Ask/Answer Engine Calibration Question

[sponaugle]

I am starting this thread as a place for Subaru tuning (engine calibration) questions to be asked, discussed, and answered. I have come across a number of interesting questions spread across a lot of different (especially build) threads, and I don’t want to pollute those threads with off topic replies.

If you have a calibration questione related to tuning a Subaru engine for your 818, ask away.

I’ll provide a little background on my experience, and do my best to provide the most technical and precise answers possible. I have been in the past, and am today but to a much smaller degree a professional engine calibrator. The word ‘professional’ is a difficult one to define in this field as there is no real accreditation mechanism or nationally recognized education or testing metric. Most people consider themselves a professional when they are doing it as part of a profession that pays them. Some do it as part of owning or operating a shop, some fly around the country applying their specialty in a specific market, and some work for a shop as an employee. I have a BS in Electrical Engineering and a BS in Computer Engineering with a focus in Microprocessor design. In college I worked at Allied Signal Aerospace Engine Controls working on software engine controls for turbine engines. After college I went to Intel to work on Microprocessor system software, and I have been in the technical engineering field since.

My particular entry into engine calibration happened in 1998 when I decided I should supercharge my 1996 Integra GSR because that is exactly what a light front wheel drive car needs.. more power. There were not very many tuning options at that time, but Hondata was one of them. Before jumping on the Hondata I built my own PIC microcontroller based MAP and IAT sensor recalibrator which used a 3D lookup table mechanism to provide MAP and IAT offsets to the factory ECU based on above atm. Boost pressure. It worked, but it was a hack of hacks.

In 2003 a group of friends and I had the stupid idea to start a tuning business and shop, and PDXTuning was born. Over the next three years we became both EcuTek and Cobb dealers, purchases a 4WD mustang dyno, and tuned somewhere in the neighborhood of 500-600 subarus. Post 2006 I left PDXTuning and my friend Tim Bailey (head of calibration at Cobb) and I started Surgeline Tuning which was eventually purchased by Cobb tuning. For a couple of years I was a research calibrator for Surgeline Tuning and I still do R&D tuning at Surgeline as time permits. It was an incredible time of learning and fun, and great way to blow a few hundred thousand dollars.  I have been dumb enough to spend more the 20 straight hours on the dyno without a break, and have had some spectacular failures especially in my personal cars.

My viewpoint on tuning is bent towards the scientific side with a focus on research, so I tend to not believe everything I read without some real world testing. There are a lot of half truths, whole truths, and complete fabrications in the ‘calibration’ world, but I believe strongly that a scientific process can sort that out.

Here are some sample threads of my style of information collection and presentation:

http://forums.nasioc.com/forums/showthread.php?t=2472813
http://www.gtrlife.com/forums/topic/94535-2014-dyno-tuning-before-modifications-lots-of-data/
http://www.gtrlife.com/forums/topic/95459-2014-dyno-tuning-stage-23-full-bolt-ons-plus-e85-lots-of-data/
http://forums.nasioc.com/forums/showthread.php?t=1645176
http://forums.nasioc.com/forums/showthread.php?t=873912
http://www.ls1gto.com/forums/showthread.php?t=522194

As most of you know there is a ton of great information on Nasioc, and I encourage any and all of you to use that resource as you can. On the flip side there are millions of posts in the archive to dig thru, and a lot of bad information mixed in with the good.

As for platforms, I am of course very familiar with the factory reflash systems including Cobb, EcuTek, and Open Source, as well some standalones like ViPec, Link, Autronic, Hydra, AEM, Haltech, HPTuners, Megasquirt, and a few others even less common.

So, ask away, answer away, comment, ridicule, laugh, and remember that the exchange of ideas is what a hobby is all about.

Jeff

[Quiny]

Great idea to start this thread. I have a question about the Cobb accessport. If I bought a vehicle that was flashed and then not unmarried to the ECM by an accessport will I still be able to access it with openport or romraider?

[Scargo]

Don't know if there is an easy answer, but my mind freezes every time I try and do research and think about solutions... There are so many combinations, options and brands... My fear is that I literally spend thousands on parts and find it's very compromised in one aspect or another.

I am building an R from scratch. No donor car. The goal is NASA racing. I'll do all the wiring and install myself. I don't want to say that the sky is the limit for the budget on this aspect of the car, but I want it to be well-supported, current technology that is compatible for expansion and "accessorizing". I don't think I want anything that requires "open-source" skills to use or adjust. For example, I have a Cobb AP and custom maps from dyno tuning for my STi. I've loaded AccessTuner Race onto my laptop and PC, but never ventured into actually using it to tweak the tuning. OTOH, I build my own PCs, have done Mil-spec wiring and have built electronics. I'm just not yet ready to venture into tuning or modifying a map myself without some hand-holding.

