Mapping Issue or 3rd party ECU?

[Monsterlime]

I shall give this a proper read today, thanks. The bible mentions pulling the cover off the ECU plug and testing from there as well, but I did look at that yesterday and couldn't see how I could remove it.

Depends on what you mean by cycling. The odd lambda (top) does change on the short term trim, but never actually goes into the negative, it just stays in the positive. It definitely changes when you rev the car, I could see it going up and down etc.

While I am not intending on driving it at the moment, from the looks of it, if needed to move it, put it back to Fuel Map 1 would seem like the right way to go?

[blitzracing]

When its working correctly the short term trim should cycle around the mid point both + and -. at around .5 to 1.5 second intervals. If it cycles on one side only, it means the long term trim is wrong is some way, as the ECU moves the long trim to get the short term trim point centralised. If it does not cycle at all, then you have a probe problem, or a sensor value is out. In your case Id suspect you have two issues- one a faulty probe, and secondly a wrong sensor value. My hunch would be that the AFM is the most likely cause- the outputs can go high, and this makes the engine run rich which matches your basic issues. Get a screen shot of RG at idle, with the AFM reading set to direct to get the % values.

1.6 volts for a 3.5 litre engine = 32%
1.62 volts for a 3.9 litre engine 32.4 %
1.75 volts for a 5 litre engine =35%


If Austec tried to remap with a faulty AFM you really dont stand a chance of getting it correct whatever you do with the map. The fact the standard TVR map 5 is not working shows you have a hardware issue(s) that needs resolving before you worry about the map you run.

[Monsterlime]

Right, ok, thanks. Will do so, but will be tomorrow now due to weather and family plans. Should I reset the ECU first or just let it idle as is?

The AFM is a lot simpler to get at/replace than the Lambda's, so much closer to my skill and comfort level! lol

What I don't really understand, is why none of this had been flagged to me already. I was told they had problems getting it to run right, and as you say, this is clearly why but why didn't they realise there were hardware issues or tell me I would need to get it seen to?

[QBee]

I've got a spare AFM (5AM, ie standard) if you want to borrow it. I got it in a job lot of spares, so am not 100% that it is working properly, but can post it to you if it might help

[ClassiChimi]

What a man,,, top bloke our Qbee,,, fancy donating £10,000 to the ALS, track day fund,,, hahahah.
Good to see your still a gent Anthony.
Alun

[blitzracing]

I cant pass comment on Austecs diagnostics, but the stock system has a lot of tolerance in it before it goes wrong, so that should have rung a few warning bells. To take the reading You dont need an ECU reset, but the idle needs to be near 800 rpm- as this is the airflow values at this engine speed. If you idle at 1000 or higher- the results may be outside the values quoted. My 3.9 runs at 1.69 volts at 800 rpm.

Edited by blitzracing on Saturday 28th February 17:22

[TV8]

Have a look at this post. http://www.pistonheads.com/gassing/topic.asp?h=0&a...

See the second post in particular. They wanted to remap the car to sort the poor running and it actually had a dodgy AFM.

[QBee]

ClassiChimi said:

What a man,,, top bloke our Qbee,,, fancy donating £10,000 to the ALS, track day fund,,, hahahah.
Good to see your still a gent Anthony.
Alun

Thanks for the compliment, Alun, but may i ask what you are smoking/drinking?
Not another consignment from our Dutch TVR friends?

[ClassiChimi]

Hahahah steady on old bean,,,
I've been getting fitter,, well until yesterday
Second expensive bike stolen from my house in less than a year, I'm glad it's going on the market.
Got the wine out last night and me and an old mate put the world to rights,,,
Effing thieving buggers,
Hey it's just good to see your still so generous to other Tvr owners,
Alun

[Monsterlime]

QBee - That is supremely generous, and really rather appreciated! I shall confirm today, once I have completed the tests!

blitzracing - Should also be said I am deeply appreciative of all your assistance as well. Needless to say, beer etc is definitely owed to both of you!

I have a 450, so I assume my voltages should be around 1.65-1.70? My idle is around 900ish, and does drop into the 800's, but should I adjust it down?

