Golf 4 ALH EDC15

Posts related to specific vehicles, or any other general tuning info.

Golf 4 ALH EDC15

Postby nexus665 » Sat Oct 20, 2012 2:09 pm

Hi,

a short introduction is in order - I've had my car remapped by a local guy a couple of times and done some hardware changes. Ever since the tuner installed an 11mm pump piston for the VP37, I've had boost issues and since have blown not one, but two turbos using this remap.

The first I was inclined to believe was old - the second dying in the same way shortly afterwards says otherwise.

In detail, both turbos blew at low load/RPM - massive overboost spike IMO because the N75 map hadn't been altered around torque max, only at high RPM. Problem being, I did not have a boost gauge installed, only logged via VCDS - and that has rather poor resolution to catch such spikes.

Both times - RPM at around 2-2.5k, WOT acceleration, after not even a second, between 2500 and 3000 RPM, a loud "pfweeeeeet" from the turbo and a large cloud of smoke out the exhaust. First time in 5th gear, second time in 2nd gear...

So that's why I decided to learn more about remapping - and I've been reading til my eyes were sore since. Still, it's a rewarding subject, you learn a lot.

Hardware setup

Platform
Golf 4 ALH
EDC15 038906012FA / 0281010650

Pump/injection
VP37 with 11mm pump piston (from an automatic, serviced and fitted with new gaskets)
Bosio .216 injectors (P764)

Drivetrain
Single mass flywheel
Sachs race clutch (organic)

Aspiration
Stock GT1749V VNT Turbo (up til now), GT1752VB hybrid VNT soon
Stock 250kPa MAP sensor
R32 DSG intake (probably aided in the 2nd turbo death - was recommended by the tuner to make it spool up more easily low down - which it did, fitted 3 weeks before the 2nd one went)

Power is or should be limited to about 140-145 HP, at least that's what I requested to make the turbo survive until the hybrid I ordered arrives. Well, it didn't make it, the GT1752VB hybrid will probably come next week, 2 weeks too late ;)

Plans

Now to what I'm planning to do - since the remap I'm currently using is decalibrated, showing 48,8mg/stroke max and limited to 51mg in diagnostic output while injecting ~58mg due to nothing being adjusted for the larger injectors - is to start over, basically, and develop a basic (not pushing anything) remap that suits my current hardware and go on from there.

All of this with extending diagnostic scales and removing the various limiters to make everything be output via diagnostics as well as have real values inside maps vs. having to recalculate everything for decalibration.

Since the tuner read locked the ECU, I haven't yet gotten a dump of the current software - however, that's not really so bad since I have the original file for my car and want to learn anyway.

Starting out

Now for the first step - extending diagnostic scales, removing various IQ limiters (51mg, scale to 70 or 80), scale pump voltage map to near real world values for the injectors (this will be tough, hope someone can help point out how, more duration for lower IQs, less for higher, but how much?), adjusting duration/SOI tables, calculating IQ for ~140HP (~56mg/stroke judging from the 60mg the ARL injects for 150HP) and the rest of the maps accordingly, boost max of 1.3 Bar. Torque max at around 2.5k rpm, not too much IQ before 2k, etc etc etc.

Even though the hybrid that will be in the car can take up to 1.8 Bar continuously and overboost to 2.05, that will be all that I will start out with - there will be plenty of headroom, but I'd like the car not to eat any more turbos or anything else.

Next will be a 300kPa MAP sensor and adjusting linearisation for that. I think I have all of the necessary maps figured out, but how to adjust them sensibly is a different question ;)

Also, a boost gauge is now finally ordered...really stupid of me not to fit one right away after the first turbo died or even before, after remapping...

I've attached my stock file, next I'll post the maps I currently have and what I think I need to adjust to achieve the goals outlined above.

Regards
You do not have permissions to view the files yet. You have to be registered and you have to make at least 3 quality / unique posts.

