Ok I think I've got the slopes calibrated pretty good, LTFT is about 0.99 above 2k, 0.96-0.98 between 1-2k but swings from 1.00 to 1.05 at idle make idle a tad rough (lean) when it swings up to the 1.05 mark.
If I richen the low slope by 1% to get a nice 1.00 at idle it goes rich by almost 10% at low loads.
Is this a good example of when adjusting the offset would be appropriate to get slightly more fuel at idle only? I will try that tonight to see what kind of affects it has.
Last edited by rolls; 10-18-2015 at 05:55 PM.
Good questions there. I think you are answering most of them yourself by playing with it.
It can be a bit of a juggling game to see what works.
You can certainly adjust the offset to get more fuel at idle. But very very small increment. Think even .01. To make things simple I just use the one number in the whole table cause really the car should be around 13.5 volts. I have seen this done on a US ford factory tune and makes understanding how this works alot easier when changing one number.
Rolls as in the PM how about trying ID1000 data with Offset + 0.0725ms and see what it does. Someone else reported it worked with their ID1000 like injectors (basically the same injector but machined by a different company).
Ok will try and see how it goes, that is about a 5% increase.
Also something else I removed all of the cold start retarded timing to heat the cat up, as most of my driving is ~5km max it has increased my fuel economy from 20L/100km to about 16L/100km. The car also doesn't drive like a bag of dicks when cold either, nor is it stupidly loud for the first few minutes.
Win win.
Oh and fuel rail temperature compensation for the slope. The ID1000 spreadsheet suggests 0.956 at 200F and 1.067 at 0F, how do I interpret the stock table. It has 0.0 at 200F and 0.5 at 450F
What is this adding, injector offset? or a multiplier for injector slope?
It would be a multiplier to slope but we don't have a rail sensor so sweet out of luck there, but how can you incorporate something into the tune to help this situation ?
FG F6 400+RWKW .....10.9 @ 132mph.... FIRST 10sec F6 in SA ( sold )
DYNO DYNAMICS 1200hp ...... Tuning weapon..... ( sold )
OFFICIAL JONNY TIG INDUSTRIES SUPPLIER ..... Biggest and lightest intercoolers for B and F falcons
SUBARU BRZ ......AVO TURBO KIT coming......soon
VF R8 CLUBSPORT SV...... 340FWKW.....ASE BIG SINGLE KIT .....one day
Jet, I would like your analysis on this but as IH8TOADS posted here:Originally Posted by JETURBO
http://www.hptuners.com/forum/showth...l=1#post342325
the rail temperature is inferred so I am interested in why you think we are "out of luck there". Just like we don't have other temp sensors but the tune is altered according to their temperature. Like to get this resolved as CAT600 is also of this belief but I believe toads is correct.
Added 0.00003 to the offset and a tiny bit to the low slope, now it idles at 1.00 -> 1.02 and 0.97 at 1-2k, everything above 2k never drops below 0.98/0.99
I don't think my wideband is accurate at the 12.0 rich ranges as the trims are great but with closed loop disabled commanded 11.5 ends up with 12.8 according to my wideband. My datalogs of my existing tunes are in the high 12s as well which don't match any of my dyno runs.
Going to revert to logging in lambda and see what I get.
I thought that rail temperature was just based off coolant temp. Eg the same as how they estimate cylinder head temperature. I imagine it would be very linear compared to coolant temp with some lag.
Why would they fill the tables out if it isn't used?
Rolls, there is a Cylinder Head Temperature sensor so no estimate on that. It is mounted at the rear of the cylinder head.
The Aussie strategies have Fuel Temperature as a logable DMR, however it doesn't move off 60. There is inferred temp for CAT, EGO & EXHV which are all active and with respond with throttle input. The Aussie strategies do have hot injector compensation tables based on soak time, Air Mass, ACT, ECT and Inferred rail temp. The relevant tables work together with the end result being a correction factor for injector pulse width. The activity looks to depend on the Strategy, with one of the multiplier tables being zero.....often.
IH8TOADS, thanks for that.
