New ACT Manifolds - Idle problem

[Rawhide]

Hi, I've recently replaced my manifolds with the ACT performance items. Whilst I was doing that I also had the ACT decat pipes put in. The car has only done about 200 miles and a trackday since having a full 12000 mile service and valve adjustment. After the service it was running smoothly but I felt it was still lacking in power. The TVR indy who had service the car mentioned the car was not making full power due to the manifolds. I've phone them since and they have said they didn't finish setting up the 'adaptives' after replacing and balancing the throttle pots.

The car has been stood for a week but I fired it up briefly and did about 5 miles on slippery roads (not ideal for testing)

The car is now idling at about 1100 - 1200 RPM and the revs don't seem to drop quite so quickly. It is popping and banging a lot more now which I assume is due to the decat pipes. It also smells of fuel a lot so wondering if it's running rich.

When on the power it's bit hesitant and is not smooth. It can rev out though but seems down on power.

I will try and poke around a bit tommorow if I get time but could do with some advice on where to look first. Could it be an air leak after my garage re-assmebled the intake after fitting the manifolds? My mechanic did a fantastic job fitting the manifolds but is not a TVR indy so something might have been dislodged or suchlike.

I'm also going to try to hook up to the ECU and see if I can get some info from there.

The car is a 2000 4.5

[Rawhide]

I've managed to get the ECU to connect to my laptop. I started up the car and let it idle.

I took a screen dump but am not sure how to host it. In summary everything looked as I'd expect.

Both throttles were showing 14.8 at idle

Adaptives are both 34.9

I think this is all ok...

When I checked the real time monitoring..

All the metrics look stable with the exception of the lambda sensors. Lambda 1 seems to jump eratically. Lambda 2 as far as I can see is just tracking along the bottom of the chart.

Would that suggest Lamdba 1 is faulty or Lambda 2 is not connected?

I'm pretty new to this stuff but I'm a quick learner. Any advice is welcome

[robsco]

Unfortunately I can't even begin to try to help with the mechanics, but to post up your screen dump for the technically minded chaps, you should be able to edit your post and there's a tab just above the box you type into. It will say Upload a photo. Just click that and you can add it into your post through there by following the instructions.

[ridds]

Lambda 2 sounds bust they sound cycle when functioning.

The adaptive value is very large but you need to look at the whole picture, different loads and speeds to see where it's bad.

Idle is too high and suspect the blades are too far open for the low pot value.

Sounds badly setup unfortunately.

[ridds]

You could also have an airleak at one of the exhaust joints that could be messing up your closed loop fuel control.

[mikesr]

Rawhide said:

Both throttles were showing 14.8 at idle

Adaptives are both 34.9

14.8 for the throttles seems low. IIRC the 4.5 should be around 18 at idle.

Definitely not OK if the adaptives are 34.9. The ECU is having to add a ton of fuel to get the mixture somewhere near. 34.9 is at the limit of adjustment that the ECU applies so the mixture may still be too weak.

Is the ECU showing AFR faults? These appear if the ECU does hit the limit on adaptation.

[Rawhide]

Edited by Rawhide on Saturday 10th December 23:40

[Rawhide]

ridds said:

You could also have an airleak at one of the exhaust joints that could be messing up your closed loop fuel control.

I made sure the garage used proper exhaust silicon as advised by ACT so I doubt this is the issue. Whilst not a TVR independent he is a trusted engineer and has done great work for us.

[Nickccc]

What do the adaptive tables look like?

[FarmyardPants]

I thought silicon killed lambda sensors. Lambda 2 doesn't look too healthy.

http://en.wikipedia.org/wiki/Oxygen_sensor#Sensor_...

Edited by FarmyardPants on Sunday 11th December 00:24

[Nickccc]

Check the tables, this sounds very much like mine did when the lambda's were back to front.
If one table shows all zero's then that's your problem.