I need it all: engine management, digital display (like Aim MXL, Race Logic Dash Pro, etc.), gauges*, data acquisition, video interfacing (I have GoPro Hero3, Black Edition) and possibly, though not initially, ABS/traction control.

I have Iphone, Ipad, Win 8.1 laptop and a OBD-II Bluetooth CAN-BUS tool. I want to be able to use a program to analyze performance data and it would be nice to have it tied into the video. Again, I currently use a GoPro.
*I have SPA dual digital gauges in my STi and though I have not used the features, I understand they can be programmed to trigger events. If a system controller can do all that, and more, I might just want a couple of analog gauges to backup or fill-in what the display might miss or not be the best at displaying in racing conditions.
Thanks PDXTuning (https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&uact=8&ved=0CB8QFjAA&url=https%3A%2F%2Fwww.pdxtuning.com%2F&ei=WgC4U7ztHIWiqAbOw4HwBA&usg=AFQjCNH5jY7-gyfcKmYTBMXCS8xY8Uziqw&sig2=9a1vm5NfAAhfM2kkdTjNXg&bvm=bv.70138588,d.b2k), in advance, for any help you can provide and for starting this thread.

[sponaugle]

Great idea to start this thread. I have a question about the Cobb accessport. If I bought a vehicle that was flashed and then not unmarried to the ECM by an accessport will I still be able to access it with openport or romraider?

It depends on when the car was last flashed. It is was pre 2008, it may not be locked and may be directly flashable. If it is post 2008 but was married with an older AP, it also might still be flashable. If it was married post 2008 with a new AP at that time, it is probably locked. The only way to tell is to try using EcuFlash to flash it. It is is locked you will get an interface error.

There are two options:

*1 You can send the ECU to Cobb and they can unlock it.
*2 If you can find someone with a 'compatible' AP for your car ( a version 2 for the same model/year range ), you can have them unmarry the AP from their car, marry it to your car, then unmarry from your car, then marry back to their car. That will remove the lock as the ECU will be left unmarried.

Jeff

[freds]

It depends on when the car was last flashed. It is was pre 2008, it may not be locked and may be directly flashable. If it is post 2008 but was married with an older AP, it also might still be flashable. If it was married post 2008 with a new AP at that time, it is probably locked. The only way to tell is to try using EcuFlash to flash it. It is is locked you will get an interface error.

There are two options:

*1 You can send the ECU to Cobb and they can unlock it.
*2 If you can find someone with a 'compatible' AP for your car ( a version 2 for the same model/year range ), you can have them unmarry the AP from their car, marry it to your car, then unmarry from your car, then marry back to their car. That will remove the lock as the ECU will be left unmarried.

Jeff

I'm wrestling with the same problem. I have installed/deleted/reinstalled etc. ECUFlash and Romraider on two computers one with Win7 and one with XP.
My donor is 2006 WRX
I cannot get into my ECU
This is what I get:
[14:52:58.498] Device Firmware Version: 1.11.3298
[14:53:41.993] kernel get version
[14:53:42.368] SSM2 init
[14:53:42.508] SSM2 ECU ID is 43 12 58 40 06
[14:53:42.602] Requesting Seed...
[14:53:42.649] Sending Key...
[14:53:42.695] interface close
[14:53:42.727] interface close

After reading lots of threads and trying lots of "home-spun" remedies, I believe I am well and truly locked.
Two questions;
1. Is there a definite way to know if my ECU is locked;
2. Is there any way to determine if it is Access port (Cobb) locked or ECU Tech locked?

I'd hate to negotiate with Cobb,send them the ECU and then find out it is not AP locked etc.

I want to avoid the simple, but expensive, solution of buying a brand new OEM ECU.

Your comments will be greatly appreciated.

Thanks

fred

[sponaugle]

Don't know if there is an easy answer, but my mind freezes every time I try and do research and think about solutions... There are so many combinations, options and brands... My fear is that I literally spend thousands on parts and find it's very compromised in one aspect or another.

I am building an R from scratch. No donor car. The goal is NASA racing. I'll do all the wiring and install myself. I don't want to say that the sky is the limit for the budget on this aspect of the car, but I want it to be well-supported, current technology that is compatible for expansion and "accessorizing". I don't think I want anything that requires "open-source" skills to use or adjust. For example, I have a Cobb AP and custom maps from dyno tuning for my STi. I've loaded AccessTuner Race onto my laptop and PC, but never ventured into actually using it to tweak the tuning. OTOH, I build my own PCs, have done Mil-spec wiring and have built electronics. I'm just not yet ready to venture into tuning or modifying a map myself without some hand-holding.