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Re: Golf 4 ALH EDC15

Postby nexus665 » Sat Oct 20, 2012 9:57 pm

Here are my currently figured out maps -

CurrentMaps.gif

(all these of course exist multiple times, only showing one coding)

I've also attached the current WinOLS file with all the map defs.

Found some additional maps that I'm not quite sure about, also got some map names from other files and probably missing one or two selector maps.

Still, it's a start :)

Regards
You do not have permissions to view the files yet. You have to be registered and you have to make at least 3 quality / unique posts.

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Figuring out fuelling for power/torque

Postby nexus665 » Sun Oct 21, 2012 2:32 am

Hi,

forgot to mention that cat is deleted and EGR mechanically deactivated, also the 5th gear has been changed from 0,756 to 0,7 for better economy and more Vmax.

A better intercooler will also be fitted, probably a larger downpipe as well as a modified exhaust.

Fuelling
To start out with I'll figure out necessary fueling for planned power levels - I'm using values from other similar cars to guess at probable IQ.

Since I'm planning a max of around 140-150HP / @ 4000 RPM to start out with, I'm using the values from an ARL (54mg/stroke for 150HP/262,5Nm@4000RPM, torque max of 320Nm@1900) to estimate about 50,4mg/stroke at HP max. Torque max will be around 2300-2500 RPM, 320-330Nm, or an estimated 65,8-67,9 mg/stroke, with a smooth reduction down to HP max.

So current fueling levels would be about:

61,7mg/stroke or 300Nm @ 1900 (if available boost is sufficient and spikes are controllable, should be doable)
67,9mg/stroke or 330Nm @ 2300
53,5mg/stroke or 260Nm @ 4000

I'm planning to scale up to about 180HP - ~315Nm or 64,8mg/stroke @ 4000 RPM, using the hybrid turbo and even larger injectors, if necessary. However, I've found lots of differing opinions on this, so I'll just have to find out myself whether they are necessary or the 216s with the 11mm plunger are enough. I definitely know it's more fuel-efficient to use smaller injectors if the requested fuel can be injected in a safe window than larger ones - but I'm just not sure about how pump voltage relates to injection duration.

For this reason, I'll be extending the diagnostic scales and removing the 51mg/stroke IQ limits. 70mg/stroke would give a projected torque max of ~340Nm, so I'll go to 80 instead (max of ~390Nm, more than enough), changing this more than once is probably a big pain in the ass.

Pump voltage map adjustment

This is what I'm having a bit of trouble with - I've read a lot about adjustment of this map with different injectors and pump plungers, but still, I have only a very general idea about the real injection quantity possible with an 11mm pump piston from an ALH automatic with Bosio .216 mm P764 injectors or any other 216 injectors in general. Figuring this out through many log runs will be very very time-consuming...

Actually, I did find a graphic here comparing .184 .205 and .216 injectors on VP37 Golfs, but with a 10mm pump plunger. Still have to figure out what exact effect the larger plunger has - different leakage losses, but increase proportional to 10mm-11mm diameter otherwise?

Also, I haven't been able to find any information about how pump voltage equates to injection duration - I'm guessing this is also different for differing plunger/injector pairings due to hydraulic differences, but not sure how to figure this out at all.

Will need to find out to be able to do a solid remap, though :) Help would be appreciated.

Regards

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

First remap

Postby nexus665 » Sun Oct 21, 2012 10:13 pm

Hi,

after more searching around, I found that an AFN has an 11mm plunger and .205 injectors (190 Bar 1st stage vs. 220 Bar on mine). There are some EDC15 AFNs as well, I found a dump file from one and the pump voltage maps inside (up to 60mg/stroke).

From what I understand, the 2nd stage is at 300 Bar for both, so main injection should be identical, but 1st stage/pilot injection to start ignition happens later with the 220 Bar versions.

So basically, the 220 Bar OEM maps for SOI will - probably - be remapped for this, and the pump voltage itself should be comparable - since it governs duration, not SOI, but SOI is changed by the different stage1 opening pressure. Will have to investigate more.

