preparations use Ethanol.

Bye

Daniel Melo - Brazil

[tempra16v]

one and a half year later...

are you guys using another fuel than gasoline in Europe?

[Guy Croft]

Sorry, no idea what folk are using!!

Some interest in this topic please, dear readers..

GC

[tempra16v]

Sorry, no idea what folk are using!!

Some interest in this topic please, dear readers..

GC

Guy,

Do you have cars in UK that uses 100% ethanol?

[Guy Croft]

Sorry but I don't know what folk are doing with fuels other than gasoline in the UK.

Most of us are on pump gas.


GC

[kritip]

At the pump the basic unleaded is 95 RON. You can also get 97, 98 and 99 RON.

Then there is LPG that is readily available and conversion for petrol cars are commonplace.

There are also regular diesel, and "performance diesel". I know nothing about these though.
Most diesel at the pump is now being replaced with "Bio-Diesel" however that generally referes to a 5% / 95%, bio / regular split. This is a government innitiative

I beleive there is also E85 at some forecourts, but availability is lower, and I've never met a person running on E85!

There is no E100 as far as i'm aware in the UK.

thank you,

Kristian