I need it all: engine management, digital display (like Aim MXL, Race Logic Dash Pro, etc.), gauges*, data acquisition, video interfacing (I have GoPro Hero3, Black Edition) and possibly, though not initially, ABS/traction control.

I have Iphone, Ipad, Win 8.1 laptop and a OBD-II Bluetooth CAN-BUS tool. I want to be able to use a program to analyze performance data and it would be nice to have it tied into the video. Again, I currently use a GoPro.
*I have SPA dual digital gauges in my STi and though I have not used the features, I understand they can be programmed to trigger events. If a system controller can do all that, and more, I might just want a couple of analog gauges to backup or fill-in what the display might miss or not be the best at displaying in racing conditions.


Easy answers.. yes I think you are correct that may be hard to find. :)

First a small correction; I am not involved with PDXTuning any more. Tim and I both left in 2006, and eventually Cobb purchased the old PDXTuning shop location. PDXTuning is still going, and I think Jarrad recently opened in new location.

On to your question: Since you are building a dedicated race car I assume you don't have to have OBDII ( or does the car need to have registered plates to do NASA? ) That makes a big difference in what you can choose. If you don't need OBDII emissions testing compliance there are a lot of standalone ECUs you could pick from. It sounds like you don't want to dig into tuning it yourself, so the biggest limitation will be what your local tuner is familiar with. Many of the higher end aftermarket standalones do dash integration. ViPec, Link, Motec, Autronic, etc all do that with the AIM dash.

If the car is not street legal, you will have to plan for some significant dyno time to get everything dialed in. Is there a local shop with a dyno you will be using?

If you did go with a factory ECU, you probably want to stay before 2006 as they don't have the immobilizer in the ECM. You can do the immobilizer, but is just adds extra stuff to wire in and you will need a dealer to make new keys, etc. ABS is something you can do, but if you try to do a 2008+ ABS with TC there is a lot of stuff to get right. You will need everything on the CAN bus needed for ABS, and it appears there are some signals from the DCCD controller in that case.

I would consider using the racelogic traction control as it is very easy to install and very configurable, and it will work with virtually any ECM choice.

If you are using a factory ECU that is supported by Cobb, you can use the Access Tuner to tune the car yourself. Depending on what kind of power you are planing for one benifit of starting with a factory ECU is you have a good place to start the tune from. Using Cobb has the significant advantage of having realtime tuning which means you don't have to reflash for every change. In capable hand it can cut down dyno time by 50% or more. Obviously you can also use the open source tools if your ECU is supported, and that can work great if you put some time into it.

Jeff

[sponaugle]

[14:53:42.649] Sending Key...
[14:53:42.695] interface close
[14:53:42.727] interface close

After reading lots of threads and trying lots of "home-spun" remedies, I believe I am well and truly locked.
Two questions;
1. Is there a definite way to know if my ECU is locked;
2. Is there any way to determine if it is Access port (Cobb) locked or ECU Tech locked?
fred

That message is typical of what happens when the ECU is locked. The ECU locking happens by changing the key exchange used to start the reflash process. By changing the keys away from the Subaru ones the OS tools can't flash. I would assume from this that your ECU is locked.

Just based on probabilities alone your best guess is that an AP was used, but as for determining for sure, I'll have to think about that for a sec. Both EcuTek and Cobb change the bootloader flash code to use a unique key, so it is hard to determine without either tool.

If you had an AP for that year/model, I believe the marry/unmarry trick will work. I'll check and see if there is another way to determine the flash type without the flash tools themselves.

Jeff

[DodgyTim]

Thanks for putting your time into this thread.

My issue is opensource tunes available for Australian ECU's
I have been able to purchase a "Stage 2" map for my 2006 ADM WRX, which will probably get me through the go-kart and registration stages of my build.
Eventually I'll be going bigger turbo and other mods, closer to a "Stage 3" setup:
-VF34 ball bearing turbo (from a JDM spec STi model)
-Bigger Intercooler
-GM Boost Control Solenoid
-Aeromotive 340 Fuel Pump
-Stock 06/07 550cc WRX Injectors
-3" Catted Turbo Back Exhaust
-Injen CAI

There don't seem to be a lot of stage 3 maps off the shelf to suit my computer (A8DH200O) as a baseline map.
XPT tuning do a map for the USDM 2006 WRX that has quite similar mods, their Stage 3 race map.

If I purchase a map like this for a USDM computer, how hard/how risky is it for me to convert it to the ADM computer?
Cheers
Tim

[sponaugle]

Thanks for putting your time into this thread.

