Scott.T

iTrader: (2)

My apexi seems to struggle on the fast idle test during an MOT. This is with standard 440cc injectors, standard pump pressure and running 'Closed Loop' O2 feedback control. Tested twice, first time with OE CATS (both) and second time a Sports CAT.

When held at 2700rpm during Fast Idle the CO goes up to approx 0.8% and the rpm is hunting by approx 50rpm.

In 'Open Loop' the rpm is rock steady, but I havn't had the chance to do the emissions test in this configuration yet.

Swapping back to a TEK'd OE ECU the rpm at 2700rpm is also rock steady but obviously the OE ECU is running 'Closed Loop'.

I done a few roadside checks to compare AFR of Apexi Vs TEK's ECU

APEXI

The OE sensor (in header) is being used to run closed loop and it's running (according to the techedge in the downpipe) 14.4-14.8 swing at idle closed loop and 14.4-14.9AFR @ 2700rpm and has audible & visual hunting by about 50rpm with the occasional pop/missfire.

When I was running NBsim it was also running 14.4-14.8 swing at idle closed loop (not sure about static 2700rpm, but on the road it looked fine)

I have checked it in 'Open Loop' and at 2700rpm I have set AFR = 14.7 (approx)and the rpm is rock steady.

As it happend with both the NBsim and the OE is this a feature of the Apexi or is there something that needs a tweak.

OE MY00 ECU

Re-ran the test using my TEK ECU (AFR check, not MOT emissions test).

OE Lambda Sensor controlling Closed loop :
Idle 14.67 - 14.98AFR
Fast Idle (2700rpm) 14.7 - 15.0AFR

It did not hunt, but held steady at 2700rpm, unlike the Apexi

The closed loop update rate of the OE ECU is alot quicker then the Apexi, so I think the Apexi may be too slow at 2700rpm to hold a rock steady rpm.
The slight 50rpm oscillation may be the result of this.


It may be a case that with the APexi it may need to be set to 'Open Loop' for an MOT, hence the title of this post.
But me being me, would like to get to the bottom of this and understand what's going on.

I will be re-running the MOT tests again one evening this week, so will post back more info....

cheers...

Scott.T

iTrader: (2)

Anyone ????

Scott.T

iTrader: (2)

For those that are interested...........I was trying to kill 2-birds with 1-stone for this test.

When I had the car MOT'd at the beginning of the month it 'FAILED' the fast idle test running with the Apexi ECU and the two OE Catalysts fitted back into the exhausts.

Due to me having a knightmare with bolts and studs when refitting the decat, I thought 'sod this every year' and went and bought a Ninja Sports Cat.

However, I was concerned the Sports-Cat was not upto the Job, so paid a visit to my local MOT staion for a check. I came away with exactly the same 'type-of' result as with the OE cats i.e PASSED natural idle, 'FAILED' fast idle.
After a bit of investigation and some tweaking/correcting of the Apexi base map I thought I would give it another try........

I Went down to MOT station today after I tweaked the open loop fuelling of the Apexi to be 14.7AFR approx from 800rpm through to 3000 (static rpm, row 2 on the Apexi map).
This seemed to help the stability of lambda and rpm when running closed loop on the Apexi.

I went down with a plan to try Apexi Closed Loop, then Apexi Open Loop and finally TEK OE ECU to see which I could get to pass.

Went straight in for the kill with the APexi in Closed Loop and it....PASSED

We started off just monitoring the varios values CO, Lambda, HC etc..etc.. and idle was fine and fast idle was also very good, so we decided to go through the full MOT procedure.

Unfortunatley during the 'quick' MOT sequence I had foot trouble and went outside the range, so being the new computerised system it then defaulted to the longer procedure.

Results were :

Fast Idle test :
Engine Speed : 2450-3050rpm Result 2640rpm PASS
CO : max 0.300% Result : 0.297% PASS
HC : max 200ppm Result : 43ppm PASS
Lambda : 0.970-10930 Result : 1.008 PASS

Natural Idle test :
Engine speed : 750 - 1050rpm Result : 850rpm PASS
CO : max 0.500% Result : 0.152% PASS


These values were up slightly compared to the dry run we done beforehand, but I was holding the revs slightly higher for the actual test.

