Discuss all things tuning in this section. News, products, problems and results.
got a customer car in : Nissan RB25det in s13.
just a normal engine with a big turbo on it , cam in for a run-in tune. Car was a non starter when arrived.
its on a nistune ecu. it has a big hpx maf , reacts a bit odd in the numbers.
problem is i cant even get it dialed in to just run on normal afr's . It idles odd , only runs on 0.8 lambda or richer. when leaner it starts to misfire. checked with 2 wideband systems. everything checks out mechanically. Maf is pre throttle body after intercooler, which is unusual for me.
when it came in it was on possible china injectors , got those swapped to original ones , still the same allthough its less rough. still doesnt take less than 0.8 lambda
What am i missing here ?
Have you tested your exhaust for leaks? MAF wiring? intake air leaks? Fuel pressure?
yes all is pressure tested , maf signal is fine , fuel pressure is set , just confirmed it with an additional gauge.
why would an engine not run on stoich ?
Perhaps the fuel is bad. Can you feed it from known good fuel.
all drained and put fresh fuel in , no difference
would bad injection timing cause this issue ? i would assume it does but not this much
Have you checked the spark plug gaps? Sometimes people use na spark plugs for turbo applications and it causes poor idle quality and misfires under load.... Also wrong camshafts alignment can be the cause of such engine behaviour.
"Big" higher overlap cams/ports on a rotary will want a pretty narrow lambda range for stable idle, if cam timing is out it will be very fussy. If they are shitty modified injectors you may be fighting minimum reliable fuel volume/atomization quality too.
High overlap rotaries usually like afr around 12.7, injector phasing can help especially if they are big, may be worth shifting it around for best idle quality.
its on stock cams but i will check timing marks next time , could explain a lot indeed
i've checked about everything , no solution found
everything is fine : all signals , all feeds and grounds , all signals.
mechanically everyting checks out , timing is on point , ...
i can make it run and drive , on the wrong afr value. Its happy with 10.5 on cruise.
pulling one injector at a time results in leaning out by the same value every time , so injectors are pretty balanced i believe
and its not bad wideband reading , tried 3 widebands by now , and its getting fuel in the oil , so yeah
open to hear from someone who has a clue about whats going on here
Since I'm not familiar with the nistune ECU -- what is the load axis for ignition? Have you changed / swapped the sensors used for that load. There might be an issue with one of the inputs that is causing a timing change related to load (causing the misfires). Or something in injector definitions that causes it to think a different mass of fuel is being used (some factory ECUs change timing based on fuel mass per cycle).
A puzzler for sure...
The other thing to check is the actual ignition timing vs commanded value. Also ignition coils themselves better to be double checked.
Whenever I start seeing odd lambda values as opposed to drivability I always try to find some way to check mixtures via old(er) school means such as EGTs and reading plugs. I know you said you checked different O2 systems but quite often we can't see the forest for the trees when it comes to technology we come to rely on. Also the good ol' schnozz test is useful for me on diagnostics.
Not an engine I've played with, so I may be talking out my bottom... ;-)
You said you checked it against two wideband systems - were these completely independent of the vehicle's system?
Three lines of thought -
There may be a problem with the Nistune board or wiring.
There are two types of lambda senders - is it possible there is an incompatability between what you're using and what the ECU is expecting - that may throw off the reading? IIRC, the RB used dual lambda (certainly might be wrong), if so, that may be throwing things off?
Is the lambda reading you're using raw data, or is there a conversion somewhere? I was wondering if it may have some conversion along the way?
Nistune have a couple of FAQ boards and a forum, which can be found from their home page - https://www.nistune.com/ - and it may be worth checking there for folks that have had similar issues.
Their 'general FAQ' has several comments about compatability issues different "RB25" versions, also under 'support' is a forum with several wideband specific discussions, that may be useful. Perusing the whole site may be worth your while, in case something triggers further thought.
I'd be inclined, at this point, to dismiss an exhaust leak, or a misfire, as those would show up as a lean mixture (oxygen content in exhaust), not a rich one.
Not if, but when, you find the solution, please return and post your findings - it may be applicable to other platforms.
the plugs are black , smell like fuel. There's fuel in the oil allready , so i assume it is effectively rich
the wideband isnt connected to the ecu , i used 3 different wideband with 3 different controllers. The ecu isnt in closed loop so its not reading any lambda sensor.
i'll be plugging in a Link ecu i got on the shelff so i can run some more tests and have a little more control.
allthough i dont think the ecu is the problem. I can control effective injector pulsewidth with the settings in the Nistune ecu.
What is the injector size? Possibly too big and not stable at low pulsewidth needed for leaner than 0.8? Have you tried a lower fuel pressure and higher pulsewidth? Any different?
stock injectors which are 370cc only
upload a log and the tune file so we can actually check.
have you verified all injectors? Low/high impedance? resistor pack present or not? (depending on Spec 1/2/3 engine Spec 3 is stupid rare)
load in a base map from the ROM pack, and start by getting the Lambda in the ball park through modifying the MAF calibration. Assuming the injectors are stock. Alternatively you could always approximate the PW you need at idle (roughly) by simple maths. and go from that.
I've worked with several NIStune cars S13/S14/R32/R33/R34 and find that some setups work, others don't.
Big MAFs tend to have somewhat different curves, depending on where you operate it. Assuming stock fuel setup with a big MAF you'll never need the exponential part of the curve, as the voltage is the load scale in NIStune. The bigger the MAF the lower the max voltage you'll achieve. The more sensitive it'll be to pulses too.
Can you confirm all 6 coils/spark plugs are firing?
Rough idle, doesn't run leaner than 0.8 lambda suggests to me one of the cylinders is taking in fuel but isn't igniting it throwing your whole AFR off.
THIS IS FIXED
i had everything ruled out and there was nothing electrical/mechanical wrong , not a single suspect of the missing. Took a Link ecu out of another car , loaded the basemap in , synct timing , started it up and drove around the neigborhood while adjusting the fuel map. No hassles
Nistune ecu was sent back to Matt at nistune. He confirmed the software was faulty.
There are times when clockwork ignition and carbies seem a better alternative...
It's great you got it sorted - if it was a corrupt install, it should be relatively simple to have it reloaded?
thats why i thought to. i've reloaded everything multiple times , didnt help.
Also money was immediatly refunded when they test the ecu themself , bit suspicious if you ask me.