×

Sale ends todayGet 30% off any course (excluding packages)

Ends in --- --- ---

Afr and black haze

Understanding AFR

Forum Posts

Courses

Blog

Tech Articles

Discussion and questions related to the course Understanding AFR

= Resolved threads

Author
960 Views

hi guys I’ve been tuning on the road for 6 years now and finally switched to a dyno now most of my builds have a target of 500hp upward the thing that has got my attention is I can see quite a bit of black smoke at wot today I pulled 537hp from an 2.4l long rod Evo at 1.4 bar of boost and a steady 11.7 afr from 1 bar of boost onward all way up to 8k rpm should I be seeing black or is my well worked wideband I did run 11.7 on 3 more evos the same day all had black haze and seemed to do the right figures stock fq300 managed 369 flywheel Hp a few months ago I tuned a Nissan 300zx to 830whp I tuned this on a 20/80% meth/super unleaded UK totalling 26% alcohol content when viewed through an ethonal content sensor I tuned this at 770 lambda which was very clear at the exhaust here’s the thing I fitted a link thunder to the 300 and wired 2 internal both ls4.9 sensors could there be an issue with my ntk thanks in advance guys and sorry for waffling on

Attached Files

What you're seeing isn't uncommon unfortunately. While the measured AFR you're reading is a pretty good place to be for that particular engine, what the wideband doesn't tell you is the level of unburnt hydrocarbons in the exhaust which is what you're seeing as black smoke. There are a variety of factors that can result in black smoke at a reasonable AFR. These include injector design and spray pattern, fuel pressure, injector location and injector timing. The important part to understand is that in order for the fuel to mix with air and combust, it needs to be in vapour form. If fuel is pooling on the valve or port wall and entering the cylinder in liquid form it won't combust properly and this can result in black smoke.

Thanks for the reply Andre how would I go about diagnosing this and now you mention it I have only noticed black smoke on motors I’ve used I’d 2000s never noticed a problem with rc or pte 1000s could you look over the map maybe see if there is something there also would the inability to burn off the fuel result in less than expected power thanks

Are you actually giving the ECU proper slope and offset data?

Yes I believe so the only occasion I never had this issue with the id2000s was when I built one with flex fuel and used the modelled fuel stratagy instead of traditional

You're welcome to post a map here and I can take a look but I doubt that it's going to show up an issue to be honest.

Essentially irrespective of the characterisation data, if the measured lambda is on target (and your target is reasonable) then the issue is most likely to be related to the injector, injector placement, fuel pressure or injector timing. The characterisation data is important for good control of the injector but ultimately a certain pulse width is needed in order to supply a specific quantity of fuel so you could have completely incorrect injector data and then just bake in errors in the fuel table to get back to the correct AFR.

I hope that makes sense?

So I’ve had a good look over the injector settings dead times seem fine what I did notice was the injector timing on these two cars it’s at 340btdc these cars being Evo 7 and a 8 When I have looked through all other evos I have done I’ve been at 320btdc all except an Evo 9 I did a while back could it be the none mivec evos like 320btdc and the later mivecs want 340 and In my laziness have over looked this by using my same base as most of my builds are the same 2.4 long rod

Injector timing can certainly make a difference and it's going to be worth trying moving that around. 20 degrees however is a relatively small amount and usually a PI engine isn't super sensitive to injector timing. The timing also needs to be considered along side the actual pulse width which is going to depend on injector size.

The master pulsewidth is currently as low it can go think around 6ms like I said I don’t remember the same spec build smoking when I ran a modelled fuel I may just have spend some time setting up modelled and see if there’s a difference also could use a little help on my latest issue trying to setup traction control on the thunder in 950hp 300zx anyone have any experience here I would turn to Dave links Uk tech guy but he’s come home to you guys until the 28th