Welcome to the SR20 Community Forum - The Dash.
SR20 forum logo

Thread: Calling all engine experts - Smog from hell

+ Reply To Thread
Posts: 21-30 of 30
2016-08-13 18:21:22
Originally Posted by Dala
So now I have spent >700$ trying to get this heap to pass.
Do you have anything in Finland like we do in the US? Most of our states let you get a "pass" for the year if you can prove with receipts that you've spent over $500 trying to fix your emissions situation. Do you have anything like that?
2016-08-13 19:59:09
For Shits and giggles you should try going back to a Maf
2016-08-13 21:52:36
Originally Posted by BenFenner
I haven't read the entire thread, an no doubt you're getting good advise from people with way more experience than me (it's been 13 years since I had to deal with emissions tests) but isn't the O₂% (Oxygen percentage) too low in all of your tests, which indicates a rich mixture, which could explain your high HydroCarbon readings?
You then went and added larger injectors and made it worse (even with returning) because the atomization no doubt is worse on larger injectors and then added even larger injectors which made it just as bad (even with the retuning). HydroCarbon readings are basically gasoline readings, right? Unburnt fuel?

O₂ should be low. Complete combustion only produces water(H₂0) and CO2. That is why there is a limit in the smog test, so people dont show up with holes in their exhausts.
If potential unburn fuel got though, closed loop will add more fuel in order to reach stoich.

I'd be curious what you'd see if you went to the place you went to do some testing and instead of messing around with the timing so much if you got those AFRs in check. Raising the base fuel pressure could help with atomization as well. But frankly I'd be trying with the smallest injectors I could.

AFR's have always been in check. Closed loop is utilized to get the AFR to swing at +-14.7AFR, to get the three way cat to do its thing. There is no other AFR adjustment to be done, other than getting the map as close to stoich as possible, making the closed loop corections as minimal as possible, preferable max 1%+-

My 259cc injectors are 3/4 , one broke with infinite resistance.
My 750cc injectors are subaru decapped, they flow like a fire hose. No good for emission testing
My 1000cc injectors are top notch Bosch, flowmatched to within 1% tolerance. Should have no issue getting these to pass.

No difference between 3Bar, 3.5Bar, 4Bar pressure. 220HC'ish at idle. Something else than fuel supply is the issue.

Edit: I see lots of talk about poor ignition causing high HC. That makes total sense to me, since that too would cause unburnt fuel to pass through to the exhaust. However, I think the low O₂ readings are telling a different story. If the ignition were bad, I'd think you'd see high O₂% and high HC... But since you're seeing very low O₂% and high HC I would think the fueling is the issue, not the ignition.

Yep, this is the best theory so far. Low O₂ can be attributed to slow filtering of smog equipment. It seems to do some very heavy filtering to the values, often taking several seconds to react to changing parameters. If you look at Lambda on the smog, vs Lambda on the wideband, there is an insane difference in responsiveness. The smog machine simply do not capture the misfires.

Originally Posted by BenFenner
Do you have anything in Finland like we do in the US? Most of our states let you get a "pass" for the year if you can prove with receipts that you've spent over $500 trying to fix your emissions situation. Do you have anything like that?

Funny that you mention it, you can re-register the car as "Ei Vähäpäästöinen" which translates to not environmentallyfriendly, and pay 5X the normal tax on it. That translates to circa 1800$ per year. Not really keen on doing that yet

Originally Posted by lynchfourtwenty
For Shits and giggles you should try going back to a Maf

MAF is long gone, threw it in the bin as soon as I went MAP. MAFs have no place on a modern engine. MAF wont solve this one. Calculating engine load is all that does, which is working perfectly here at no load! Idle and fast idle for smog.
2016-08-14 04:05:14
Originally Posted by Kyle
All the peppering is too much heat, detonation, and/or a dirty chamber. My B14 plugs looked just like that when I was running too much timing, with incorrect t-stat, and my ignition system was plagued by loss. I had moisture corrupting the signal from the distributor to the wires to the plugs. Huge mess, but those plugs he posted looks similar to mine. Maybe I have them somewhere....

Appreciate this post, I see the peppering in Dala's pics too now.
2016-08-15 14:12:52
Originally Posted by Andreas
If you put on a MSD ignition with coil it would clear up that mess

I totally respect the person giving this advice, but I'm hesitant to recommend adding parts when there may still be some mechanical or electrical issue causing the problem. You always need to tackle the underlying problem and get a handle on that before complicating the setup with non-standard parts. If you're back at a NA setup, the stock ignition if it is in working order wouldn't be causing this problem. And if the stock ignition is fine, going with a fancy ignition box would be wasted money.
2016-08-15 18:33:14
There is some great advice in this thread. I also talked to some colleagues, going to borrow a compression tester for the weekend. The two main theories right now;

1. Spark issues
2. Valve damage from running lean

Spark is gonna get totally renewed this weekend too, already have all the parts for COP conversion, so why not. If that doesn't change anything, I'm gonna do a compression test to see if my valves are sealing like they should. I had some lean mishaps with the 750cc injectors, but not for very long. These motors probably can take 13.0 WOT F/I briefly?

