TQS progress thread, From STD-Scroll-Eliminator.

[aragorn]

Bill: We're under no illusion that its perfect. However we worked on the assumption that as long as we run it on 98octane fuel, the management (which is designed to run perfectly well on 95) would handle any over advanced timing, with the fuel taking up the slack.

Looking at the vagcom logs from the car, the highest knock ****** logged is 8degrees. This is no different to a completely stock S4 running on 95 octane fuel. I can show you logs of a standard S4 running on 95 and 98 if you'd like to see this.

The MAF and Fuelling may be "frigged" but the maf is underreading by 38%, and the injectors are 40% larger than stock, which for the most part will balance out. This can be seen by the fact that the Lambda multiplicative correction % is sat around 3, as opposed to some wildly high number. The N75 is completely under the ECU's control. AEB has no boost sensing, so the ECU is pushing the N75 duty cycle by itself to get to what it thinks is the correct maf values.

Ideally the car needs some timing pulled out, and when it gets mapped, thats exactly what will happen. But given the values the ECU is pulling seem the same as a stock S4 running on 95, i cant see how they're dangerous?

 

[aragorn]

An exceprt from a vagcom log, unfortunately the RPM's havent captured properly, so we'll need to do it again, but thats the worst correction out of all the driving.

MAF is the 5th column, and the last four columns are the timing ******.

 

[Broken Byzan]

Not much going on atm, just some cheapy bits.

MPG seems to have settled around 34 according to the DIS and 32 in real life, so not too bad.

Stuffed some silver ring things into the S4 dash

Ordered an inlet mani gasket so i could paint the inlet manifold with wrinkle paint. £4.79 delivered,

check out the packaging.....

Using some spare bits in the "audi" box i sprayed a set of door handles black like i had before.

Not sure on whether to leave them silver to match the beltline trims, or black to blend in with the card better

Have done some more searching on immo problems, so i opened my 018CJ ECU to locate the 24c02 immo chip.

Searched around and found for a sensible amount of cash you can aquire the equipment to clone them yourself.

Using some quite basic tools,
Programmer


25 blank chips for testing...

Adapters to allow the SMD chips to fit the DIP progger.

 

[Geeman]

immo problems...?

Rings look good!

 

[Broken Byzan]

The immo issue is self inflicted mate, i am putting a non native ECU into the car as it is more programable and bench flashable.

For some reason, the ECU adapted to the immo fine on the purple one, but wont have it on this car. Having done some searching the way most tuners "fix" it is by cloning the immo info from your old ECU, can't go wrong then !

 

[jcb]

so long as you keep your boost sensible and your intake temps don't go through the roof teh ECU shouldn't need to pull timing should it?
You don't appear to be running lean and have matched fuelling to charge volume

 

[aragorn]

As a rough rule of thumb, every 1psi increase in boost pressure needs a degree of timing pulled out.

Becuase we've "fooled" the ECU into increasing the boost and have effectively hidden it by using the larger MAF and matching injectors, it will still be running the timing for 12psi, rather than the timing for 18psi so will in effect be overadvanced.

The knock sensor will detect this and ****** the timing as long as its not too overboard, and running super unleaded helps too.

 

[Broken Byzan]

so long as you keep your boost sensible and your intake temps don't go through the roof teh ECU shouldn't need to pull timing should it?
You don't appear to be running lean and have matched fuelling to charge volume

ED You are slightly luckier than I, you have a proper AFR gauge to watch for sure. I imagne looking at the corrective vaues, it's not far off on mine

 

[jcb]

As a rough rule of thumb, every 1psi increase in boost pressure needs a degree of timing pulled out.

Becuase we've "fooled" the ECU into increasing the boost and have effectively hidden it by using the larger MAF and matching injectors, it will still be running the timing for 12psi, rather than the timing for 18psi so will in effect be overadvanced.

The knock sensor will detect this and ****** the timing as long as its not too overboard, and running super unleaded helps too.

granted but that assumption is on all the same hardware being cranked up 1psi.
if you change turbo and intercooler, pipework etc then your thermal efficiency will be much better and you are probably running roughly the same inlet temps as a Ko3/Ko3s pushing 12psi.

