Despite my early reports that my car was fine with 16vt injectors, the opposite has turned out to be the case.

The car is really struggling with them

The problem appear to be the ECU "learning" function - or more accurately, the adaptation of the ECU to information it is getting from the various sensors. The problem is that with the Unichip, what the ECU sees and what is actually happening are two different things.

Basically, mapping the Unichip onto a freshly-reset ECU gives good figures and the car feels absolutely great.

But within a few miles, the ECU decides that there's too much fuel (even though there isn't), and starts to back it off. At 20%, the ECU gives the injector light with a "learning function" error (Code 06, I think). At 25%, the limit is reached, and the ECU throws a 23 code - "possible ECU error"

By this time, the car is seriously underfuelling everywhere. Its very difficult to pull away, it hesitates and hunts on a light throttle and it doesn't pull well on a full throttle. At tickover, its a lottery as to whether it will run or stall.

Simply resetting the ECU makes it all good again, but 50 miles later, its just as bad.

I've PM'd Graham L to see if the learning function can be disabled, but I'd really appreciate any feedback from anyone who knows what they're on about, or anyone thats fitted 16VT (or bigger) injectors.

I'm hoping for feedback from Barbz, Mavric, JohnS and Nyssa (Trev - didn't you have a management system for sale )

Sorry to hearthis Nig. , I was warned against going to these, as they have been known to cause these sorts of problems,mines are still sitting in the `to do one day` box,did Barbz(sorry `the advisor` ) not just fit and run with these ?

the problems u describe sound similar to the problems 16vt Evo chip owners experience on cold starts. The injector light comes on for about 30 secs and the revs are all over the place, i've also noticed that during this time, whilst my headlights and dash instuments remain lit, my boost gauge fades in and out dependant on the revs. (eg lower revs, near stalling = very low light on boost gauge).

The fact that this always stops after the injector light goes out makes me think its a 'learning function' problem too. Maybe GrahamL can pass his findings onto Paulinho so he can disable it on the EVO chips too.

Hmmmm, this is proving troublesome Nigel Are the problems you are having only during closed loop, I assume open loop the modified fuel maps are working fine?

Barbz didn't appear to have any idle problems when he ran the 16vt injectors for a fair few months. This was on the standard ecu and normal fuel regulator. You could try an adjustable fuel regulator which may be able to force more fuel and tweak it accordingly for idle and then adjust the fuel maps once that has settled.

I can't remember how Jari got round this as he was/is running 440cc injectors irc. I guess you could try to get the ecu reprogrammed by someone who can read it all and modify all parameters e.g. Squadra Tuning, although JohnS is working on something at the moment but not sure how close he is.

hi Nigel,

I thought that you had sorted out your troubles with that
split re-circ hose ?

.

Nigel,

The main problem is the air flow meter. With this inline the ecu will be constantly varying the map and 'fine tuning' it to create its own perfect fuelling. with ur turbo etc this is a problem as u need alot more fuel and the ecu is backing it off to compensate bak to its original map. If you disconnect the air flow meter u will notice ur cutting out problems low down will disapear as the ecu will set a base map, however u will probably run lean after say 4k rpm, but try it for the low down off boost and u will c.

1 wasy to get around this problem is to install a map sensor and get rid of the afm entirely. That is why i havent had my unichip mapped on my car and i have disconnected it for now, because of the problems u are having. Hopefully I will be mapping my car with the bigger injectors and a map sensor in the next few weeks so ill keep u posted.

Mark

Hi Flea, Nigel

This is the same ECU learning problem Juergen from German reported, was it last year. Now there is at least two more who have similar kind of problems, Nigel and also one other guy in German.
What I have found commonality to these cases is, that all of you are using same type of injectors if you think the design of them. Both Juergen and the other German is running 440cc Bosch injectors. I think the 16vt 384cc injectors are same design.

I did also have problems with my 465cc injectors, but now I have learned (one month ago) that my problem was simply too low boost level. Last autumn, when having 357 bhp, the car was overfuelling very badly, A/F near 10. I ran max. 1.4 bar boost and the behaviour was terrible. Now I have the very same setup, but I have adjusted the boost level up to 1.8 bar. So running between 1.6 - 1.8 bar, the car is functioning very well, but after 3500 rpm. Exactly in 3000 rpm I do have terrible feeling that the car does not want to go anywere. My guess is that the fuelling under 3500 rpm does not match with the boost I can get (+1 bar).
I guess Flea did have same kind of experience last year (?).

