Audi/VW MQB 2.0T Tuning Thread

[aaronc7]

Hello, I wanted to make a new thread to discussion MQB 2.0T EA888 Gen3 Tuning stuff! My MPI/Flex thread was kinda turning into that...seems this may be more appropriate to cover it all. Plus, Flex, MPI and other stuff was recently added to some of the Operating Systems.

On my to do list are some writeups on enabling Flex fuel, MPI, burble etc. If this turns into a good information thread, I will try to keep the OP updated with links etc.

But feel free to post any questions, share your findings, show off your recent dyno, datalog etc.

[sidejobjames]

I've got a question/comment about my 2016 gti s, 6mt. I've got some mild bolt ons (twin-cooler, 200 cell catted downpipe, injen 3000 enclosed intake, turbo inlet pipe, turbo blanket, catch can) and i've been using a JB4 on map 1 for a long time now. I'm pretty familiar with HP Tuners, but not on this MQB platform. I'd like to get more out of my GTI, and possibly replace my JB4 with some custom tuning, or use the JB4 to supplement some mild re-flash changes. (a jb4 is a pretty advanced piggy-back tuner, that allows for user adjustment and data logging, but a piggy back non-the-less).

My question: What parameters should i be logging in order to determine the most beneficial changes to my tune? And what should i be looking for in those parameters? What maps/modifiers/limitations ect. most directly effect the power output of these MQB simos ECU's? I know they're torque target based, which isn't a system i'm totally familiar with, but i think i get the general idea.

Eventually i'm going to go mpi and flex fuel, for the numerous benefits of that style system. Most of which being, for me, the reduction of the carbon deposits on the backsides of the intake valves being cleaned away by fuel getting sprayed on them. I don't know how many of you guys have pulled the intake manifold on your gti's, but after 40k i pulled mine just to see what the build up was like, and i was not happy with what i saw. I used the media blast method of removal, and although it wasn't terribly difficult, it's not something i wanna do all the time. Had i know how quickly the carbon built up i probably would have performed the de-carb every 20k, which is what i'm going to do going forward until i get a mpi system going.

Anyways, thank you in advance for any advice. Let me know what i should log and i'd be happy to post the log file, and my tune file to the thread

[aaronc7]

I'm actually redoing my logging list right now, as HPT recently added quite a few. There's some duplicate/similar items and I'm trying to figure out which one(s) are best.

- Engine RPM
- Engine torque - actual engine torque
- Driver Intent Torque 1 - requested torque
- Throttle position
- Intake air temp
- Desired MAP
- Charge Pressure Desired (PUT)
- Charge pressure actual
- Cylinder airmass (mg/stk)
- Wastegate duty cycle
- Cyl 1 Advance
- Knock 1-4
- Actual Lambda
- Mean Injection time (DI injector pw)
- STFT
- LTFT
- Fuel Rail pressure

There's certainly more I would like to have, but it's a balancing act between logging everything you need vs speed.

Cobb and Ecutek actually have decent big picture tuning guides....I won't try to duplicate that effort right now...but big picture its something like this (just shooting from the hip here, I might be missing a few details):

- Accelerator pedal % requests a certain percentage of max torque ->
- Max Torque table to give you a target Torque (let's say 400 lb-ft) ->
- 400 lb-ft gets converted to an airmass target (mg/stk) per the TTA (torque to air) tables ->
- Spark efficiency factor gets worked into airmass target to hit desired torque (let's say 1500 mg/stk now) ->
- The Speed Density / VE modeling takes 1500 mg/stk goal and converts that into MAP Desired ->
- MAP desired goes thru the TIP desired table (also known as Charge Pressure) ->
- Charge Pressure Desired is your target boost

Along the way there's various torque, airmass, air pressure, turbo speed, turbine temp limiters, etc. Some of these are mostly a nuisance and you just want to get them out of the way (some of the "max torque" type tables). Some of these limiters are actually very useful and often used to tune the car (for example, the Turbo pressure ratio table).

