Board logo

still no joy with emissions and iva this friday, any honda enthusiasts give me some ideas?
daniel mason - 28/2/12 at 08:59 PM

following on from my thread a few days ago we still cant sort out co levels on the s2000 engine.
tried new cat and still no good, as far as im aware, i have only removed ac pump and fitted new belt to suit, blanked off the brake vacuum boost and disconnected air pump and blanked the valve off.
lambda and hydrocarbons good but Co way out still. and my test is friday so too late to cancel.

[Edited on 28/2/12 by daniel mason]

fesycresy - 28/2/12 at 09:05 PM

Are you sure the cat is 100% spot on?

daniel mason - 28/2/12 at 09:10 PM

well they have both been used on almost all mnr's so cand imagine both their test cats are goosed?

eddie99 - 28/2/12 at 09:12 PM

Are you definitely getting the cat hot enough? It's scary but you have to rev the nuts of it to get it hot.. Also is the cat close to manifold?

fesycresy - 28/2/12 at 09:15 PM

I assume they are testing at the end of the IVA with the cat virtually melting?

Do you know how many cells the MNR cat is and how many cells is on the S2000?

BigLee - 28/2/12 at 09:24 PM

How far away is the cat from the manifold? Needs to be quite near to get hot enough. If it 'ain't hot, it don't work!

johnay - 28/2/12 at 09:31 PM

Hi dan
This might be of some help, I found it on the net when you said you had removed the Air Pump

Quote
S2k pumps extra fresh air to the "exhaust" for cold start which can help reducing CO and help increasing catalytic converter heat up, in order to cut emssion.

Hope it helps I think the big BMWs use this.

It might just be bo77ocks

John

daniel mason - 28/2/12 at 09:43 PM

thanks for that,but i think thats just on initial start up mate when engine is cold.
cat is next to manifold and emissions done at engine tuners so they know what they are doing! still struggling though

l0rd - 28/2/12 at 10:04 PM

can you run both cats for IVA?

BigLee - 28/2/12 at 10:07 PM

Have you tried wrapping your manifold to keep the heat in?

franky - 28/2/12 at 10:15 PM

Maybe an air leak? how long did you run your engine for? It might need a good 25mins with high rpm to get properly warm. Whats your oil temp at? Ignore water temp, you need your oil at 80ish deg.

jack_t - 28/2/12 at 10:16 PM

drain the tank and fill it with shell v power and disconnect the battery for half an hour, the ecu wil reset itself anf v power burns hotter, also how new is the engine oil?

daniel mason - 28/2/12 at 10:18 PM

i think there is an underlyimg issue other then tha temperature of the cat. i am not with the car so not too sure if its wrapped or not? its still awaiting me to collect this week

afj - 28/2/12 at 10:19 PM

Well your out of time for any big changes. But I would try in the time you have available. New plugs. New engine oil. Clean/check air filter check any air leaks on engine and exhaust then fit a new lambda/ O2 sensor. Maybe some cataclean in the fuel tank. Then fingers crossed

daniel mason - 28/2/12 at 10:20 PM

engine oil brand new last week. not sure on oil temps as i dont have a gauge!
new plugs fitted 2 weeks ago.
new high octane fuel in tank

[Edited on 28/2/12 by daniel mason]

adithorp - 28/2/12 at 10:34 PM

What are the readings? That'll give us a chance of working out the problem.

The most common problem people have is getting the cat hot. The cat has to reach almost 500c before it does it's stuff. That'll take a least 15min at 3500rpm. Leave the test prob out while you do it or you risk melting it on a short system (like to guess how I know). If you haven't got a heat shield or the cat is a long way from the manifold it's harder to achieve. It'll do nothing untill it hits that temp.

franky - 28/2/12 at 10:41 PM

can you drive it there and leave it running? Seems like a lack of real heat or an air leak. As above it does take a lot to get the engine oil properly warm. My car failed until the oil temp went above 75deg C then passed every time, it wouldn't reach those temps just ticking over.

mikey p - 28/2/12 at 10:44 PM

check for exhaust leaks would be my first point of call.
you may well not be running lambda 1 even if the gas analyser says you are...
are you on standard honda ecu?

daniel mason - 28/2/12 at 11:01 PM

dta s60 ecu. cant understand why it wont meet readings.

daviep - 28/2/12 at 11:44 PM

Do you know what all the gas readings are, need to know HC, CO and lambda to make any kind of informed guess.

Did the tuner check the emissions without the cat fitted (or before the cat) to try and establish whether the cat was functioning.

Davie

stevebubs - 28/2/12 at 11:45 PM

My 2p - Air Leak / Lambda Sensor (perhaps a mismatch...)

PeteS2k - 29/2/12 at 08:09 PM

Same engine and ECU as mine. My exhaust silencer has a cat integral with the silencer can, which may have helped to keep it hot compared to a totally exposed cat. No problem with the emissions. I had to tweak the target lambda a teeny bit after a pre-IVA check on a local friendly emissions tester. Both my pre-test mechanic and IVA examiner were prepared to let the exhaust warm up for a good few minutes before taking the readings. IVA was passed with the DTA F20C base map, running closed loop for the bottom corner of the map. It seemed to sort itself out very nicely. I didn't get a full mapping done till after the car was on the road.

