MK4 R32 : 200 Cell Sports CATs; struggling with P0420 & P0430 codes

Fitted 200 cell sports CATs a few days ago and the engine has been throwing the following dreaded codes every 40 miles or so:

16804 - Catalyst System; Bank 1: Efficiency Below Threshold
P0420 - 001 - Upper Limit Exceeded
16814 - Catalyst System; Bank 2: Efficiency Below Threshold
P0430 - 001 - Upper Limit Exceeded

The car has apparently been remapped at AMD sometime in the distant past 2004/2005 ish so prob is running ancient coding and missed any codebase updates e.g. 6463 update?

The car will also throw the codes with decats with rear O2 spacers but will run absolutely fine with no codes with an OEM CAT fitted.

Ive checked the following in VCDS:

Block 034 - Aging of Pre-Cat Oxygen Sensor:
1.866 B1-S1 OK 1.819 B2-S1 OK
These vary a little anywhere between high 1.7 to high 1.9s so Im assuming the variance is okay and both front wideband O2s seem pretty new?

Post CAT O2s are also showing as ok in block 36 as both show as B1-S2 OK AND B2-S2 OK but is there any way to correlate the post CAT o2 reading with the wideband pre CAT O2 sensors to see if anything spurious is going on?

Block 20 & 21 are showing ok with a little retard now and again but nothing major.

Additive fuel trims for both banks in block 32 are very low: both around 0.8% and multiplicative for both banks is a little lean at both around 2.3% to 2.5%.

Not sure what else to log and check for?

MAF readings seem a little spurious like signal drop out of a good few % esp so when post 4200 inlet port switch and wonder if the MAF is on its way out or not:



Ive cleaned it using isopropanol which made the car run a little smoother but its still chucking the above codes

Kinda struggling now and wonder if Is either a vacuum leak in the inlet manifold itself, possibly dodgy plug/s , spurious MAF readings or actually (unlikely) a dodgy pair of CATs causing the ECU to spit the dummy?

What to check for now?



Any R32/VCDS gurus out there?

 

P0420 on any car usually means the cat(s) are inefficient. Can you log the lambda voltages in graph form? If the post cat voltage trace mirrors the pre-cat it shows the cat(s) is/are inefficient. Post cat voltage should stay a constant voltage in closed loop.

As far as I remember from my ex GF having her R32 remapped by AmD back in 2004ish their map fixed the 6463 update.
Her car was a year old when she bought it and the 6463 update was already in use (IIRC)

She had Milltek sports cats (not sure what cell) on her's and never had any codes/CEL.

If you have a vac. leak the fuel trims would be high (adding fuel) to try and counteract the leak (remove the FPR pipe and watch the fuel trim rise!) Bedtime reading on Fuel trim. http://wiki.ross-tech.com/wiki/index.php/Fuel_Trim_Info

If it was plugs/coils it would show a misfire code.

MAF readings, yes look a bit crazy, but still hitting acceptable peak reading (200+ g/s) so I'd say OK. Have you run it with it disconnected to see if it performs better?

Certainly points towards the sports cats themselves if it's all good with OEM cats.

 

He's already done that dude.

My experience of P0420 is usually on Ford Fiesta/Focus that have suffered coil pack failure, and been run around pouring fuel in the cat, soon after the coil pack has been repaired it will show the P0420 with CEL/MIL on. As I mention above I test them by monitoring the pre and post cat voltage traces on Ford IDS, if the voltages match the cat is fecked. New cat = fixed.

 

So let me get this right Gary as I'm not overly sure of the data values in the measuring blocks... Looking from the above info that Mushy posted (Cheers bud) - Is it blocks 33 and 36 I need to log and correlate - In short I'm looking for realtime voltages from O2s pre and post CAT on both banks. The post CAT voltages should be relatively uniform ( 0.5V? ) and not tracking/following the pre CAT voltages?

 

Oh Doh... Now I'm doing it too... Must be this Northern air Block 31 appears to show all 4 sensor voltages for a 2 bank system?

 

Correct. It's easier if you can view it in graphical form rather than numerical. On our stuff it shows the realtime voltages in graph form.

 

Was Stage 1 with Milltek for a start,then sports cats and K&N intake thing, then had Schrick 264 cams and Revo software which needed tweaking at HQ to cure a cold running stutter. This was all before the days of coding out rear sensors etc as far as I know. Used to go very well in the end and sound great! Remember doing 160mph without too much issue coming back from the 'ring one time.

 

There are 2 branches on VR engines with EOBD capability.
With EOBD (and with OBD2) you monitor the condition of the cat looking looking at rear CMS behaviour B1/B2_S2 in relation to B1/B2_S1.
When the catalyst is aged, overheated or poisoned ( oil contamination), the conversion efficiency and oxygen storage diminishes and the rear sensor mimics the front sensor volatage between 0.1 to 0.9v. Once the rear sensor voltage drops to min voltage the OBD threshold will be breached and you will have the mapped faults as displayed.
Having no catalyst is what is causing this concern and it is a case of switching off the rear sensor feedback and heater control via an updated calibration flash.