There's many "paths" that torque/airmass/charge pressure get computed....the LOWEST value = the target. So in the above example, you might be targeting 1500 mg/stk from the initial torque request...but if the Pressure Ratio table is telling you you are limited to a PR of 2.5....that's 22psi, which then would covert back to only be 1250 mg/stk at the point in time. (making up numbers, you get the idea)

That's obviously very big picture/high level, but it's definitely important to understand so you're not just aimlessly changing stuff.

One annoying/frustrating thing about these cars is that there are many different box codes / Operating systems. The GTI vs R tunes are very different (we think different teams worked on them separately) but even within the specific model, there are multiple box codes and slightly different strategies for that box code. The big picture stuff like this is all the same...but you'll find that a 2015 vs 2016 vs 2018 GTI tunes are all a little different and have different limiters that may or may not need addressed etc. So there can be some year/OS specific "gotchas".

@Exley on here has a 2016 as well, you guys probably have the same OS of 5G0906259D

[vw1320]

309whpmk7.jpg

So here are the results of tuning my car with HP Tuners. 2015 GTI with just a dp, turbo inlet, and IE intake. Pump 93. The tq down low is kept intentionally low as its 30* as a high here and traction is at a premium. With the recent updates HPT has become a pretty viable option. Though there are still a few limitations and things to work out its possible to get a pretty decent tune with the software as is. Thanks to @aaronc7 and a few others for the crash course on how this ecu operates and some of the differences between models and years.

[sidejobjames]

I'm actually redoing my logging list right now, as HPT recently added quite a few. There's some duplicate/similar items and I'm trying to figure out which one(s) are best.

@Exley on here has a 2016 as well, you guys probably have the same OS of 5G0906259D

Wow, man, I really appreciate the explanation. I have read through cobb's and one other companies (can't think of the name, possibly ecutek) explanation of the strategy behind these ecu's, and your explanation has helped my understanding the most. Thank you!

Do i need to keep updating the VCM suite to get the most recently added parameters for logging? Or will they be added when i sync the device?

You listed knock 1-4 as one of the parameters to log. I wanna be sure i'm understanding you correctly, you are talking about knock retard degrees for all four cylinders? Measures in negative degrees of timing for each individual cylinder.

[aaronc7]

309whpmk7.jpg

So here are the results of tuning my car with HP Tuners. 2015 GTI with just a dp, turbo inlet, and IE intake. Pump 93. The tq down low is kept intentionally low as its 30* as a high here and traction is at a premium. With the recent updates HPT has become a pretty viable option. Though there are still a few limitations and things to work out its possible to get a pretty decent tune with the software as is. Thanks to @aaronc7 and a few others for the crash course on how this ecu operates and some of the differences between models and years.

Lookin good! Hope to see more HPT results coming out. I only got a vdyno so far lol. Dyno is quite a ways away or else I would have probably hit it up already.

[aaronc7]

Wow, man, I really appreciate the explanation. I have read through cobb's and one other companies (can't think of the name, possibly ecutek) explanation of the strategy behind these ecu's, and your explanation has helped my understanding the most. Thank you!

Do i need to keep updating the VCM suite to get the most recently added parameters for logging? Or will they be added when i sync the device?

You listed knock 1-4 as one of the parameters to log. I wanna be sure i'm understanding you correctly, you are talking about knock retard degrees for all four cylinders? Measures in negative degrees of timing for each individual cylinder.

Yeah that's correct, definitely want to log knock for each cylinder. The factory system is pretty hyper active though, definitely normal to get some here and there.

Make sure you're using the latest beta always for these cars....stuff gets added pretty often and you won't see it unless you're on a new enough version. Honestly I've just got in the habit of updating it almost every time I use it. There's about 5-10 beta releases a day it seems like (encompassing updates for all platforms).

[aaronc7]

While it's fresh on my mind....datalogging. Here is my current (new) general log list:



While the new PIDs are a very welcome addition, the logging isn't particularly fast (yet), so you need to be mindful of a few things and how many items you're logging so you don't get slow logging rates.

The scanner receives right at 100 frames a second which means the overall logging rate is around 100hz. Log 20 params at full speed and you'll get ~5hz etc.

There's some weird behavior when logging a mix of SAE params and Mode 22 params (the ones not marked SAE), which I documented here: https://forum.hptuners.com/showthrea...n-Optimization

Bottom line...I recommend logging only mode 22 PIDs or weird stuff will probably happen and it will probably wreck the logging speed of the stuff you really care about.