I'll slowly update this thread as time goes on and I rule out more stuff. The last day is 26.8.2016, after that it becomes illegal to drive.
2016-08-15 18:41:09
A leakdown test would be easier and prove to be more valuable of a tool if you are having cylinder loss. (along with not compressing everything via battery and starter. You turn the engine over by hand for each cylinder to be at TDC, firing aka all valves closed and ready to spark the combustion inside. Any leakage, you will hear where the leak is. The timing chain area, means the pistons are leaking. If you hear it up top, then the valves are leaky.
2016-12-28 19:06:00
very late reply since i don't come here much anymore. But i feel your pain. I live in CA and every 2 years, i have to go thru the same SMOG test that usually fails the 1st time. Last year, it failed 3 times and luckily, my last resort, which was to replace a CAT, passed. I have a stock SR20de. So my advice is, is it too late to go back to stock? Not sure if it matters or makes a diff, but get some stock injectors, get a stock fuel rail and stock ecu. Clean out the EGR, new plugs(I get the NGK copper and replace them every 2 years before the test) and maybe new wires. Get a new fuel filter and a fuel cap. Seems like you already went thru the trouble of checking everything else. Good luck.
2018-06-04 11:28:14
Thread revival!

Decided to have another go at the NX while I construct my EVNX in parallell. The EV is more of a long term project, but it sure would be nice to get some ttop action this summer...

So, troubleshooting continues.

Pulled the car out of 7month storage, It was quite dirty:

Washed and ready to work on:

First order of business was to perform a cold compression test.

Seems good enough for a VE with metal HG?

Then I started it, still slightly misses at >2200rpm;
-Took an oscilloscope, and counted ignition pulses coming from the AEM COP driver. No missing pulses were observed at 3000rpm
-Also checked the injector driver to Cyl4, whilst locked at 2ms injection duration@3000rpm, no change in pulse waveform is seen when misfires occurs.

Then I played more with the tune.

-No closed loop. No knock correction. No offsets.
-Running pig rich, 11.x AFR is not affecting misfire. Still misses
-Running lean, 15.x AFR will maybe cause it to slightly miss more? Dunno really, kinda hard to hear.
-Running with low ignition timing (20*@3000rpm) will not affect misfire
-Running with high ignition timing (30*@3000rpm) will not affect misfire

Other oddities/stuff noticed
-Load signal is good, misfires with MAP or TPS as load source
-Knock voltage jumps between 0.5V -> 2.5V when the car misses.

Eh, this pile of scrap is not giving me an easy time. Will try to measure all 4 injectors along with 1 channel for ignition, and trend everything on a better scope. I am highly suspecting something mechanical still...
2018-06-04 14:47:53
I have a picture what I believe are good plugs, though from a SR20DE.

For passing emission test I had this funky thing, I own three MAF sensors and one I don't know it's voltage and the others are 0.7 volt and 0.8 volt according to datascan.

Now I couldn't tell you still have a MAF sensor in a DET setup and I know nothing about turbo's so if it looks ambiguous what I write then it's probably me.

The funky thing, I couldn't pass easy with 0.7 volt MAF but I didn't know, the expection guy stomped the gas for a while three times and said, passed and wasn't too happy about the smoke in his diy garage.

Now I switched MAF for mpg issues (no it didn't better) and I passed with one rev at 3000 rpm with the 0.8 volt MAF. Though another garage but it was a less then 30 second job for the inspection guy.

I can't explain how so but maybe another thing to throw in your puzzle, I don't expect that debris on your plugs photo to go away and I do have a new NTK lambda it didn't had the Nissan logo but it's NTK for my DE. I also hope to get notification if there are replies after me and hope that these $0.02 are helpful. Btw ashes on plugs I think about burned oil, oil has ashes and as far as I know gasoline doesn't. For the cat, I sure hope to find a used original one, the new ones look weird to me.
Last edited by richardwbb on 2018-06-04 at 14-49-39.
+ Reply To Thread
  • [Type to search users.]
  • Quick Reply
    Thread Information
    There are currently ? users browsing this thread. (? members & ? guests)

    Back to top