Are you letting the ECU 'self tune' closed loop?


Edit to add: I couldn't get the Idle Air Control Valve working with the DTA, so I just removed it, blanked the holes in the manifold, and set idle directly via the throttle stop. Needs some throttle to start (about 4%) and manually (or with the foot!) manage the idle for a couple of mins, but otherwise it's fine.

No idea whether the IACV would have an impact on CO, but it is another potential difference between yours and mine.

[Edited on 29/2/12 by PeteS2k]

daniel mason - 29/2/12 at 08:24 PM

i have had no input whatsoever in the set up.the car went to mnr for pre iva work,then to rolling road who have struggled like hell to get it sorted. so have no idea really

franky - 29/2/12 at 08:26 PM

did you pay for the 'map'? if so take it back and get it done properly? Have you checked for air leaks? do you know what the readings are?

You'll get there in the end

daniel mason - 29/2/12 at 08:38 PM

not paid for any of it yet. and my power map seemed good with 180 rwhp, and that was with the collector cracked and leaking. i believe this is sorted now. but cant get back to dyno before friday

franky - 29/2/12 at 08:41 PM

180rwhp seems a little low for a 240bhp engine? Maybe the crack affected the mapping so it'll need sorting/mapping again?

daviep - 29/2/12 at 08:44 PM

quote:
Originally posted by franky
180rwhp seems a little low for a 240bhp engine? Maybe the crack affected the mapping so it'll need sorting/mapping again?


Was the collector cracked when they were trying to set the IVA emissions? That would explain a lot.

Davie

daniel mason - 29/2/12 at 08:46 PM

honda produced figures of 237 but not accurate. im a member of the s2000 forum and they said to expect 190 bhp max at the wheels. probably less with a non standard manifold as the oem honda one is superb!

daniel mason - 29/2/12 at 08:46 PM

no davie. its been back since and still no joy?

rodgling - 29/2/12 at 09:04 PM

High CO means it's rich (if it had more oxygen available it would be producing CO2 instead). I would have a read of this:

http://www.autoshop101.com/forms/h56.pdf

which goes into a lot of detail about causes of various emissions problems.

DJT - 1/3/12 at 12:54 PM

On my Webcon Alpha set up I had to activate the timing retard function. Apparently this was built into the ECU software for Westfield (I think). When the timing is retarded the exhaust gases run hotter, thereby lighting the cat. I recently passed the MoT and IVA with the timing temporarily retarded. Without retardation it was failing. I've hard wired a switch so that I can activate this future MoTs.

The other thing Webcon advised was to make sure the tester is using a long enough probe as with the side exit exhausts the pulsations in the air pressure can draw fresh air back in to the silencer, which then gets sniffed by the probe and the readings messed up.

This was just my experience. Maybe something in there of help. Good luck.

PeteS2k - 3/3/12 at 08:53 AM

How did things go on Friday? Did you find the problem in time, fail on just emissions or postpone?

daniel mason - 3/3/12 at 09:52 AM

potponed again, have had enough of it all now. wish id never started

franky - 3/3/12 at 09:56 AM

You'll get there, could you not get a map off someone else to see if its just a mapping issue? did you find any leaks?

BigLee - 3/3/12 at 10:07 AM

quote:
Originally posted by daniel mason
potponed again, have had enough of it all now. wish id never started


Don't be down mate. If car building was easy everyone would be at it! This will buy you enough time to sort it.
Mine has been running rich for weeks. Been chasing my tail figuring it out. Changed pretty much every sensor, wire, coil pack etc to no joy. Figured it yesterday. Tiny leak on inlet manifold gasket. Simple things.

Best of luck

Lee

daniel mason - 3/3/12 at 10:20 AM

its nor running rich though. its had hours on the dyno and power map was good, but no matter what emissions could not be sorted, so as it stands ive got the cart back home in iva trim,imout of pocket and still high Co at idle and fast idle!
and i have no idea about mapping ecu's for emissions.
looks like im going to have to take it somewhere else to try but its expensive. if i could find a buyer id sell it in non road legal/track form

daviep - 3/3/12 at 11:12 AM

You need to tackle this in a more logical manner and start eliminating components.


  1. If you have high CO then you are running rich, it's simple chemistry - there isn't enough oxygen to burn all the fuel
  2. You need to establish that the engine is running nice and clean without a CAT
  3. If you cannot get the CO down to reasonable limits without a CAT then you need to diagnose which sensor is causing the ecu to over fuel - eg an exhaust leak will allow oxygen into the gas stream, the lambda sensor will see the oxygen and think that the engine is running lean and so start adding extra fuel to try and richen back up. You need to check for sensible readings from all sensors.
  4. Once you have it operating correctly in closed loop then fit the cat and compare readings to check for correct operation of the cat.