Anyways, in the list above, I am logging the first ~18 items at 5hz, then about 5 at 2hz, then the various temps are a much slower rate. With the goal of having no slower than 5hz for the stuff I really care about.

Not the best logger in the world, but you can make it work, you just cant get lazy and select everything for every single log. Might not be a bad idea to create a few different log lists depending on what you're trying to get after.

[sidejobjames]

While it's fresh on my mind....datalogging. Here is my current (new) general log list:

Bottom line...I recommend logging only mode 22 PIDs or weird stuff will probably happen and it will probably wreck the logging speed of the stuff you really care about.

Anyways, in the list above, I am logging the first ~18 items at 5hz, then about 5 at 2hz, then the various temps are a much slower rate. With the goal of having no slower than 5hz for the stuff I really care about.

Not the best logger in the world, but you can make it work, you just cant get lazy and select everything for every single log. Might not be a bad idea to create a few different log lists depending on what you're trying to get after.

Yeah i hear ya, with regards the bandwidth of the logger i mean. I'm a technician by trade and frequently set up custom data lists on the techstream (toyota's factory scan tool) in order to get better sample rates. I didn't know that you were able to adjust different sample rates for different logged parameters.

When you're logging, do you only use the graphing section of the scanner? and get rid of the numerical values typically on the left side of the screen, and the gauges on the top?

When logging with the jb4, I would only log wide open throttle runs, usually from 2nd or 3rd through 5th or 6th (depending on the road). Is that the same way people typically log data using hp tuners as well?

Sorry for all the nube questions, I really appreciate your time in helping me out with this. And hopefully I won't be the only one finding this information useful.

[aaronc7]

Yeah, despite the VCM scanner still being a bit slow for this car...the ability to log certain things a much slower rate is a huge plus. On the Cobb AP the logging rate is something like 1900 params/s....so logging speed is never an issue.

I display both the numerical values on the left and use the graph view. Graph view is most useful for me, but I also don't graph every parameter so the left hand view is nice. Left hand view also helps you identify what a value is actually updated vs just the software interpolated value.

Just play around with the software and set it up how you like it best. I totally turn off the "gauge" view, so it's just the Param list on the left, and 5-7 graphs on the right hand side. I haven't really gotten into histograms with this car yet but I have a tab for those as well.

One pro-tip: space bar starts and stops logging, which is useful when out and about driving/logging. For most WOT tuning...the best is a single gear pull, 3rd or 4th depending on what trans you have and what you can realistically get done safely. It's good to check multi gear stuff once in a while, a good single gear pull (2500-3000ish to redline) will give you best data for WOT tuning stuff.

[this guy]

Lets talk about MPI modes and combustion modes. I'm testing out DI/MPI Mode because I don't know what wobbling mode is. Now I looked into EcuTek and they talked a little about the combustion modes but not very much. So far I just left them alone.

[aaronc7]

The MPI stuff really makes my head hurt. It gets really confusing regarding all of the different combustion modes, combustion modes index and all this. The simos18 funktionsrahmen is really the source for this... the combustion mode section.

What got it working for me may be specific to my OS (8V0906259K), others might require additional digging/changes etc... honestly I am not 100% sure. This is yet another reason I believe many of the tuning platforms for S18 go to a consolidated ROM of sorts.

Anyways, here's a start:

- MPI Mode- Wobbling mode (I have no idea what this means, some bad german to english translation probably...maybe something related to electrical term?, it's referenced in the FR...but this is the option you want)
- Combustion mode 5 - OPP MPI 2, this will be your DI/MPI operating mode. Which per the FR is combustion mode request 11 or combustion mode index 9....(it gets very confusing lol)
- Max RPM - Set Mode 5 to 65,535 or some high value
- Comb Mode Priority- Mode 6-14, Mode 12-20 (swap the two values). I literally had to dig into the FR again to figure out why I did this. I think you are giving priority to a MPI/DI mode when operating in scavenging (low rpm, high load) over the traditional DI only.
- Injector constant/deadtime - input values
- DI portion- defines the portion of overall fuel requirement that DI will provide. 1 = DI only, .50 = half/half etc. Something like .5-.6 is probably reasonable at high loads
- Valve diag- set these low so valve diag is not disabled
- Torque - Set min and Max for mode 5 to something reasonable. When you want it to kick in and set max to something higher than you'll ever see

