Pfc Tuning For Hot Weather

[GTRNUR]

Ive been re-tuning my car for summer driving, and have been strugling with issues steming from the Air temp corrections. Basically ive found that I can get a perfect tune during the day, or at night, but not both. The problems appear to be stemming from 2 area's.

Firstly, the air temp corrections in Setting 2 dont seem to be having an effect (or my understanding of how these values apply to the base map is wrong).

Secondly, the air temp sensor location (in the plenum in an R34 GTR) seems to be getting heat soaked when stuck stationary in traffic, and it takes sustained driving at speeds above 80km/h to rapidly re-cool that sensor sufficently so that the engine isnt being over-fueled again.

The actual engine temps are being reported as being at the lowest 79 degrees and a peak of 85 degrees when stuck in traffic in the middle of the day(PWR radiator). If the car is tuned in the middle of the day, as air density is lower (hotter air) the fueling requirements are less. However the same tune at night is as much as 0.5 to 0.7 AFR incorrect as a result of changes in air density.

A little background....

Im tuning with a Datalogit linked to a 0-5V PLX devices wideband AFR.

Crusing at 80-100km/h in the middle of the day I am seeing air temps around 55-58 degrees. At 60km/h and below when stuck in traffic, air temps are being reported as being as high as 66 degrees. I have seen the temps past 70 if the car is parked when hot and then allowed to heat-soak, and then restarted. Both of these situations are when outside ambient temps are about 36-38 degrees (a scorching hot day).

So can anyone think of what Im doing wrong here? Is it my lack of understanding of how the air temp corrections are implemented in settings 2, or is this just a grin and bear it situation that can't be effectively tuned around?

Also, has anyone ever considered moving the air temp sensor from the plenum to be in the hose, a little further away from the plenum so that it is less suseptable to heat soak, and also possibly updating the sensor so that it is of the new NTC style air temp sensor, so temp changes are more rapidly detected?

Cheers,

Ian

[STATUS]

send me the file and i will have a look over it:

[email protected]

[GTRNUR]

Thanks mate, ive just sent a copy of the most recent ecu config.

Cheers,

Ian

[paulr33]

the PFC will dial in injection correct based on airtemp

it's the PFC saying, the air is hotter, run richer to avoid detonation city

if you take out the correction - that is, unplug the airtemp sensor, or move it

the same airtemp is still going in (ie sub 70deg after sitting there) and the injection will be 1.00 instead of sub 1.10x ish

so it will likely detonate as the airtemp may be too high

so by moving the sensor, you just give the ECU false readings

which in turn wont enable the injection correction for excessive intake temps

which in turn will likely make it ping if the engine is tuned correctly

so your options are

1) tune the car in winter and setup the airtemp correction correctly so that in summer, by design it backs out timing (the PFC supports this on airtemp) and richens up

2) move the sensor, give the ecu false readings and hope it doesn't ping

3) unpiug the sensor and give the ecu no airtemp reading and hope it dosen't ping

[paulr33]

also check on the datalogit forums if someone has worked out any alternate sensors to use, and if they update faster

this may help and blow away the hotair heat soak more quicker

[GTRNUR]

I was hoping you'd chime in Paul.

I think I will probably have to go for option 1, and just grin and bear the crappy fuel economy/rich running on a hot day. It just seems that the rate the PFC richen's up the mixtures as temp increases is a little too exponential.

Can you please explain how the INJ vs Air temp + Boost setting on the setting 2 page is supposed to work? Is this only really relivent for when you have a boost control module connected so that the ECU can reference an external map senor too?

Cheers,

Ian

[Adriano]

Can you adjust the voltage vs temp? Im assuming its a d-jetro. I would consider fitting a different sensor, as i know on an r32 they are the same sensor as the water temp sensor.

[GTRNUR]

Just had another idea.... I could look at wiring a resistor in parallel with the air temp sensor, so that as the sensor approaches 50 degrees it will only really be repoting 40ish or so. It will take a bit of stuffing around, but since I cant seem to alter the rate at which the AFR's are richened up, tricking the ECU to think that it is further down the scale might work.

The ignition retards vs air temp will still work, and will just need to be tweaked to accomodate the new scale.

[GTRNUR]

Can you adjust the voltage vs temp? Im assuming its a d-jetro. I would consider fitting a different sensor, as i know on an r32 they are the same sensor as the water temp sensor.

No sorry its a MAF sensor version. The car is more or less standard except for a nismo exhaust too. Its no radically modified car by any means. Im just after a better tune. The standard ECU was running 11.5 - 11.2:1 AFR when on boost (on a normal 28 degree day, not the 35+ degree days we are getting at the moment).

Its the same sensor for air and water on the R34's too. The picture I posted of a sensor in the first post is what is used in the wolf/haltech ecu setups, and they react a lot faster. You can actually see temps change when data logging air temps vs boost.

As far as the datalogit forums went, I only found one reference to a guy with an MR2 that had simular issues. He was advised to try a simular sensor upgrade as well, but there was no follow up as to how well it went.

Edited December 30, 2009 by GTRNUR

[discopotato03]

I would have thought that the Mass Flow Sensors'd take temperature/density into account ?

Cheers A .

[GTRNUR]

I would have thought that the Mass Flow Sensors'd take temperature/density into account ?

Cheers A .

I think its like Paul was saying. The ecu is going into "over protective mother mode" and dumping in a ton of extra fuel to starve off detonation. It just seems that for every degree above 60 degrees the amount of fuel added is at least a 0.100 correction. Tuning in my climate then means I have about 5 degrees of tunable area before the ecu goes into mother mode.