My question was in repsonse to Bill's.
Unless you run lean at WOT or are running crazy boost and/or crazy inlet temps though the OEM map it shouldn't have any problem

 

[rasA4]

love the S4 cluster loads better rings looks good also, you saying the gasket was the only thing in that box?
quite a lot of chips on the silver beltline trims. IMHO i prefer the chrome handles

 

[aragorn]

jcb: it isnt all about temperature...

If you put more charge into the cylinder, it tends to burn quicker, becuase of the higher dynamic compression ratio etc. As a result, more power will need less timing, even if you've completely changed the turbo for a much more effecient model.

 

[slappy_dunbar]

I want to say thank you for the terrific info to be found on this thread. For some time I have been flummoxed by the skewed AF readings coming from a stock sensor placed in the 3" MAF. I would never have trusted myself to "do the math" on this question, and your multiplier factor of 1.38 seems dead on. Just as a corroboration, I had been seeing uncorrected numbers in the mid 140s with a hybrid turbo very similar to the Scroll you are using. The compressor wheel is less efficient than yours but the housing is better-flowing, so it was a big confirmation of my rough guesses when I was able to use your math to arrive right at our 200g/s target!

I would say you have every reason to be pleased with seeing anything above 200g/s on this turbo. Your tuner has done a great job dialing you in and Scroll's design seems really solid. Congratulations!

 

[Broken Byzan]

Thanks Slappy, your hybrids should be pretty good once the teething period and decent files are written. I am subscribed to your threads on VW vortex and the Zine, we all need to pool info to enable searches to be of more use IMHO

 

[slappy_dunbar]

Thanks Slappy, your hybrids should be pretty good once the teething period and decent files are written. I am subscribed to your threads on VW vortex and the Zine, we all need to pool info to enable searches to be of more use IMHO

I think it'd be really helpful to have a centralized resource of software applications to these turbos. Secondarily would be a comparative listing of the turbos themselves. Over the last couple of years pricing has come steadily down on these kinds of kit (hah! I used British) while performance is going up. It's a hoot to be part of it. I plan to do something informative like that on my own site, but that's not as impartial as we'd like.

 

[gmx]

An exceprt from a vagcom log, unfortunately the RPM's havent captured properly, so we'll need to do it again, but thats the worst correction out of all the driving.

MAF is the 5th column, and the last four columns are the timing ******.

Where is the BTDC column?

Log that and "knock detection in degrees" at the same time.
Let me know what you discover....

That is not a real correction factor block. FYI.

 

[aragorn]

so the "CF" values arent in degrees?

I'm not sure which values will be available on his M3.7 ecu, but i'm sure BTDC timing should be there.

 

[Broken Byzan]

I got bored and gave my car the first clean of 2010. OMG she was one heck of a state.

 

[Purple Princess]

She looks amazing babe xx

 

[rasA4]

looks good myman!

 

[ian52]

I still need to give mine a good machine polish like that. Dons think it will ever look that shiny thou.

 

[Broken Byzan]

I had to change the rocker cover gasket as it was leaking so thought it was time for a freshen up.I already had a can of that nice wrinkle paint so used the opertunity to paint the IM too. So it only costed me a VC and IM gasket and some time

Maybe too much black? Meh too late now.lol

While i was in there i noticed a water leak from the rear flange thingy, sent the Mrs into the closest VW stealer and £2.21 later we had a nice new O ring to seal the flange to the head.

 

[rasA4]

looking nice and clean as i said already the red brightens it up a bit, whats next?

 

[aragorn]

Hmm maybe i should get the paint out too!

Looking good.

 

[Lee Goodall]

I'm sure I've seen one similar somewhere??

 

[Broken Byzan]

I'm sure I've seen one similar somewhere??

I don't know what you mean,


I like how my nice yellow injectors satnd out !! lol

 

[Broken Byzan]

Hmm maybe i should get the paint out too!

Looking good.

Doooo eeeeeettttt It's not like you don't have the paint !!

 

[Lee Goodall]

I like how my nice yellow injectors satnd out !! lol

Hmmm.. I think red and black would be better lol

 

[docurley]

Nice will do mine when I need to pop the rocker off I quite liked the look on my old B5.

 

[gmx]

looks great!
Any details on the FMIC + piping?

 

[slappy_dunbar]

I have some added info. I had the 6.4 last time, this time it's the 7.6mm.Not sure what traits that will bring..