[longbow]

Its been a long time since I set my pots up, but from memory the adaptives should be set to 15 when the throttles are both completely closed and the bar disconnected with the stop wound back in fully. Then, with that done, you add the bar and adjust the stop, tweaking both until both adaptives read 17 or 19 - can't remember which. This should be the idle setting. At the minute your adaptives are set to 14 on idle which is way too low and hence the correction (or adaptive value) is at its maximum showing that the ECU is adding lots of fuel against the ECU map to try to get the mixture right. When you set the pots, you should try to get the adaptives at idle as low as possible - i've had mine at around 5 or less on each bank before but it can be tricky if you have leaks or worn butterflies etc.

Edited by longbow on Sunday 11th December 07:49

[ukkid35]

I'm pretty certain you're only seeing half the picture with the water temp at 42, what happens when the engine is at normal operating temp?

[tvrgit]

ukkid35 said:

I'm pretty certain you're only seeing half the picture with the water temp at 42, what happens when the engine is at normal operating temp?

This. You'll only get correct readings when the engine is warmed up.

However, it does look as if one of your lambdas is dead, and the other is cycling erratically. They should cycle regularly (like the middle bit of your diagram shows Lambda 1 doing).

If the ECU isn't getting the right lambda signals, the adaptives will be all over the place.

However - I had a very similar graph - I had Lambda 1 cycling properly, and Lambda 2 looking completely dead, adaptives (in my case only on one cylinder bank) at the end of their range. It turned out to be a duff connector in the loom - the lambda connectors are on top of the bellhousing, under the starter motor. One of mine had water ingress and it had stopped the signal reaching the ecu - a bit of a clean, some waterproof contact grease and it was right as rain (which was what caused it n the first place...) - all lambda signals normal, minimal adaptives required.

Investigate the easy stuff first!

[Rawhide]

Thanks guys! Massive help. In between buying christmas tree and child care I'll check the connections and get it nice and warm to see how the numbers are looking

[ridds]

Rawhide said:

ridds said:

You could also have an airleak at one of the exhaust joints that could be messing up your closed loop fuel control.

I made sure the garage used proper exhaust silicon as advised by ACT so I doubt this is the issue. Whilst not a TVR independent he is a trusted engineer and has done great work for us.

Indeed but don't discount the possibility, many an hour and £ is lost chasing problems when you assume the simple things are fine.

As said above silicone does kill Lambda sensors so it needs to have fully cured before firing up

[Rawhide]

ridds said:

Indeed but don't discount the possibility, many an hour and £ is lost chasing problems when you assume the simple things are fine.

As said above silicone does kill Lambda sensors so it needs to have fully cured before firing up

Ok. Understood but it was Tim at ACT who recommended using silicon as the exhaust comes in sections.

If this was the problem could I expect any excess to burn off before replacing the failed sensors? It's not like it's easy to remove

[Rawhide]

ridds said:

Indeed but don't discount the possibility, many an hour and £ is lost chasing problems when you assume the simple things are fine.

As said above silicone does kill Lambda sensors so it needs to have fully cured before firing up

Ok. Understood but it was Tim at ACT who recommended using silicon as the exhaust comes in sections.

If this was the problem could I expect any excess to burn off before replacing the failed sensors? It's not like it's easy to remove

[ridds]

It is the right stuff e.g. Loctite 5900 or 5910 (high temp) but leave it 24 hours before starting.

Same with using on the engine casting joints, the sealant leaches silicone until fully cured and silicone causes huge aeration issues with oil.

[Byker28i]

That was the issue with that dodgy tesco fuel a while back. It had been contaminated with silicone so killed lamdas.

How did your ACT manifolds fit. Mine had to be 'adjusted' just slightly as they touched a chassis rail. My indy said they were a pain to fit and he had them on and off the car adjusting until he was happy.

However I didn't have to adjust my mapping after fitment and Charlie at Surrey RR said my car was set up fine. I do have the quiet ACT system which I know robs a little power, but I've still got 380bhp so prefer the 106db to another possible 5-10 horses.