I am quite happy with this result considering it's running the Apexi in closed loop and also running with a Ninja sports cat. Especially as my tests over the weekend showed that the Apexi Closed Loop control isn't as tight as the OE ECU, so an OE ECU should get through with the Sports Cat even easier. The tweaks made to the Apexi were by no means required to get the Sports-Cat to pass, but were corrections made to get the Apexi to run closed loop a little better, infact I think there are still some minor improvements that can be made

Find below the results of my previous MOT's inc Catalyst and ECU tune. Even some of these are close one way or another, which just proves what a farce this MOT emissions test really is......

2006 MOT :
Car Spec : MY00, Apexi ECU, VF28, Both OE Cats, 440cc Inj, STi Panel Filter

Results were :

Fast Idle test :
Engine Speed : 2450-3050rpm Result manual check PASS
CO : max 0.200% Result : 0.194% PASS
HC : max 200ppm Result : 3ppm PASS
Lambda : 0.970-10930 Result : 0.994 PASS

Natural Idle test :
Engine speed : 750 - 1050rpm Result : manual check PASS
CO : max 0.500% Result : 0.162% PASS

note : The above test had a manual rpm check as it failed fast idle due to the Apexi issue described above.

2005 MOT
Car Spec : MY00, TEK3 ECU, VF28, Both OE Cats, 440cc Inj, STi Panel Filter

Results were :

Fast Idle test :
Engine Speed : 2450-3050rpm Result 2682rpm PASS
CO : max 0.200% Result : 0.00% PASS
HC : max 200ppm Result : 13ppm PASS
Lambda : 0.970-10930 Result : 1.004 PASS

Natural Idle test :
Engine speed : 750 - 1050rpm Result : 805rpm PASS
CO : max 0.300% Result : 0.24% PASS

2004 MOT
Car Spec : MY00, TEK ECU, TD04, Centre CAT only, 440cc Inj, STi Panel Filter

Results were :

Fast Idle test :
Engine Speed : 2450-3050rpm Result 2776rpm PASS
CO : max 0.200% Result : 0.18% PASS
HC : max 200ppm Result : 19ppm PASS
Lambda : 0.970-10930 Result : 0.998 PASS

Natural Idle test :
Engine speed : 750 - 1050rpm Result : 794rpm PASS
CO : max 0.300% Result : 0.30% PASS

2003 MOT
Car Spec : MY00, Standard ECU, TD04, Both OE Cats, 440cc Inj, K&N Panel Filter

Results were :

Fast Idle test :
Engine Speed : 2450-3050rpm Result 2757rpm PASS
CO : max 0.200% Result : 0.00% PASS
HC : max 200ppm Result : 9ppm PASS
Lambda : 0.970-10930 Result : 1.005 PASS

Natural Idle test :
Engine speed : 750 - 1050rpm Result : 796rpm PASS
CO : max 0.500% Result : 0.49% PASS

Jay m A

iTrader: (2)

Scott FYI I have an Apexi and it passed on closed loop fast idle with a twin cat centre section in place, Andy.F mapped and I haven't changed a thing.

MarkFitz

are you sure you can run an O2 sensor pre-turbo?
could be worth logging the O2 reading to see if it's stable...might explain why the CL control is all over the place.

vulnax999

iTrader: (7)

Early cars had the Lambda / O2 sensor in the header collector before the up-pipe, and worked ok!

MarkFitz

its a 00MY

Scott.T

iTrader: (2)

Works fine...
Previous emissions test was with Wideband in the downpipe running narrow band simulation into the Apexi, this gave the same results even with both CATS fitted..

I could probably now pass running the narrowband simualtion, now that the open loop fuelling of the Apexi is set up better.

The OE was only put in the header to try and run tighter then the narrowband simulation.

The OE ECU does run tighter Closed loop control then the Apexi, but only marginally once the open loop Apexi map is set closer to 14.7AFR.

