MPG has dropped drastically

[jazzdude]

Since getting the car back on the road following its winter's hibernation, and just pootling around at lowish revs until I am confident that nothing has gone amiss while it was sitting in the garage.

The thing is I swear the fuel gauge is moving with the rev counter as I have been through nearly a tank and have only covered less than 150 miles. Used to get over 200 miles with spirited driving, not driving like my old mum as I have been the last few days. I have got out a couple of times to check for leaks but can't see anything.

The car is fully decatted, on green tune with a new EPROM with one of Steve Sprint's tune files (TVR 400 Precat and 400 CAT based on LR R3652) so not sure what Rovergauge will show if I used it. Since the car was put away I changed the ECU temp guage (although the previous one worked fine, just upgraded it) and checking that on Rovergauge it works, although it seems the fans come on at an indicated 98-100 deg, when I am sure it should say 92 deg shouldn't it?

So I first completely cleaned the stepper motor, as it started doing that racing thing before coming to a stop but the idling has improved, but still not steady, occasionally dropping to nearly turning the car off sometimes. I have disconnected the stepper motor while the idle was low, and the car idles more smoothly but goes erratic again after a while.

Other changes are: I retrofitted a/c, but independently of the ECU, ie new relays and switches work off the trinary switch and an internal switch on the dash, the return fuel feed goes through a fuel cooler now. I also fitted electric power steering and put the battery in the boot, the alternator is pushing out 14.4V, and I measure 14.2V at the battery. All the earths were renewed in the engine bay to nice fat wires which seems to have added an extra volt on my voltmeter that used to read about 13.5V (it reads over 14.5V now). A remote clutch servo was also fitted and that takes a vacuum from the plenum. Also changed the dash so wires from instruments removed and pushed back in.

The vacuum pipe from the dizzy is on, although I can't see if it is blocked nor know how to check.

The car feels like it is running rich and I sometimes get a whiff of petrol when I start it. It drove shunt free last year after the chip change, yet now it feels shunty again when decelerating at lower speeds.

Lots of factors I know, but I really don't know what to check first.

Edited by jazzdude on Wednesday 10th May 12:59

[rev-erend]

I had that when my AFM connector was falling off.. pushed it home and cable tied sorted it.

[jazzdude]

When you say falling off, the connector was not sitting in the AFM?

Mine is definitely in place, although I can check to see if it is seated properly. I suspect the ECU temp sender, I will also check that by putting the old one back in, as I write I am wondering if the connector on that is seated well.

[blaze_away]

Do an mpg check using your phone and an. App. I had this on my previous chim. Turned out the speedo wasnt recording any mileage when travellint under 25 mph.

[jazzdude]

Haha, that would be neat.

My car's fuel gauge needle seems to move over to the left by 1mm every 10 minutes of driving though. I will do an app check but it definitely seems to be moving very fast, and there is this lumpy idling to think about.

[eric450]

Sounds like a daft suggestion but I had a similar issue with mine.
I'd just changed the HT leads for the ceramic ended ones to get rid of the extenders.
Car had always run well with very little shunt etc until I put these on.
Then at odd times it started to run very rich and lumpy for a few minutes then correct itself.
I'd taken great care to route the new leads (or so I thought)
The issue turned out to be the king lead running too close to the ecu temp sender.
When it happened, rovergauge would show the temp dropping to sod all so more fuel chucked in.
Re routed the king lead away from the sensor and not had a problem since in several hundred miles.
My own fault and probably not you issue, but worth a look.

Cheers ....

[Classic Chim 13.111]

You say the Ecu temp is at 98/100 when fans come on!
I'd stick to looking for faults on the CUX first, Otto switch etc etc, start by putting your original sensor back in.
Sounds like you've done some nice work on the car, little choice but check and check again your earths etc and routing of wires or anything you've been near. Lambda connectors etc


ETA
You've got good voltage, you've done a good job

[jazzdude]

My first suspicion is the ECU temp sender, to be honest.

The replacement TVR parts did not fit, and then they sent me another one, but the weird thing is that I recall that the connector did not seem to 'click' when pushed down, like the old one did. I am going to try putting the old one back on. But also these are the areas I also want to check:

1. Check Rovergauge for limp home mode and any fault codes
2. The base idle on the plenum
3. the voltage at the AFM
4. that the vacuum pipe from the dizzy is not blocked or leaking
5. that all the HT leads are on nice and snug and that those in 7 & 8 are not perished due to heat (ceramic socks on them, no extenders).
6. the throttle body voltage

When the auto electricians wired up the power steering and new a/c, they decided to tidy up the engine bay and put everything in conduit, which is appreciated and looks good. Not sure if, as Eric 450 alludes to, that they might have strapped the ECU sender cabling to something or any other cabling for that matter to an HT lead. I doubt it but worth a check.

