Cranking / starting part 1

Tuning, troubleshooting and the nitty gritty of using rusEFI to make your engine run nicely!
Post Reply
blundar
contributor
contributor
Posts: 141
Joined: Tue Jan 07, 2014 4:38 am
Location: Cincinnati, Ohio
Github Username: blundar
Slack: Dave B.
Contact:

Cranking / starting part 1

Post by blundar »

Again... it was asked to preserve some slack discussion on getting appropriate settings for cranking and starting the engine. Much like idle, this can be kind of tricky and there isn't an instrument you can put on an engine like a wideband for cranking.

Overall strategy:
-You need to get the motor spinning fast enough for your crank sensor to reliably measure how fast the motor is spinning. This is a hardware dependent thing - there is no universal answer. If you have timing faults, you're going to have a hard time getting the motor going. Resolve these.
-How fast the motor spins will greatly affect the settings you use to try to make it "go." Not all starter motors are created equal - some rusEFI controlled vehicles have even been kick-started motorcycles
-During cranking, motors need the same things they need while running - air, fuel and spark.
-When the motor is not spinning fast, it needs different timing because of the low speed
-When the motor is cold it needs different amounts (and injection timing) compared to when it is warmed up fully
-Getting the right amount of air helps too
blundar
contributor
contributor
Posts: 141
Joined: Tue Jan 07, 2014 4:38 am
Location: Cincinnati, Ohio
Github Username: blundar
Slack: Dave B.
Contact:

Re: Cranking / starting part 1

Post by blundar »

Edited for clarity / more information Dec14-2021 DB

For a rule of thumb, the the cranking RPM limit should be about 2.5x the fastest speed "starter" spins is a good estimate (i.e. 120 x 2.5 = 300 RPM).
The cranking RPM limit is the speed at which the ECU stops using "cranking" logic to run the engine and the "normal" ECU logic (be it MAF, speed-density or Alpha-N) starts to take over.

If you're dealing with a slow or unreliable starter, lowering the cranking transition may be in your best interests. Just be sure to adequately tune the low RPM area that the engine will end up on immediately after starting.

It's generally a good idea to setup your tables such that the first RPM index is close to the crank/run threshold, or at least the lowest table entry wouldn't be horrible if it were used at the crank/run threshold. Keep in mind that particularly in speed-density, the engine is going to see very different loads while cranking than under normal operation or idle. This can sometimes be tricky to tune by direct observation and you need to guestimate / extrapolate from parts of the tune you can more easily get to.

Vocabulary:
-Priming fuel is typically a SINGLE pulse that is fired ONCE with the intention of wetting the walls and getting the process of liquid fuel evaporating into combustible fuel vapor started
-Cranking fuel is fuel supplied to the engine while it is spinning on each revolution instead of the amount of fuel typically prescribed by MAF, Alpha-N, Speed Density or whatever other method is used to determine fueling during normal engine operation

It is easier to start with too much fuel than too little - you just need to be careful not to foul / wet plugs. If your plugs get so soaked with fuel that the electrode is covered in liquid fuel, the vehicle will NOT start. Remember, liquid gasoline does NOT burn - only once it has *vaporized* can it light off.

The priming pulse shoots some fuel into the engine to start the process of evaporation, which will happen on its own at a rate determined by temperature. You can play with the "priming fuel pulse" to help starts, particularly cold but this is something to do AFTER getting the cranking settings dialed in somewhat first.. A guideline for a first idea of a priming pulse wile warmed up fully would be to get the engine started, log the injector pulsewidth at a normal idle and then multiply this pulsewidth by 1.5-2. Cranking fuel will similarly be substantially larger than fuel idling, if for no other reason than a lambda target of 0.6-0.8 is within expectation for cranking vs. closer to 1.0 for idle.

Starting fuel (bot priming fuel pulse and cranking fuel) will MASSIVELY increase as cold temperatures decrease. If the data for vapor pressure of the fuel you are using is available, it is an excellent mathematical guide to how much the priming and cranking tables need to be increased with respect to temperature. At these conditions, coolant temperature will be the one that matters most. Keep in mind that ethanol has a VP close to zero below ~ 8C so cold starts will suck/beimpossible without a little gas in the mix. I found this https://rjpatelioc-petroleum.blogspot.com/2017/04/rvp-vs-tvp-of-gasoline.html handy dandy site with some decent charts. If you have hot starts good, look up your vapor pressure at a certain colder temperature. For your cranking/priming tables, multiply by (hot vapor pressure / vapor pressure at this cold point) to get a good first estimation of how much extra fuel you are going to need.