[JonnoHR31]

have you tried setting them all to 1.000? aslong as you stay on the safe side with your afr's and keep the air temp ign retard then it should be fine

[paulr33]

you shouldnt need to, and you definelty shouldnt modify any of the sensors, please don't do that

if you think the PFC is adding in too much fuel when it shouldnt due to airtemp correciton

either adjust the amount of correct, ie less, so if its say 1.095 then try 1.045

or adjust the airtemp range

either will give you the same result

ie less overfueling at a given airtemp

or a higher airtemp thresohold

the key to getting it right will be

if you shift up the airtemp range or lower the injection correciton, do you get pinging (youll see check light come on)

so thats when you know you've gone too far

ie - if you didnt have this airtemp correction or you turned it off, its likely it would ping, so its a good thing, it just sounds like its maybe a touch too sensitive or too much correction

[GTRNUR]

Thanks Paul, i'll have another shot at playing with the air temp INJ corrections. I should have mentioned earlier that I have tried setting these to 1.000, and also very large values in the past, but they didnt seem to have any effect on the AFR or delivered INJ value.

I suppose it is possible that these changes do not come into effect until the engine is driving under load though (like the ignition lock at 20 degrees when at idle), as when I was making these changes I was stationary.

Ive been keeping the knock levels pretty low (<20 at max rpm/boost) and have my threshold for the warning set to 30. I havent noticed any knocking so far. Quite possibly because the ignition is already tuned somewhat retard compared to cooler climates. This will no doubt change in 6 months when it cools down 10-15 degrees though.

Cheers,

Ian

Edited December 31, 2009 by GTRNUR

[paulr33]

yeah check it when you are driving or look at the logs

another way is to view SETTING, ING/INJ

which is the temp adjust screen, check out the Kij and Fin values

these are injection value in real time

unsure if the PFC will show you corrected value (it should)

[STATUS]

just got your email, i will look over the file 2nite over a few beers.

edit: actually im playing with it now, i think i have found the load hysteresis settings.

can you email me a afr log? or at least a few known afr points on the map say?

[JonnoHR31]

try it at 1.000 on all of them and see if you get consistent afr's under load, if not then something else is causing your problems. you still have ign retard vs air temp which you should use to keep knock levels low at higher temps.

in the sample rb26 file it has the temps at 100, 70 and 40 with the correction at 1.195, 1.117 and 1.000. if your temps getting up to 100 with these settings it'll be adding 20% more fuel ontop of whats already in the map. thats enough to take you from a steady 12:1 down to 9.6:1, way too rich. correct me if i'm wrong but dont most people aim for a steady afr (ie 12:1) under full load regardless of temp then use timing to stop it pinging on hot days? which would make that correction a little too much

[GTRNUR]

Well I went for a quick lap today and have altered the settings for air temp vs INJ on injector 2 as JonnoHR31 has suggested. As Paul had confirmed, these settings 2 air temp corrections have no effect at idle, yet the corrections are applied when driving. So thanks for your assistance guys.

Thanks for your assistance to Trent. Especially offering to look at a map for me in the middle of the xmas break period.

I changed the 70 degree air temp value from 1.147 to 1.050 ish(I think.. that was 6 hrs ago and im in holiday mode!). This seems to have done the trick as the AFR's are no longer diving well into the 11's when on boost. It actually went to 12.0:1 (so it seemed I guessed close).

So I think this has resolved the issue I was having. In the end it was just my lack of understanding about how the corrections are applied. Every ecu sure does work a little diferently and I am finally getting to know the FC's quirks. Now all I need to do is wait till it cools down a bit so I can get a cooler air temp tune into the car, then it will just be a matter of tweaking the air temp corrections... now that I know that they actually do work!

One more question... With the Wolf ECU's that ive tuned in the past, it is possible to re-scale the MAP sensor so the tuning map provides more resolution at the low end of the load scale, and max boost pressure = 100% load on the fuel map. It is also possible to re-set the lowest manifold pressure point to be the zero load point on the map. I hope that makes sense?

Ive noticed that the FC seems to never be able to achieve low enough airflow (at idle or in de-cel) so that the active load point is in the 1st row on the map. Also as the car is only running 0.8 bar, the max load band achieved is only 14 (15 occasionally if I really push it and get a boost spike to 0.85 bar). So essentially I am only able to tune with 13-14 load bands.. Not 20.

Is there a guide somewhere for re-scaling the load points of the FC's map so it is appropriately scaled to the engines operating ranges? The Settings 3 page relating to the MAP airflow sensors allow altering of the airflow curves, but I dont quite understand how the 1-32 points on the air flow curves relate to the 0-20 load points.

Cheers,

Ian

Edited December 31, 2009 by GTRNUR

[STATUS]

Load scaling is fully programmable, right click on the map referrences across the top of the map and select map refferences.

Then got to settings 3 and look @ the left hand table 01-20 is your load points and rpm points

as far as my previous post INJ vs IAT (load hysteresis) there are 3 parts to the table (the 1st two are self explanatory) far right column is the load hysteresis and you can change the constarints to suit your application.

i am playing with your map as it has a few holes (peaks and troughs) i only have the demo versin on my travel lappy but i will have my tuning laptop 2moro so will email it 2moro night.

i can set all the load poits if you like but i would really like a afr log if possible, as your load line 14 is super lean in one spot then rich and timing jumps up 4 degrees from 14 to 15. fuel also gets leaner as it goes past 14.

Interpolate it richer down to the bottom line and drop timing off as it travels down too.

[JonnoHR31]

as trent said, load point setup is under settings 3

i'd personally leave p01 and p02 the same, then set p20 to the same as p16 or 17 if your reaching 14 and 15. then use the graph function in the bottom right corner to smooth the curve out similar to the way it is with the sample scaling. make sure all of your cells are currently richer and more retarded than the ones above them, so that changing the scaling doesnt make it lean out or knock the first time you put it under load to start tuning.