Discuss all things tuning in this section. News, products, problems and results.
Theres alot of secrecy about this. Can anyone shed some light?
In a direct injection engine.
Inject later/earlier up high ( rpm ) ?
Inject later/earlier on high load ( turbo) ?
How long should there be between SOI/EOI and ignition?
Give the engine what it wants. There are many reasons that the results from one engine may not be the same as another. For example:
-- Injector location -- what is the transport delay from the time the injector starts (or stops) spraying, until the atomized mixture reaches the valve. In addition to the distance from the valve, it may also depend upon flow velocity (so related to RPM, engine displacement, intake runner / port size, valve size).
-- Injector spray pattern -- how atomized is the fuel leaving the injector -- does it need to be atomized by evaporating off the back of a hot intake valve, this might vary with fuel pressure, too.
-- Injector size relative -- a large injector relative to the fueling requirements of the engine will mean that you have more opportunity to pay with injection timing (especially at high RPM). If the maximum injector duty cycle is 75% then injection timing will make a lot less difference that if the duty cycle is 30%.
Of course, there are webinars telling you how to go about tuning this (i.e. figuring out what the engine wants):
Thanks for the reply. I should have specified direct injection. Do these same principles apply?
Sorry I missed that. My experience is only with Port Injection.
All good i totally left it out. Ive been trying to get the basic idea. but it seems to be a mystery at the moment. lol
I don't believe there's any secrecy. It's more to do with a limited amount of knowledge and experience in the general tuning market on DI engines compared to PI. While David was talking about PI engines, his comment still stands - Give the engine what it wants. There are no absolutes here because the specifics will depend on a million factors - Injector spray pattern, DI fuel pressure, is your ECU specifying SOI or EOI, what is your injection angle, what pulse width do you require to achieve target lambda?
The key point is that with DI engines we're heavily restricted on injection window as we can only inject during the intake stroke and part of the compression stroke. Even with this in mind you need to be careful of the SOI point as this has the potential with some injector spray patterns to damage the piston if you're injecting near TDC (this fortunately hasn't been an issue on the DI engines I've been involved with). You also need to have the injection event complete prior to the ignition point which probably goes without saying. I've found that I'm keeping a window of around 20 degrees between EOI and ignition but this is also seemingly engine dependent - That works nicely on a VW Golf TFSI engine but will cause all sorts of ugly running on the Subaru FA20.
Basically I'd suggest holding the engine in steady state and moving the timing around and watching the effect on AFR and torque. You will soon see what your particular engine likes. This needs to be done at multiple load and rpm points so that you can start building a map. It's not necessary to tune every point and I'd pick too at each rpm range and interpolate. Likewise you can be quite coarse in your rpm points - Maybe 2000, 4000, and 6000 rpm and interpolate/extrapolate to suit.
We will be running a webinar covering this in more detail as soon as we have the 86 mobile again so hopefully that will help you.
This is a big topic. Since you are talking about an FA20DIT Subaru engine (I think we may have exchanged some PM's on NASIOC), it is side injection, intake and exhaust variable valve timing, single injection event for all speed and load and I think it's 150 bar fuel pressure. From the perspective of torque, earlier SOI cools the inlet air and can give a very slight improvement of charge air (volumetric efficiency). A later SOI cools the charge in the cylinder and reduces knock, but gives less time to mix.
Inject later/earlier up high ( rpm ) ? Generally, earlier SOI as rpm increases.
Inject later/earlier on high load ( turbo) ? at low engine speed, a little later with higher load. At high engine speed, it is less load sensitive. On side injection engines you may see less smooth maps than central injection (think BMW & Mercedes for central DI). This is because the interaction of the spray pattern, piston, and cylinder walls can be a more non linear situation.
How long should there be between SOI/EOI and ignition? It's more about the position of the piston. When start of injection is before about 330-340 deg or EOI is after about 40-20 deg BTDC, you start hitting the piston (you'll see stock maps spraying this early anyway). At minimum you have an oil dilution or smoke concern (depends on injection pressure, spray pattern, etc).
EOI is less of a concern unless you are near the effective injector duty cycle limit. Due to injection timing window, max duty cycle is more like 40-45%. So if Start of injection is 350 BTDC, end of injection is 20 BTDC, duration is 330 crank angle degrees. 330/720 is 46%
As far as AVCS (variable valve timing) goes, well you generally want to tune the cams first. Then try adjusting injection timing. Rail pressure is typically max or very close to max for WOT.
Thanks for the response Andre. Ill definitely do that. Im still working on the formula on my own reading SAE papers trying to sift between emissions based stuff and actual performance. all the engines in the papers seems to be 1500 rpm static rpm.
Yea we did talk together sir Raymond! HELLO AGAIN! Thank you for your time man. both of you.
and yea FA20DIT.
Ive been getting better with knocking out LSPI with moving the angles later in the stroke. About 20 degrees up to about 270 or so watching noise diminish.
Ive been applying that concept to higher rpm with some predictable response ( higher angle less noise without touching ignition ).
and my duty cycles rarely peak 38%. I havent pushed further than that and have settled at about 8 degrees timing high load high rpm and 17psi boost.
What platform are you making your tuning changes on?
Fa20dit in a 2015 fxt
You will really benefit from the webinar we have planned then. As an aside, in N/A form the stock injection timing table is pretty good. I spent a lot of time playing with this using the ecutek software (which is time consuming in itself since you need to flash each change) and found minimal differences. There are gains to be made through the VVT timing but these also are quite limited in N/A form. If I recall back to our development car, the best I could get through just tuning changes without hardware modifications was in the region of 6-8 kW atw or thereabouts.
Thanks for the insight Andre. I will be investing in a standalone in the future. To speed up the process. Just a matter of which platform. Which has the easiest software to deal with and most capability of control.
And I will be there for the next webinar!