Note: these are just a guideline a starting point. OEMs spend a majority of their tuning budget figuring out optimal starting procedures that also minimize emissions. Don't expect to nail startups in all temperatures on the first go.

I'd set the After Start Enrichment multiplier to 1.1 or even 1.25 to get a little more fuel in there. Too much fuel is a thing but it's generally a lot easier to get the party started with things running a little on the rich side, figure 0.75 lambda as a lower limit wit h a goal to stay above 0.8lam.

If your engine is easier to start when you open the throttle a little bit, that can be a sign you have too MUCH cranking fuel

In terms of timing... A rule of thumb is that the faster the starter spins, the more timing advance you'll want to see. I'd recommend setting the "Use Advanced Corrections for Cranking" to false so that crank timing exists independently of other corrections. Crank advance is going to be determined by the engine. In general, 5-15 degrees LESS timing than idle is typical for cranking. It is not at all unusual for higher compression or slower spinning motors to fire at 0 degrees while cranking, or even negative(AFTER TDC). Doing so ensures that whatever energy from combustion only pushes the pistons in the "right" direction, at the expense of efficiency. If your motor sound like it is "fighting" itself or struggling to crank, this is generally a sign that you have too much cranking timing and you need to regard it. There is often a slight increase in cranking advance when the motor is cold compared to hot.

Getting air right is also a big deal. The throttle plate is a huge restriction to the engine when closed. If you want to see how big of a deal this can be, try cranking the engine with coils disabled so it definitely won't start. Try it with the throttle closed, log RPM. Then go to WOT and repeat, logging RPM. You'll see a significant difference. On engines with a slow or weak starter, opening the throttle more will let the motor spin faster which can help a bunch with crank sensor sync up. Opening the throttle more will also tend to let more air in, requiring more fuel to get a happy mix to light off. Be aware of what you're doing.

There isn't a hard and fast rule for how much air an engine needs. There are some important differences between cable throttle and ETB setups.
With cable throttle, the amount of air that enters while cranking is (Air through throttle body + Air through idle valve + any leakages)
With ETB, the amount of air that enters while cranking is (Air through throttle body + any leakages)
"Any leakages" here means both accidental leaks like bad gaskets, cracks in stuff, etc. and deliberate leakages like bleed screws or holes in intake manifolds.

With cable throttle, the throttle stop is an important factor affecting the minimum amount of air that can go through the throttle. When it closes, the throttle rests against the stop, but that doesn't mean that it restricts airflow completely. It's generally a good idea to adjust the throttle stop so that when the engine is idling fully warmed up at its lowest RPM (i.e. lowest airflow requirements) the Idle Air Control (IAC) is between 20% and 40% duty. This gives the idle valve plenty of room to open to increase idle but still leaves room to close it to bring revs down quickly.

With ETB, the IAC is modelled by using a range of ETB motion for idle. There is a parameter which determines the range between 0% IAC (ETB commanded filly closed - note some air will still move through the ETB) and 100% IAC (ETB positioned at the "Electronic throttle idle range(%) location). Essentially, you can choose how big of an IAC you want to have by changing the range.

Regardless of whether you're using ETB or cable+IAC, a good starting point for cranking airflow is 1.25X - 1.75X where the engine operates idling at a given RPM. Get the motor started, log where ETB / IAC is once things stabilize, do a little math and there's a place to start. This will generally be enough air to make the engine run at a higher RPM than target idle. The "cranking IAC taper" controls how long it takes to transition from the cranking airflow logic to the idle airflow logic and is a good way to keep the engine operating at an elevated RPM before entering idle.

Remember, you need to blend changes to get a smooth transition from cranking behavior to running behavior. It is extremely likely that the settings required to get the motor moving initially are going to be way different than those required for idle or normal operation. When setting up cranking parameters, keep in mind that the engine will need to make this transition.
blundar
contributor
contributor
Posts: 141
Joined: Tue Jan 07, 2014 4:38 am
Location: Cincinnati, Ohio
Github Username: blundar
Slack: Dave B.
Contact:

Re: Cranking / starting part 1

Post by blundar »

Matthew Kennedy suggests:

"60mg fuel / liter of displacement for each cylinder is a much better estimation for cranking fuel."

i.e. 5.3L V8 = 5.3L/8 = 0.666L / cylinder * 60mg/L = 37mg cranking


As of today (12-14-2021) priming logic is too new to have well defined suggestions.
deffie83
Posts: 5
Joined: Sun Feb 06, 2022 11:05 pm
Github Username: gdiciocco
Slack: Giacomo Di Ciocco

Re: Cranking / starting part 1

Post by deffie83 »

thank you for your guide!
Post Reply