[blitzracing]

The AFM tend to be a long way out when they do fail, so its will be more than a percent or two, but as a guestimate I think you are there voltage wise. The percentage RG shows is the voltage against 5 volts being 100%. Dont worry about trying to wind the tick over down at this stage.

[SILICONEKID345HP]

How the hell does mine work without an air flow meter ? The temp sensor is not going to do a lot .

[Monsterlime]

Ok, so, I am really rather confused. To me, it looks like the AFM is fine?



Also, after reading the thread that TV8 linked, I tried the EasyStart leak test, and it didn't seem to have any effect. Although it is very windy, so I may have to try that one again on a calmer day.

However, the reason I am really confused is that this time, both Lambda's were working -

Some revving -



After a bit more revving -



Basically, both times I revved on the drive, I could see both Lambda's cycling. The long term trim didn't change though. I did also clear the fault code from within RG.

And this is after I took it for a quick spin to see if it carried on "working" -



I also logged the entire drive and have uploaded it here - http://pastebin.com/phMtEGtn Although I think it missed a couple of minutes because the laptop went to sleep (I forgot to amend the power settings before I set off).

I did look to see if I could get some paper clips into the lambda's to measure their voltage, but it all seemed sealed with glue, so I didn't think I'd be able to get them in. However, when I did look quite closely at them, on both sets, the wires right near the connection point into the sensor are quite frayed/damaged.

The car drove and idled fine while driving and stopped. Plus, when I restarted it shortly after the drive, the rev's did not do the normal of go high, drop to like it might stall and settle, they just stayed level.

Looking at the raw numbers in the log, it still looks to me like the even lambda definitely has a problem as it is logging 255 a lot, but is definitely doing something at other times which it hadn't in my other logging sessions.

So, in conclusion - I am deeply confused.

QBee - Not sure you should send me the AFM at the moment, as it doesn't look to be a problem, but I do stand to be corrected. Thanks

[blitzracing]

Id agree the AFM is fine. Ive seen it before that the short term lamdas wont move until you give the car a rev' then every thing is works- so dont worry about that. You need to swap the probes over and see if the 255 moves to the other bank- then you would know you have a faulty probe for sure.

Edited by blitzracing on Sunday 1st March 16:29

[Monsterlime]

Ok, will give it a try.

Where do they plug into the loom? I've tried to follow the wires, but they disappear into dark places. Plus, they are currently cable tied to the chassis so I don't want to start cutting stuff until I actually have the probes out.

[blitzracing]

The TVR has extension cables over the original Range Rover loom so there are two lots of connectors- - you can follow the lead back from the probe, and you will find a circular connector strapped to the chassis down near the top front wishbones. You will find two little areas on the side of the connectors you press together to allow the connector to pull apart. There is also connectors down by the fuel rail on each side of the V8, but unplugging at the chassis front will be easier.

[Monsterlime]

So, that ploy failed, unfortunately. I really just don't have the right tools. I tried to use molegrips, but they are quite thick ones, so all I ended up doing was squeezing the casing at the top of the sensor (likely to kill it completely?), rather than turning it as they wouldn't reach the bolt.

Is it worth buying those sockets from Ebay? These ones - http://www.ebay.co.uk/itm/LASER-4984-Difficult-Acc... Not totally sure they have the reach either, but I can certainly try.

The sockets I do have are too short to clear the end of the probe, and they were my fathers and I don't really want to modify his tools if I'm honest (plus I don't have the tools to do so anyway).

Would I be completely wrong in suspecting that this could be a manifold leak? Enough of one that at idle the probe sees nothing, but when moving/revving enough gas is hitting it?

[QBee]

If all else fails i suppose, at an absolute last resort, before you kill the sensor completely, you could try taking it to a TVR independent, who will have the right tool and the knowledge.....

[Monsterlime]

Hehe, yes, that is pretty much my next course of action. Was just trying to do as much myself as I can, but I'm not sure its worth spending £30 on something that may or may not do the job.

I suspect a call to Taylor TVR will be made tomorrow!

But, again, I must thank all that have assisted in getting me this far, I would never have gotten here without the very kind assistance I have had and am very much appreciative of it.

[Monsterlime]

Well, she is booked in. However, they are very busy and they can't look at her until the 15th April!