46x42x8mm K04-015 (clipped tip)
50x42x6.4mm (S3 TT)
50x42x7.6mm (non VAG)
50x46x_._mm (RS6)


Please make sure to mention we are not building "outlaw" upgrades from K03 turbos, only K04. The time required to alter a K03 is astronomical and we no longer have the time at our new workshop as the day to day rebuild work is hectic.

Mark -- I just re-read this thread and came across this posting. This means you are using a 7.6mm blade tip height turbine rotor in place of the 6.4 height rotor found in the K04 turbos. This change in spec will slightly alter the flow characteristics to reduce back pressures. You aren't wired for EGTs are you?

Point of clarification: I'm not bringing this up as a criticism. Quite the opposite. I concur with this design alteration.

 

[Broken Byzan]

I wish i was wired for EGTs

No critisism taken, it was a trial Turbo Sean did for me, slightly different to the normal specs. I don't think it will have a massive impact on output, just in the way it spools. Either way the tune is going to be done for it on the RR

 

[ALPINE]

what made you paint the inlet marky boy?? hahaha!

 

[slappy_dunbar]

Mark

I have a tester over here who is showing what I believe to be inaccurate mass airflow numbers. He is using an OBD2 VR6 MAF housing (this is the version which retains the sensor orientation of a newer 1.8t transversal OEM housing. So the sensor is oriented in the airflow exactly as it had been, but now it is in a (presumably) 38% higher volume housing. Why then are the uncorrected and corrected numbers not lining up with what you'd expect? I am going to have him run fuel trim logs to see if he is running super rich, but he says the car is running "perfect".

 

[aragorn]

Has he increased his fuelling by 38% too?

I see a peak of 242g/s which is around 300hp, were you expecting more or less?

 

[Broken Byzan]

Thats basicly what i am running.
My std correction figures are a few % Mult and Add.

I will find out actual lambda figures and power figs etc next week , She is booked in for a day on the 7th @ Rtech.

I will get him to log as much as is practical, boost, MAF etc.

 

[slappy_dunbar]

Has he increased his fuelling by 38% too?

I see a peak of 242g/s which is around 300hp, were you expecting more or less?

This turbo should put out ~200g/s. This is a Mk1 F4H-T, so it's using a compressor wheel that is smaller than a TT225/S3 wheel.

The graph above is the actual numbers reported by the sensor to VAG-COM and secondarily shows my estimates using your 38% multiplier. Remember his car has a MAP sensor so yes we have larger injectors. The Unitronic software file calls for an OEM 1.8t sensor in a 3" MAF housing (either TT225 or VR6 version OBD2) and 380cc injectors. I asked him to get me his Lambda correction values, expecting to see the ECU working like mad to address the improper mixture. Instead corrections are +3%, meaning he's running an insignificant amount LEAN.

I've forwarded his log to the tuner to see if they can sort this. And, oh yeah, the car is driving perfectly. No idea how. Should I be fretting, really?

 

[aragorn]

If the map is designed for a 3" MAF, then the vagcom output might already be corrected, or at least scaled so the numbers no longer make sense.

IE it might only making be 170~

The fact he has a MAP sensor means you cant fool the ECU like you can on the AEB, as the ECU would notice its overboosting and cut the power back.

 

[slappy_dunbar]

If the map is designed for a 3" MAF, then the vagcom output might already be corrected, or at least scaled so the numbers no longer make sense.

IE it might only making be 170~

I've thought about this. The tuner can answer this question, although in my recollection they said their software does not alter the data coming from the sensors.

The fact he has a MAP sensor means you cant fool the ECU like you can on the AEB, as the ECU would notice its overboosting and cut the power back.

Exactly. Can't be sneaky like you've been. At least not on this particular car,

 

[aragorn]

might not alter the data coming from the sensors, but they're altering the data the sensors are being referenced against....

An S4 ecu will have a MAF table that says 5v = 300g/s, the 1.8t's will say 5v=220g/s (or whatever) if they altered that table to reflect the new housing 5v would now = 300 (or whatever)

 

[slappy_dunbar]

Understood. But then there's the fueling. We're showing 20+/ms injector duty cycles. Basically the 380s are being maxed. This is at 3bar.