The otto switch was new last year and I am sure it was an 92/88, not higher. When I calibrated the Rover temp sender, Rovergauge was spot on with the temps that the fans came on, so its strange that it shows a much higher temp. Also firing a laser thermometer at the brass ring of the otto switch seems to confirm that the fans are coming on and off at the 92/88 temp.

Strangely enough, the stepper motor seems to be getting the idling down to an acceptable (950 rpm) point after the car has warmed up, but it still hunts and slows, it cant seem to keep a steady idling speed. Taking the stepper off calms things a bit, but then that is not dealing with the cause, just the ECU's reactions to the info it is getting, so I dont think that is a solution.

[blitzracing]

Problem with green tune is you loose a lot of error codes if something is wrong as it does not use the lambda probes for feedback, so few fault codes can get logged. All is not lost however- the first is how did you set the CO trim on the side of the AFM? I know Steves chips have a specific voltage to get the correct AFR, but that's as he set them up on his car. Ideally you want an MOT gas tester to set the correct mixture, but you can do it by ear- wind the pot one way and it will start to sound like it on choke- so that's richer, and then the other way until the engine becomes lumpy- this is too lean, so then wind it back for a smooth idle by no more. Give it settle time as well. One of the basics however is check the plug colours- if they are black and sooty, thats where your MPG is going as the mixture is too rich, and there is a big adjustment available with the CO trim screw. This alters the mixture up to around 2400 rpm, which is just where the car spends a lot of its time. Always worth double checking the AFM voltage is in spec at idle, as a high output will make the mixture rich as well.

[jazzdude]

blitzracing said:

Problem with green tune is you loose a lot of error codes if something is wrong as it does not use the lambda probes for feedback, so few fault codes can get logged. All is not lost however- the first is how did you set the CO trim on the side of the AFM? I know Steves chips have a specific voltage to get the correct AFR, but that's as he set them up on his car. Ideally you want an MOT gas tester to set the correct mixture, but you can do it by ear- wind the pot one way and it will start to sound like it on choke- so that's richer, and then the other way until the engine becomes lumpy- this is too lean, so then wind it back for a smooth idle by no more. Give it settle time as well. One of the basics however is check the plug colours- if they are black and sooty, thats where your MPG is going as the mixture is too rich, and there is a big adjustment available with the CO trim screw. This alters the mixture up to around 2400 rpm, which is just where the car spends a lot of its time. Always worth double checking the AFM voltage is in spec at idle, as a high output will make the mixture rich as well.

I know re the green tune and the limitations on the error codes.

When I last checked the AFM with a multimeter, I got around 1.4V out of it, but I couldn't set the CO trim on the AFM because the allen keyed bolt that slots into the little brass screw in the AFM, it pops out, so turning it does not change the voltage. I'm thinking of putting a tiny dab of superglue onto it, and the pushing it in, and hopefully the little screw will turn with the allen keyed bolt.

If this indeed works, then I will check the voltage again, but this time using Rovergauge, and am aiming for 1.42V or 13.5 AFR which is what Steve recommends in his chip.

[blitzracing]

If the co trim is broken- its just a 1k 10 or 20 turn trim pot to ground, so you can cut the trim wire where it joins the AFM and pop the pot into the loom instead.

[jazzdude]

You might have completely lost me there Blitz.

As I understand it you mean get a 1000 ohm variable resistor, wires both sides, attach one end to earth and the other, where?

[Steve_D]

Mark may be along to clarify but I think he will be saying you need to cut the Blue/red CO trim wire either at the MAF or in the footwell (which would be my choice).
You then need an ignition switched +12v ( the Brown/Orange coming out of the main EFI relay would be a good choice) going into your new potentiometer with the output going to the Blue/Red wire and into the ECU. You will then need to adjust the pot so the ECU sees around 1-1.5v then get it on a gas monitor at a MOT test station.

Steve

[jazzdude]

I have found the mod, is it the one described at the bottom of this page?

http://www.v8engines.com/carbs-2.htm#setting

Some references say for a decattted green tune car should read 1-1.5v here (as opposed to the white tune 1.8v), but as I said previously it is recommended for this map to get 1.42v. How crucial is that extra voltage?

[blitzracing]

Steve_D said:

Mark may be along to clarify but I think he will be saying you need to cut the Blue/red CO trim wire either at the MAF or in the footwell (which would be my choice).
You then need an ignition switched +12v ( the Brown/Orange coming out of the main EFI relay would be a good choice) going into your new potentiometer with the output going to the Blue/Red wire and into the ECU. You will then need to adjust the pot so the ECU sees around 1-1.5v then get it on a gas monitor at a MOT test station.