 

Okay done some logging...

Looks like block 31 on the BFH engine only shows Lambda Factor:

031 Linear oxygen sensors :
Lambda actual value Bank 1 / Lambda specified value Bank 1 / Lambda actual value Bank 2 / Lambda specified value Bank 2

Specified values on cruise always show 1 except on the overrun where the Specified values always appears to increase to 1.47 and actual value goes way rich typically up to 1.9 or so.:
1.078 1.047 1.055 1.047

Guess this explains the crackling and popping

On closed loop typical values:
0.93 1 0.93 1 79.72
0.969 1 0.969 1 80.23
0.938 1 0.922 1 80.74
1.031 1 1.031 1 81.25
0.906 1 0.977 1 81.76
0.984 1 0.969 1 82.27
1.031 1 1.039 1 82.78
0.992 1 0.992 1 83.28
0.961 1 0.938 1 83.79
1 1 1.008 1 84.29
0.938 1 0.938 1 84.8


On transient (open loop throttle):

0.797 0.899 0.797 0.899 90.88
0.891 0.899 0.906 0.899 91.4
0.906 0.899 0.899 0.899 91.91
0.891 0.899 0.891 0.899 92.39


As soon as it goes back to part throttle the lambda goes back to closed loop as above...

No cigar here I guess?


So logged gtoups 33 and 36:

'033 Group C: '036
Lambda Voltage Speed Voltage Voltage Voltage
TIME
% V km/h V STAMP V V
-1.6 1.54 0 1.52 0.45 0.445 Test OFF 0.445 Test OFF
-1.6 1.54 0 1.52 0.94 0.445 Test OFF 0.45 Test OFF
-1.6 1.54 0 1.54 1.43 0.44 Test OFF 0.445 Test OFF
-1.6 1.52 0 1.54 1.91 0.445 Test OFF 0.445 Test OFF
-1.6 1.52 0 1.52 2.39 0.445 Test OFF 0.445 Test OFF
-1.6 1.54 0 1.54 2.87 0.445 Test OFF 0.445 Test OFF
-1.6 1.54 0 1.54 3.35 0.445 Test OFF 0.445 Test OFF
-1.6 1.54 0 1.54 3.87 0.445 Test OFF 0.445 Test OFF
-1.6 1.54 0 1.54 4.4 0.44 Test OFF 0.44 Test OFF
-1.6 1.54 0 1.54 4.89 0.45 Test OFF 0.445 Test OFF
-2.3 1.52 0 1.54 5.38 0.445 Test OFF 0.445 Test OFF
-2.3 1.52 0 1.54 5.89 0.445 Test OFF 0.445 Test OFF
-2.3 1.52 0 1.54 6.4 0.45 Test OFF 0.445 Test OFF


I assume TEST OFF means I need to let the ECU finish setting itself up?



Here's some group 37 & 38:
(not yet sure what I'm looking for here TBH)

Group A: '037 Group B: '038
Load Voltage Difference Load Voltage Difference
TIME TIME
STAMP % V STAMP % V

48.33 21.1 0.69 -0.004 System OK 48.5 21.1 0.325 -0.004 System OK
48.67 21.1 0.075 -0.004 System OK 48.84 21.1 0.095 -0.004 System OK
49.01 20.3 0.295 -0.004 System OK 49.18 20.3 0.735 -0.004 System OK
49.35 20.3 0.755 -0.004 System OK 49.52 20.3 0.755 -0.004 System OK
49.69 21.1 0.71 -0.004 System OK 49.86 21.1 0.38 -0.004 System OK
50.03 21.1 0.085 -0.004 System OK 50.2 21.1 0.155 -0.004 System OK

 

I hear you Eddie... Trouble is the bucket R32 is a daily so unless I can get a lift to work for a week or so the damn ECU will be staying in the car... Just moved location but couldn't get anywhere even near work and now I'm in a worse position than I was before. lol Gonna need to try and sort something out soon though somehow.

So just to confirm Ed - It couldn't be caused by a vac leak or poor combustion given that the car will not throw a code at all with the normal OEM CATs fitted and in short the 200 cell CATs are sheight?

 

Must have went like stink - I have a mate in the central belt with pretty much the same setup and it goes pretty well for a fat bus.

 

Echoes my post with added Toyotecwerke..lol

 

I think you may well be right Mushy.

 

A vac leak will be picked up by the feedgas sensors as a lean voltage and will make the lambda controller compensate by multiplying a fuel factor to get the engine to lambda 1 for part load conditions and during idle conditions add extra fuel to correct.
In such a case you would have a different fault something along the lines of Lambda controller fr limit exceeded (greater than +25%)

The space velocity effect from the 200 cell cat may mean less time for the feed gas to interact with the cat washcoat and your rear CMS for B2 and 1 are mimicking the front sensors because of low conversion. Spacers on the rear HEGOs are a waste of time and bodge for a proper calibration where you can switch off this function.
You can purchase another ECU though...