This is an assumption - but based on this, I'll combine the .205->.216 changes and the AFN 11mm pump map with 205 injectors and that should give me a starting point for extending the pump voltage map.

Without disassembly and remapping, the pump voltage map on my ECU cannot be extended - there's another map right after it. So the stock amount of values will have to suffice and I'll have to change some axis values.

So this is my resulting pump voltage map up to 75mg/stroke- everything > 50mg is interpolated and probably needs more adjustment, the rest are taken from the maps above. I realise that there are more limiters and scales to be changed to make this work - this is just the pump voltage map, for now, one thing at a time.

My very first remap ;)
PumpVoltageMap-11mm-plunger-216-injectors.gif


These values need to be multiplied by 1.221001 to be millivolt, that's straight from the dump. Maximum safe pump voltage is 4,85V/3999 decimal in map from what I have read, so I'll stay below that.

From what I have calculated from this pump voltage map, I can inject up to about 73mg/stroke @ 2500 RPM, falling to about 67mg @ 4000 at 4,85V - these would be the practical limits that we can't cross, but that would be enough for my project IMO - 180HP @ 4000 RPM, ~350Nm @ 2300-2500 RPM. SOI will probably have to be adjusted as well, after further research, I'll also have to find out VE to check how much I need to modify my values.

My interpolation has yielded lower numbers, though - will need to find out whether it's wrong or whether the pump cannot deliver enough fuel for my long-term goal with the 11mm plunger and .216 injectors after all.

Regards
You do not have permissions to view the files yet. You have to be registered and you have to make at least 3 quality / unique posts.

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Re: Golf 4 ALH EDC15

Postby RPM freak » Mon Oct 22, 2012 12:51 am

11mm plunger is used only in cars with automatic gearbox. With 11mm plunger you will have smaller injectors then .205.

User avatar
RPM freak
Site Admin
 
Posts: 621
Joined: Thu Mar 01, 2012 9:49 pm
Location: European Union

Re: Golf 4 ALH EDC15

Postby nexus665 » Mon Oct 22, 2012 1:17 am

Yes,

I'm aware of that - however, I've fitted .216mm injectors together with an 11mm pump from an automatic.

So that's why I need a pump voltage map to reflect the hardware I have - which I tried to make above up to 75mg/stroke projected maximum for my long-term goal of around 350Nm.

My real question is, how I can calculate this for-real instead of using someone else's maps/observations - or can I only do this using logging?

Should be calculable using hydraulic formulas, at least approximately - but I don't know anything about those. In any case, a working pump voltage map IMO is the first step towards a solid remap without decalibration. It's just that I haven't found any 11mm pump with .216 injectors in another car so far...so can only interpolate the maps since I don't know how to calculate them correctly.

Regards

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Re: Golf 4 ALH EDC15

Postby RPM freak » Tue Oct 23, 2012 1:38 am

This kind of remap when a lot of hardware is changed needs a lot of work. You can not get superb results in a firs go. A lot of calculations are needed for this kind of work. You will need to make few maps and then do tests on the car. The best start for you is to look for Volkswagen Transporter T4 2.5 TDI 150hp ori maps. This engines use .216 injectors and 10mm pump in standard setup.

One more thing that you must do is to add a lift pump in the tank to achieve the desired 180hp. Your VE pump cant flow enough fuel from the tank to supply the injectors.

User avatar
RPM freak
Site Admin
 
Posts: 621
Joined: Thu Mar 01, 2012 9:49 pm
Location: European Union

Re: Golf 4 ALH EDC15

Postby nexus665 » Tue Oct 23, 2012 11:56 am

Ah,

thanks for commenting on the additional fuel pump, I wasn't sure whether it was necessary, either.

I realise this kind of map doesn't get perfect without testing - will take a while to get it right even if you calculate it beforehand.

Yes, I've read about the T5 - will need to check out a dump file.