Steve

Not quite-

The CO wire open circuit has 5 volts on it- so you put one end of the variable resistor to ground, and the wiper goes onto the CO trim wire and the other end of the resistor. As you alter the resistance value, it effectively pulls the 5 volts down towards ground by however many volts you need- You can do this as there is a resistor in the 5 volt feed inside the ECU, so it forms a potential divider with the CO trim resistor, so you don't need a secondary voltage supply from anywhere. The RPI wiring diagram is fine.

http://www.v8engines.com/Gallery-2002/4.6-airflow-...

In terms of "how critical"- you are effectively setting the long term fuel trim with this- just like adjusting the idle mixture on a carb- and it will go from so lean the engine wont run towards 0 volts , to very rich at around 3.5 volts. Its simply a case of getting the voltage right to what suits your car, the base voltages are guidance only that they should be in that range if all else is well on a stock engine.

Edited by blitzracing on Friday 12th May 11:35

[Steve_D]

jazzdude said:

I have found the mod, is it the one described at the bottom of this page?

http://www.v8engines.com/carbs-2.htm#setting

Some references say for a decattted green tune car should read 1-1.5v here (as opposed to the white tune 1.8v), but as I said previously it is recommended for this map to get 1.42v. How crucial is that extra voltage?

Yes that is the mod and shows that I had it the wrong way round.......this has me very confused so I hope Mark can help me out.
If you tested the MAF normally you would probe the blue/red and earth to see what voltage the MAF is sending to the ECU.
To my mind this means the other side of the pot in the MAF has +12v which is the opposite to the mod you liked to.

If your ECU is mapped for 1.42v then set it that way but I would then take it and get it check on a gas tester.

Steve

[jazzdude]

blitzracing said:

Not quite-

The CO wire open circuit has 5 volts on it- so you put one end of the variable resistor to ground, and the wiper goes onto the CO trim wire and the other end of the resistor. As you alter the resistance value, it effectively pulls the 5 volts down towards ground by however many volts you need- You can do this as there is a resistor in the 5 volt feed inside the ECU, so it forms a potential divider with the CO trim resistor, so you don't need a secondary voltage supply from anywhere. The RPI wiring diagram is fine.

http://www.v8engines.com/Gallery-2002/4.6-airflow-...

Edited by blitzracing on Friday 12th May 11:27

Edited by blitzracing on Friday 12th May 11:29

Thanks Mark, like I said I came across that RPI thing and it makes sense now.

The only thing that doesn't though is that it says pull apart the socket, which is fine, add the resistor to the wiring, which is also fine, then it says put it all back together. How can it go back together if there is a resistor in there as well? Does it fit and if so how can you make any adjustments without pulling it apart again?

I think I might try with the blob of superglue first to see if I can get the pot in the AFM turning with the allen key, and if not then I will take apart the plug.

[blitzracing]

There is no need to mess around at the ECU plug end- you can do this at the AFM end as long as you waterproof it with tape after you have soldered it into the loom and set the voltage. You MUST cut the CO trim wire from the back of the AFM socket if you fit a new variable resistor , or you end up with two trim resistors, and that will make setting the voltage impossible. The RPI pages are about fitting a different AFM where no trim resistor is fitted in the first place.

[Steve_D]

blitzracing said:

There is no need to mess around at the ECU plug end- you can do this at the AFM end as long as you waterproof it with tape after you have soldered it into the loom and set the voltage. You MUST cut the CO trim wire from the back of the AFM socket if you fit a new variable resistor , or you end up with two trim resistors, and that will make setting the voltage impossible. The RPI pages are about fitting a different AFM where no trim resistor is fitted in the first place.

Hi Mark
So you cut the wire at the MAF (or somewhere along the wire) and fit your resistor/resistance to that wire running to the ECU but where does the other side of the resistance connect? is it to signal earth or +12v?

Steve

[blitzracing]

Steve_D said:

Yes that is the mod and shows that I had it the wrong way round.......this has me very confused so I hope Mark can help me out.
If you tested the MAF normally you would probe the blue/red and earth to see what voltage the MAF is sending to the ECU.
To my mind this means the other side of the pot in the MAF has +12v which is the opposite to the mod you liked to.

If your ECU is mapped for 1.42v then set it that way but I would then take it and get it check on a gas tester.

Steve

You are thinking the wrong way around- the AFM CO connection is completely passive- just a resistor. The ECU supply a 5 volts on the trim wire that you pull towards ground as you lower the resistance. So the other end of the resistor goes to ground.

Edited by blitzracing on Friday 12th May 11:51