How to Tune the Fuel Charge Cooling Coefficient in a Link G4+

If you are planning to use the ‘Modelled’ fuel equation in the Link G4+ or Vipec i-Series ECUs running firmware 5.2.2 or later, one of the first tasks once you hit the dyno is to calibrate the ‘Fuel Charge Cooling Coefficient’. This defines how much the fuel being injected will cool the intake charge and has a significant effect on how accurately the ECU can track requested changes in air fuel ratio. In this webinar we will look at the correct approach for tuning this parameter using our Nissan 350Z fitted with a Link G4+ ECU.

Watch: How to Tune the Fuel Charge Cooling Coefficient in a Link G4+

Post your questions below. English subtitles coming soon.

Just curious if you have any tips for configuring the Charge temp approx table?! Seems to be a lot of trial and error! Have you seen much benefit of this over the old school IAT correction only?!

You're right Dan, it isn't a straightforward table to tune. I'm hoping to do some testing on our 86 shortly and see if I can develop a sensible approach to dealing with it. Right now the best place to start is with the suggested starting numbers that Link offer in the help file - I've found these to be reasonable starting points.

If the lambda seems to change too much with heat soak at idle, I tend to weight the charge temp table more towards the ECT in these areas (normally idle/low rpm). It is a lot of trial and error though which I know isn't overly helpful.

Would the coefficient be any different between direct and indirect injection?

Yes it is, although the Link G4+ currently can't control DI engines so it's not strictly relevant. Motec have a similar parameter in their M1 ECU and on our 86 with port and direct injectors, the charge cooling gains are slightly different.

That you're aware of, is there any method to do this on the Elite 2500 series (say with a generic compensation)

Or is this handled in the background if at all?

Hi Jaydn, under 'sensor properties' you'll find a coolant temp bias table which achieves the same aim as the charge temp estimate table. 100% will give a value that is the same as the ECT while 0% gives a value that's the same as the IAT.

Hey Andre, yeah I've already tinkered with that as per your other video (Dropping the bonnet and increasing the IAT and monitoring the AFR)

I was more referring to having actual AFR accurately follow target AFR when changes are made as you showed in this webinar (without requiring tweaks to VE)

While you're here, is there any plans to do a video on setting up the Elite Closed loop O2, particularly the PID values.

I have watched your video on PID, however the elite didn't seem to behave anywhere near as nice.. Proportional seemed to either not be anywhere near enough or WAY too much, and Derivative seemed to cause further oscillations rather than dampen them..

Maybe it's just due to the inherit delays and i've got unrealistic expectations forcing the system into a too aggressive state.

Granted, I haven't spent overly long trying to dial them in as of yet, more just curious if there's a good methodology you've come across for their O2 control. Thank you, appreciate all your hard work!

The aim of the coolant temp bias table is to have the AFR track the target accurately as air temp changes. You do need to remember that the Elite gives you the option to have IAT compensated automatically in the main fuel equation (just using the ideal gas law), or you can use the IAT compensation table. If you choose automatic then the IAT comp table 'should' be zeroed, however often we find that the fuel equation tends to over compensate so some fudging within the IAT comp table may be beneficial. Just understand that if you've selected auto, then the IAT table shouldn't be making gross compensations of 2.5-3.5% per 10 deg C.

As far as the closed loop control goes, to be honest I found the defaults for our 350Z worked pretty well but perhaps we could cover this in a webinar. I know that the way Haltech deal with PID gains is a little unusual compared to many manufacturers.

Yup, has to be correctly calibrated to estimate heat transfer into the charge air. It's set to auto currently, and I don't have that other table enabled.

I think we got a wire crossed, because I was originally was asking if the elite has any way to factor in the fuel cooling coefficient, already knew it had air temp estimation :)

Though it is good to know you've found the auto calibration to be too aggressive, I'll try your suggested manual adjustments.

Yeah the base values work ok, found it a little slow to respond ppersonally.Did make the integral axis relative to AFR error, gave me a little more flexibility with how quickly it'd wind up (which I'm much happier with)

But a future video with more in depth Haltech tips and tricks would be fantastic if you ever get the chance!

Also If you're interested I came up with a way to make a generic output toggle on and off via a hysteresis, something that's not natively supported. Also found a way to have it support a momentary switch (Say a steering wheel button) to toggle a mode (a little more complex to setup)

Feel free to give that a try, both worked a treat for me; at least until haltech add native suport to both of these.

http://forums.haltech.com/viewtopic.php?f=31&t=33783

http://forums.haltech.com/viewtopic.php?f=31&t=33793