I will see if I still have an old log but mine did change but not by much and that is why I thought you were on the money with the original post. As you say if one of the multipliers is zero then they are probably not using it and it might just be a left over. I logged injector slope modifier once and it stayed at 1.0 so it is not a factor in the calculations as well.
Yeh what froggy said
It would be nice to have the sensor there but these cars were built for stock applications and the sensors that are there care good enough to have a compensation for what is needed on a stock level, it's only when we start replacing hardware we would love to have further more precise tune ability
Can only piss with the dick we have
FG F6 400+RWKW .....10.9 @ 132mph.... FIRST 10sec F6 in SA ( sold )
DYNO DYNAMICS 1200hp ...... Tuning weapon..... ( sold )
OFFICIAL JONNY TIG INDUSTRIES SUPPLIER ..... Biggest and lightest intercoolers for B and F falcons
SUBARU BRZ ......AVO TURBO KIT coming......soon
VF R8 CLUBSPORT SV...... 340FWKW.....ASE BIG SINGLE KIT .....one day
Cool interesting info.
So I ran into a problem, possibly due to messing with the offset. At idle it will sit around 1.00 long term trim then occasionally drop to 0.78 and go lean about 16.8:1
If I tap the throttle or turn the Ac on it will go back to 1.00ish and the afr will come good agian
Any ideas what could cause this? I believe the stock o2 is still in place so will log the volts to see if it is giving a false reading.
Under some circumstances it will go troppo, especially when you are not where you should be with scalling.
Rolls, I suspect high slope is too high. Don't know how you are going to fix this if you don't think your wideband is accurate.
I just played with the ID1000 like injectors in my car. Very quick to get them pretty much spot on. ID1000 data and just scaled both slopes until LTFTs went to 0. Unfortunately the damn LPG injector with mods did not improve enough to get to the point where the two injectors would work and the LPG was sitting with LTFT at about +24%.
FYI I could rev the car out to 4500rpm and it was still the low slope so don't think revving the engine will get you there, you will need some load to test.
Why would my idle/cruise be fine 99% of the time with it reporting 0.78 trim and going lean at idle occasionally? At cruise and medium load the mixtures seem spot on with trims around 0.99 - 1.00, idle for the first 20 minutes or so is around the 1.00 figure as well maybe a few points either side.
The only time it seems to go wrong is after driving for 20+ minutes and sitting at idle it will just pull fuel making idle go lean and pop, apply the tiniest amount of throttle and it goes back to 0.98 trim and back to 14.7 afr, turning the AC on also fixes it.
It is like for some reason it is just using a wrong trim figure at idle occasionally leaning it out.
You have the car, you're the best person to figure it out. Given that there is not much info to work with, apart from a lean condition, load a tune with a big scaling change. Reduce the low slope by 5%. Ignore what the trims tell you for now.
Unless you log pulse width you will have no idea which injector slope you are on. Mine cruises fine on the low slope at 60kph and that is with much smaller injectors. What have you calculated your change over pulse width at?
Also don't forget fuel trims take a while to settle after you put a tune in.
Now for your other question. In SCT you can log Aircharge and you will, more than likely, find your car with an idle aircharge of 0.000400lb of air which translates to about 12mg of fuel when the accessories are on. Using standard ID1000 data for our pressure this is pretty dam close to the change over point for and ID1000 so idling with air on fans kicking in will go onto the high slope. Changing slopes will alter this.
Without accessories being on I expect your car to idle with and Aircharge of 0.000272 about 8mg fuel. So it will be on the low slope here.
Last edited by DarrylC; 10-22-2015 at 05:22 PM. Reason: Change 0.000400g to 0.000400lb
I see so perhaps I'm not on the low slope as often as I believed. I will recalc my changeover point and do some more datalogging. Have just reverted the tune as I haven't had time to do much more this week.
Just FYI eric from hptuners has updated VCM suite to allow reading of scrambled OSID files that were created in XCAL4. He even went to the effort of adding this functionality over the christmas break. 3 cheers to eric for being a top bloke
Reply With Quote