Tick over is normal, like with OE injectors. Fuelling I haven't check, but comparing last year overfuelling, I guess now it is quite good.

In some point of this month I will go to RR my car. There the fuelling can be measured properly with wide band lambda.

My injectors are EV6 desing from Bosch, they are much sensitive compared to the "older" design. The shortest open time can be set much lower than in OE type of injectors.
Bosch number: 0 280 155 968 (Volvo S60/V70 R).


Regards and I hope you can solve the problem Nigel,

- Jari -

Jari, those injectors are rated at 420cc @ 3bar and 465cc @ 3.8bar. Do you have a 3.8 bar regulator? It's very interesting that you have no idle problems, I know you have a custom chip but do you know what exactly has been done with it?

Markus, are you going to use a map sensor with the Unichip? Hope it works well

Has anyone spoken to 'Fergie' as he runs 16vt on his motor ??

Jamie

@flea no unichip, but yes to the map sensor...im gonna chuck the unichip in the skip

I think Fergie is still stuffed for internet access.

I've got his mobile no. PM me if you want it Nigel and you can see how he's getting on with them.

Flea,
I have adjustable fuel pressure regulator, at the moment I'm running with 3.2 bar fuel pressure. Now my eyes opens once again , this might be the reason for the problem under 3500 rpm, last year between 3000 - 4000 rpm the car felt much stronger than with OE injectors and now much weaker, but last year I was running with higher fuel pressure(3.8 bar).
Because of last year over fuelling I dropped the pressure, but this might cause now the "weakness" problem. I have to adjust the pressure higher to see the difference and see if it tries to choke in high rev's like it did last year.

Yes, I don't have any idling problem what so ever. I have now ran over 1000 miles in one month without noticing any difference, just that weakness in low rev's. Also starting the engine cold or warm has no effect at all.
About idling, last year there was no problem either, although the CO -value (idling) was app. 8% and it should be 0.5%.
I don't know if it makes any difference, but I'm using the Audi RS2 spark plugs, which are platin and have 0.6 mm gap.

- Jari -

Nigel I replied this on your other thread, but didnt get any reply

so I'll post here, especially as Jari is on board here,.. Jari this was something you were talking to me some time ago about the adaptive function of the standard ECU,.. I'm sure the lambda eventually changes values throughout the whole fuelling maps, like what is happening at idle affects elsewhere.

Quote:

---

Whilst getting bored mowing the lawn this afternoon , I had an idea

What else could possibly be the problem ?,.. what assumptions have we made about the running of the ECU?,.. and what has changed?

The only thing that has changed is that you have fitted 16vt injectors?,.. but your car doesn't really know this,.. but then maybe..it does?

So the only difference is that the car will be running rich, and the only way it knows this is via the lambda.

The assumption is that the lambda is ignored on full boost and it has no effect on full throttle running?

Jari, emailed me long time ago about his experiments with larger injectors, and he found that after a short while the lambda was making influences throughout the whole running of the car, as the ecu tries to compensate for the larger injectors. An ECU reset settles it, but only temporary .

It's like the ECU adds subtracts values based on the lambda readings throughout the maps.

How you'd get round this I don't know , maybe by bypassing the lambda with a fixed value?

---

Of course ,with the poor idle, that just Nigel is getting with the larger injectors, it could be just down to a sensor issue, but then you've bunged an acr reader onto it anyway Nigel?

Joe

Joe - I did spot your original post - my brain was too scrambled to post a meaningful reply

All my sensors are fine - PTS did think the lambda might be faulty yesterday, but their diagnostic (running off a wideband sensor) was showing pretty much the same as my on air fuel ratio meter, running off the stock lambda, so they sussed that two seperate sensors couldn't both be wrong in the same way.

The problem is so bad that when the adaptation circuit reaches its limit (75%) and PTS tried to adjust the fueling back up, the lambda would go open loop at tickover I also got error messages from the AFM, as the voltage was dropping to the point where it thought there was a short circuit.

I'm not conversant with what the ECU does and how it reads the sensor inputs, but I guess that it goes something like this

Lambda - "Ooo - too much fuel - ECU, tell the injectors to give less fuel"
ECU - "Okey dokey - injectors, give a bit less fuel please - 20ms pulse should do it - ta"
Lambda - "Oi, ECU, injectors are still giving too much fuel - sort them out, will ya?"
ECU - "Doh! - OK - injectors, sort yourselves out"
Injectors - "Eh?, you told me to give 20ms, thats what I gave"
ECU - "Oh, OK, er try 15ms then"

Loop back to top

Basically, the ECU doesn't know that it has 20% bigger injectors, so no matter what it tells the injectors to do, it'll always be incorrect.

the messages coming from the various sensors muct be all over the place.