In my definition file, there was an additional "calibration table of inhibit assignments" which I THINK you also need to disable that for combustion mode request 11. But I do not see this defined in HPT. I did not test MPI without changing this table though, so I am not sure if it was a needed or unnecessary change. Maybe someone can try these changes and chime in. If so we can put in a ticket for it. In the logger, the Injector pulsewidth 1,2,3,4 are the MPI specific ones (the ones that have the option for pulses 1,2,3 are the DI ones).

[aaronc7]

Anyone else get Flex Fuel and/or MPI working?

[aaronc7]

Guess not lol!

I purchased an OS patch to give me direct memory address logging and greatly increased logging speed. I think I am finally ready to proceed forward with my hybrid turbo Vortex install and tune it with HPT.

I'll post updates in here as I progress.

https://datazap.me/u/aaronc7/hpt-aar...=36-39&solo=36

[LAGP]

Guess not lol!

I purchased an OS patch to give me direct memory address logging and greatly increased logging speed. I think I am finally ready to proceed forward with my hybrid turbo Vortex install and tune it with HPT.

I'll post updates in here as I progress.

https://datazap.me/u/aaronc7/hpt-aar...=36-39&solo=36

Hi, could you elaborate more on the "OS Patch" that you mentioned?

[aaronc7]

Hi, could you elaborate more on the "OS Patch" that you mentioned?

Sure, it's a 'generic' OS/ASW patch applied to a raw bin file. I flashed this modified bin file, then continuing to use HP tuners to flash calibration only flashes.

So you need a way to flash a .bin file, but thankfully there are open source tools now, that you can flash with either a raspberry pi or Tactrix Openport. The above mentioned logging patch uses OpenPort to log (and others), so that's the obvious choice here. And you just need a .bin for your car...if you can find a .frf, you can easily use and convert that as well.

So yeah, it's a little bit of a cumbersome process and price adds up if you need an OpenPort as well, plus now I have to lug around 2 dongles etc......but best solution I've found so far.

Here's to hoping that HPT can implement a logging patch or similar in the future, since there is otherwise no real way to log direct memory addresses.

Shoot me a PM and I can point you to more specific info if you are interested

[aaronc7]

Quick pro-tip:

Problem: VCM Scanner "Cylinder Airmass" caps out at 1389 mg/stk....which is not a lot of torque/power--it's really helpful to see where you're operating in the timing table, a bunch of other tables reference it etc.

Solution: Log "Mass Airflow" and convert to a cylinder airmass value in mg/stk.

Mass Airflow (g/s) / RPM * 30000 = Cylinder Airmass (mg/stk).

Here's a custom math file if it will let me upload it.

[aaronc7]

If you have 5G0906259A, 5G0906264M, 5G0905259L, or 8V906259K (probably others), take a look at the latest VCM update. They added a bunch of tables in editor across the board.

Kudos to the HPT team to continuing to add more stuff that we are asking for.

[sidejobjames]

Took me forever to finally get some logs, been busy with a lot of other stuff recently. I'm really glad i finally did though, I want to get off of this piggy back deal, and start working towards my end goal of an mpi flex fuel set up. I'm really hoping by the time I'm ready to dive into the hardware side there will be more examples to learn from with regards the software requirements.
Anyways, just posting these logs for anyone who would care to comment on them. The set up is a 2016 GTI, jb4 piggy back, map 1 (+4.5 psi across entire rpm range i believe). Twin cooler, high flow cat, Injen Evo 3000 intake (enclosed). 5G0905259L operating system.3rd gear on ramp.hpl3rd gear on ramp rnd 2.hpl

[aaronc7]

Looks fine, I don't see any red flags. But I don't think you're logging anything to see boost level (MAP or Throttle inlet pressure). 5G0905259L is now well supported on the editor side, time to get moving with hp tuners! Be happy to answer any questions on getting MPI/flex fuel going.