Regards

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Re: Golf 4 ALH EDC15

Postby nexus665 » Wed Oct 24, 2012 8:51 pm

I meant T4 2.5 not T5 :) That was the one with .216 and 10mm, thanks for the reminder - wanted to look at that but haven't yet.

The hybrid GT1756VB is about to arrive - once the car is running again, we will see what's what.

Also, the boost gauge just arrived, will be useful in keeping the turbo alive. In the meantime, I'll see how far off my map is when comparing to the T4 2.5 one.

Regards

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Removing diagnostic boost limits

Postby nexus665 » Sat Oct 27, 2012 11:35 pm

Hi,

still searching for a T4 2.5 file, probably just blind :) I'll find it eventually.

Adjusting diagnostic boost limits

Instead, I've adjusted the diagnostic limiters for boost so the max displayed value now is 3315 mBar, since the stock linearisation up to 4,85V goes to 3300 mBar, as well, and I'd like to see any spikes.

In case anyone is wondering how to do this, there's a great thread in a parallel forum explaining how to do it.

In short, you have several places in your EDC15/MSA15 dump file where diagnostic limits are set for boost - 2601 mBar max for 90/110hp models, 2601 max requested and 2988 max displayed for 130/150hp models and so on.

These limits are set in two 8-bit values and one 16-bit value - the two 8-bit values are message type and factor, the 16 bit value is also a factor, just in 16 bit...


How to do it

It works like this - models that have 2601 as SVBL will also have the diagnostic limits scaled so ~2620 is the max displayable value.

Here, search for "18 255" in 8-bit mode. There are several places in your file where these exist, change them all or none of them - do NOT change just one or two! Reason: the factor we're changing is used by ALL boost/pressure calculations - so if you change it, you need to change all of them to make the values be correct.

Close to these "18 255" 8-bit values there will be another 16-bit value, for 2601 mBar SVBL cars this will be set to 980.

Now, we need to calculate a bit. Say, we're mounting a 3 bar MAP sensor which reads up to 3300 mBar @ 4.85V, which is where the stock linearisation ends. So let's calculate what values to use for diagnostic limits.

  • Message Type
    This is set to 18 by default, looking up this message type in the KWP1281 spec tells us it means "0.04 * A * B".

    The ECU gets 8-bit values from 0-255 which it needs to convert into usable boost values. The max boost pressure needs to fit inside the 8-bit space, so the max boost pressure we can get with the stock formula is:
    0.04 * 255 * 255 = 2601

  • 8-bit Factor
    We want quite a bit more than that, so we need to find a suitable message type (type 8 works nicely, "0.1 * A * B") and need to calculate backwards:

    X = 3300 / 255 / 0,1
    X = 129,412

    So by setting the message type to 8 and the 8-bit factor to 130, we can achieve a maximum diagnostic display value of 3315 (0.1 * 130 * 255).

    In my file, there are 4 places per coding where the 8-bit message type and factors need to be changed.

  • 16-bit Factor
    Now to make the diagnostic interface display correctly as well, we need to adjust the last value, a 16-bit factor. Search for "980" in a 2601 SVBL car relatively close before or after the 8-bit values. There are more places where "980" appears, but those are clearly within map descriptors, which you can easily see by looking at the values around them. Do not change those.

    Now, we had 980 as a starting value, and we need to change this accordingly. This value is calculated as follows:
    255 / mBar max = 0,0xxx -> xxx is the number we need, 980 in the stock file -> 255 / 2601 = 0,0980

    So for 3300 mBar:
    255 / 3300 = 0.0772 -> our 16-bit value needs to be changed from 980 to 772

    On my file, there are two places per coding where this needs to be changed.

Hope someone finds this helpful, I sure did :)

Regards

P.S.: hint - on cars with higher boost limits than 2601, search for "8 120".

nexus665
Silver
 
Posts: 71
Joined: Wed Oct 17, 2012 6:04 pm
Location: Austria

Next

Return to General tuning

Who is online

Users browsing this forum: No registered users and 2 guests