What I can't get my head around is the fact that it all seemed OK for a couple of weeks, but then (when I started to get an air leak) it went to pot and we can't seem to get it back

I'm intrigued by the suggestion above about fitting a MAP sensor, rather than MAF - why and how?

I'm at the limit of spending now - I really can't afford to throw much more at this

Its almost at the point of putting the 20VT injectors back in, adding a 3.8 bar FPR and geting PTS to do the best they can with it. I'll end up with about 360 - 370 bhp, but at least it'll run properly off-boost and in traffic.

Nigel - sorry you've not solved the issue(s) yet. My car is missbehaving bit on startup - but nothing like yours. It got much worse after adding a 3.8 bar reg. I was told the reg would not effect the idle - is this the case? (sorry if this is a slight hi-jack Nigel...) Isn't there a unichip that can control the injectors availiable now? If so could you get one of these + load your current map to it + get PTS to tweak the injectors below 2,500 rpm - would it cost a load of money?

Joe - when on boost I thought the lamda can't change anything. Are you saying that an inncorrect idle will mean the ECU will change the base map throughout the rev range?

I don't know, but that's what Jari found out.

He found that after a short period of time the AFR would slowly change, he would adjust his FPR, but the ECU would always try and compensate.

Jari has a trick ECU, and volvo injectors (I think ) and runs very rich.

However what is confusing me ,is why Fergie got on ok for a short time, and Barbz did.

Nigel, I do think you are within the realms of using the 20vt injectors, running some aquamist and some advanced timing could well see you not far off your current figures.

both myself and flea are running 360-370bhp, depending on the day and weather,... your turbo will be more iffecient than our gt28rs at full power, so no good reason why you wouldnt get more power than 370bhp.

The other option for increased fuelling Nigel ,is to run a 6th injector , the Unichip can run this, Paul has chatted to me about this, granted the fue ldistruibution woukld potentially be uneven between the five cylinders, but with you getting 370bhp+ from just 5 20vt injectors a fifth one might let you hit 400bhp+ and if you are runnign mist ,the mist will give you some safety if you do run sl lean on one cylinder

joe

On boost i.e. open loop, the narrow-band lambda can't react fast enough or accurate therefore fueling is based on fuel maps etc. Regarding the 3.8 bar regulator, I have had this in mine for 14 months now and I have never had a problem with a smooth idle (if left on idle for a few minutes it may cut out) or closed loop so shouldn't have any problems Si. The new Unichip isn't currently available on general release in the UK although Dastek in Scotland have had them since the start of the year. They have a few bug issues with the software (now Windows not DOS) but they have been fitting them to cars up there.

You mean 6th right Joe!??!?!

Yep, sorry Simon ,I meant 6th injector

Nope it only applies to using larger injectors with the Unichip and then adjusting the idle map

Normally the Unichip does not affect the idle map so the ECU at idle with the standard 20vt injectors says everything is fine, so don't worry we'll be fine .

Joe

The new Unichip allows for provision of a MAP sensor so you can bypass the limitation of the MAF sensor, and runs a separate injector map, so you can run any size injector,..didnt know Dastek had them yet....

It may be worth trying the standard 20vt injectors with an adjustable fuel regulator. At 3 bar they flow 330cc, 3.8bar - 371cc, and 4 bar 384cc i.e. the same as the 16vt injectors. Obviously running them at greater pressure can have it's own effects but I can't think running them at 25% extra capacity is rediculous, at least not in the coupé tuning world

I think the most you can get extra out of them is 5-10% due to design limitations of the injectors themselves??

Joe

claudio at perfect touch said that the new unichip is available?

its more money though, and i dont know of it being able to control injector cycle.

Markus - ive spoken with Paul at PTS and Paul Hills at engine advantages about converting the OE ECU to run a MAP sensor, both have said it simply isnt possible.

id be interested in your method of doing it, as it would certainly be a great step in tuning and reliability.

nigel - i really think you need to do what Frag has said, run the 20VT injectors with methanol Mist or just be happy with lower power output, for a driveable car.

for you to get the 16VT injectors working properly seems like it could take alot more money, and RR time can work out very costly.

Does Matt know something we dont about this problem, he's running two fuel rails with 20vt injectors.

See Picture

Can the ECU signal for the injectors be split to run 10 injectors and then map it on those ?