You are paying for a service from MNR and the tuner, you are entitled to detailed breakdowns of what was / wasn't tried or tested and also the relevant data which was produced when trying.

Just my 2p

Cheers
Davie

theconrodkid - 3/3/12 at 11:39 AM

if the CO is out then it,s too rich,try measuring the ohms readings of the water and air temp sensors and is the wiring to them ok?

adithorp - 3/3/12 at 12:29 PM

quote:
Originally posted by theconrodkid
if the CO is out then it,s too rich,...


Possibly, but it could also just be that the reading are too high to pass because the cat isn't reducing the CO. For example, 0.5%CO isn't rich for pre-cat readings but with the cat not working (either goosed or just not up to temp) it'll fail emmision testing as too high.

In order to comment on it and to give any meaningfull advice we need to know what the actual reading are.

daniel mason - 3/3/12 at 01:07 PM

i have a seperate air fuel ratio gauge not linked to ecu and at idle with the iva map,its up over 15-16. it has 2 cats on it now also. think there is a sensor not working somewhere but how do you go about eliminating these with no mapping gear? there are sensors everywhere and i am not too familiar with this type of fault finding?

[Edited on 3/3/12 by daniel mason]

daviep - 3/3/12 at 01:29 PM

quote:
Originally posted by daniel mason
i have a seperate air fuel ratio gauge not linked to ecu and at idle with the iva map,its up over 15-16. it has 2 cats on it now also. think there is a sensor not working somewhere but how do you go about eliminating these with no mapping gear? there are sensors everywhere and i am not too familiar with this type of fault finding?

[Edited on 3/3/12 by daniel mason]


You need to identify which sensors are feeding the ECU, you will have all or some of these: MAP,TPS, coolant temp, air temp, lambda.

You then need to find the spec sheets of the relevant sensor which will tell you what the resistance / output of each sensor is at certain values.

15-16 is already running lean, you really really really need to find out what the actual readings were and under what conditions and also whether base line readings were taken before the CAT(s) were fitted.

Once again you need to be logical and methodical to avoid wasting time / money no point and guessing about sensors you need to establish whether there is a fault with the cat(s) by testing with and without them. and then if the cat(s) are OK you then need to identify where the fault lies.

Davie

daviep - 3/3/12 at 01:34 PM

quote:
Originally posted by daniel mason
i have a seperate air fuel ratio gauge not linked to ecu and at idle with the iva map,its up over 15-16. it has 2 cats on it now also. think there is a sensor not working somewhere but how do you go about eliminating these with no mapping gear? there are sensors everywhere and i am not too familiar with this type of fault finding?

[Edited on 3/3/12 by daniel mason]


Are you running 2 lambda probes or are the gauge and ECU looking at the same probe?

If they are looking at the same probe then it is possible that the lambda probe is giving a false lean reading, either due to air leak or damaged lambda probe.

Davie

daniel mason - 3/3/12 at 01:46 PM

I nEver wired the engine/ Ecu but can't see how Ecu can be connected to an O2 sensor as the exhaust only has one which is totally separate fro
The Ecu. Or do these Ecu's have lambda monitoring built in by measuring fuel and O2?

BigLee - 3/3/12 at 06:18 PM

A cheap test to eliminate air leaking (as suggested by several forum members) is a can of brake cleaner. Leave the engine idling, then spray each point of the intake from MAF to head. Listen to the engine note. You'll know when you find it. That's how I found my running problems. For £3.50 it's a simple test to find simple problems. If there is a leak and you stop it, you'll need another mapping session as it'll change so much. You may get a better headline figure too. Chin up Dan. You've built a cracking car there mate, don't let something like this ruin it for you.

PeteS2k - 4/3/12 at 08:33 AM

If your tuner managed to get a good engine map, with a smooth idle, no rough running or misfires, then things can't be that far off. It will be something simple, just a case of finding it!

When looking for air leaks in the exhaust ahead of the lambda probe, don't forget to check where the auxiliary air pump valve sits. (The bit that sits towards the back of the engine on the intake side). I think you've got this removed and blanked off, like I have. If not properly sealed, it could suck air through into the header. Could be missed if just looking for leaks in the exhaust side.

Also worth checking with your ECU installer and tuner to check whether or not the ECU/map is using the lambda or not - and if the lambda the tail-pipe probe agrees with your AFR gauge (and/or ECU, assuming it's wired in). This could highlight if you've got any leaks between your lambda sensor and the tailpipe. If you're trying to tune for emissions, and the two probes are seeing different values, you could have the ECU battling against you!

I had a few issues when getting my engine running initially - including crank, cam, MAP sensor and Idle control valve wiring. From my experience, if there was anything wrong with these, you wouldn't have got as far as you have - and your tuner would've spotted obvious bad readings, so I would reckon your engine sensors will be ok.

Hope you get it sorted soon.

[Edited on 4/3/12 by PeteS2k]