It turns out that if the Open Loop base map is not close to 14.7AFR i.e in the high 13's or mid 15's that the Closed Loop can struggle. Any minor change in rpm causes the ECU to use the Open Loop base setting then try to achieve stoich. The closer the Open loop is to 14.7 the quicker and more stable Closed loop stoich control will be.

I don't think the position of the Lambda matters much, although they do take a bit more aof a hammering in the header.

MarkFitz

not sure if this helps, but my base fueling at 0 load is...
0.754 (800-2800rpm), rising to 0.844 at 5200rpm+.
timing wise i've got...
12 deg @ 800rpm, rising to 40 deg from 2600rpm+

based on 97MY UK. (DeCat DP Only, StgII Legal)

Scott.T

iTrader: (2)

Cheers, but won't help as it depends what the injector and maf scaling are set to.
Your probably running a MAF rescale to run with a UK MAF instead of the default Apexi STi MAF and also maybe an injector rescale to run 380's instead of the Apexi default of 440's.

There are several ways to skin a cat on the MY97/98 and yours should have done one or the other (or both).

MarkFitz

you got it, power fc set up to run a UK car with 380's+oem maf/green panel....should have mentioned first time. Cyl trims and map resolution will also differ so absolute values will be of no use but trends might give you an idea of how well the mapping has been done in the region your focusing on.

Richy B

iTrader: (3)

Scott, how have you got on with this?

My 22B is also having problems passing the MOT with the Apexi in place. I had a ScoobyWorld Ninja Sports Cat Downpipe and it wouldn't pass emissions with either the Apexi or OEM ECU fitted. I now have the Hayward & Scott Sports Cat Downpipe fitted and it passes the emissions on the OEM ECU but not with the Apexi ECU.

I am running standard 22B spec (VF24 turbo, 2.2 displacement, 440's, STi TMIC) except for GT Spec Headers, Blitz Nur Spec R and the sports cat.

Scott.T

iTrader: (2)

Hi Richy,

I guess it was your thread on 22B then....it made be double check the Emissions when I fitted the Ninja. Initially I got the same result as the OE CATS i.e failed fast idle.

As mentioned above I went back to the MOT station once I had plumbed in an OE Lamda into the header, as I was initially running with the Wideband Lambda in the downpipe producing a Narrow Band simulation into the ECU.

I thought this narrow band simulation may be a problem hence why I put an OE Lambda in aswell.

I went to the MOT station armed with the Apexi and OE TEK mapped ECU with the aim of seeing which would get through the test.

My options where :
Apexi Closed Loop Control
Apexi Open Loop
OE TEK

Since it's last failure, as well as fitting the OE Lamda had been playing with the Apexi map to try and get closed loop running a bit tighter. And to my suprise the MOT test passed running with the Apexi in Closed loop mode, so we didn't bother running the Apexi in Open loop or the OE ECU.

I expect you need to ensure that in Open loop mode the Apexi is running at 14.7AFR approx at idle and fast idle. after mine failed I checked the open loop fuelling and it was running around 13.9 at 2700rpm.

I then set mine to run 14.7AFR (approx) from idle through to 3,000rpm i.e 800, 1200, 1600, 2000, 2400, 2800 & 3000. all of this was done by adjusting the Cells on row 2 of the fuel map (check on the comander map trace what cell it sits in then adjust this and the corresponding rpm cells described above).

You will need a wideband lamda do to this though. If you havn't got one then the MOT station lambda probe up the exhaust would suffice.

When you then switch back to Closed loop, it should run a bit tighter and get through the test.

The other option is to run Open Loop during the test, which will cause it to run at what you set it too above. This was my next plan of action that didn't need doing.

I hope this helps......Good luck....

Andy.F

You can do that with the OE narrowband signal read from the commander, thats all its good for though, setting up stoich mix.
Surprised you needed it to be so close though Scott ?

Andy

Scott.T

iTrader: (2)

Andy,

Your may be right it may not be required to get that close.
My problem may of initially been a combination of a narrow band simulation only running 14.4-14.8 i.e the rich side of stoich and the Apexi open loop being about 13.9AFR.

We're always learning.........