My issue is opensource tunes available for Australian ECU's
I have been able to purchase a "Stage 2" map for my 2006 ADM WRX, which will probably get me through the go-kart and registration stages of my build.
Eventually I'll be going bigger turbo and other mods, closer to a "Stage 3" setup:
-VF34 ball bearing turbo (from a JDM spec STi model)
-Bigger Intercooler
-GM Boost Control Solenoid
-Aeromotive 340 Fuel Pump
-Stock 06/07 550cc WRX Injectors
-3" Catted Turbo Back Exhaust
-Injen CAI

There don't seem to be a lot of stage 3 maps off the shelf to suit my computer (A8DH200O) as a baseline map.
XPT tuning do a map for the USDM 2006 WRX that has quite similar mods, their Stage 3 race map.

If I purchase a map like this for a USDM computer, how hard/how risky is it for me to convert it to the ADM computer?
Cheers
Tim

You certainly could start off with a purchased USDM map and convert tables over, and it would help if you started with the closest map possible (same model year in the USDM market). Based on your modifications it looks like you are using the same stock injectors, better fuel pump, and a larger turbo. A setup that like can be tuned assuming you are willing to start slow. If you are not going to take to the car to a dyno, you need to make sure you have a good wideband installed, even if you get a basemap from someone. If you have a good setup to do logging including a wideband, a remote tuner can do a lot.

Since your fuel is different, I would have the basemap start with reasonable but conservative timing, low boost, and work on getting the AFRs in line. After that is well calibrated you can work on timing and boost. While you can do a lot with road tuning, it is difficult to really get a perfect calibration as you can't see how small changes effects torque... and of course a loading dyno makes that much better.

Jeff

[Scargo]

... I assume you don't have to have OBDII ...there are a lot of standalone ECUs you could pick from. It sounds like you don't want to dig into tuning it yourself, so the biggest limitation will be what your local tuner is familiar with. Many of the higher end aftermarket standalones do dash integration. ViPec, Link, Motec, Autronic, etc all do that with the AIM dash.
Is there a local shop with a dyno you will be using?...

I would consider using the racelogic traction control as it is very easy to install and very configurable, and it will work with virtually any ECM choice.
... Obviously you can also use the open source tools if your ECU is supported, and that can work great if you put some time into it. Jeff
I think I want a standalone ECU. One thought is to put an EGT sensor on all cylinders. I believe standalones have the potential for many inputs?

I use EFI Logics in Bethel, CT for my dyno tuning. They are one of the best. I believe Mikey can tune using any major system. I am not opposed to working with someone to supply me with maps based on my data. OTOH, I do want to put the time into gaining the knowledge necessary to evaluate the data and eventually do on-track adjustments myself. I actually will have weeks while at my cabin to read and research more fully the self-tuning aspect of this. Can't fish 24/7. Well... I guess I could, but I think I'd tire of it.
Thanks for your input. I'm hoping you can narrow my options for me on components or give me some pros and cons.

[Sgt.Gator]

Hi Jeff, great to see you here in the forum.
I have an 818R on order, due next December. In the meantime I'm forcing myself to learn the Subaru world by racing the original Subaru factory road race team Legacy GT Wagon across the Pacific NW. I just finished a race at PIR, nxt race at ORP July 25-27. Here's a thread about the car: http://legacygt.com/forums/showthread.php/cool-article-05-scca-wagons-floating-around-still-73859.html Cobb/Surgeline did a lot of the work and the tuning to get this car back on the track. And they are still helping me everytime I have a question. They are great folks there.

My question for you: Does opening a duct from a high pressure area such as thru a headlight or somewhere else in the front of the car and routing that cooler air to an SRI that is not sealed, but just the ducted air blowing on/around the filter, have beneficial effect? And what if the duct is actually sealed up all way so the higher pressure cooler air is actually forced thru the filter and into the intake tract past the MAF and on to the turbo inlet. My top speed at Spokane was 156 mph, so I expect there could be quite a bit of pressure on the air entering the intake tract.

Thanks and hope to see you at ORP or PIR. We are racing at PIR the same weekend as the Big NW Subaru Meet, August 10th, it should be fun to be racing a Legacy GT at the same time thousands of Subaru gearheads are out there. http://www.bignorthwest.com/ Will you be there? if so please come by the paddock and say hello!
31140

[DodgyTim]

You certainly could start off with a purchased USDM map and convert tables over, and it would help if you started with the closest map possible (same model year in the USDM market). Based on your modifications it looks like you are using the same stock injectors, better fuel pump, and a larger turbo. A setup that like can be tuned assuming you are willing to start slow. If you are not going to take to the car to a dyno, you need to make sure you have a good wideband installed, even if you get a basemap from someone. If you have a good setup to do logging including a wideband, a remote tuner can do a lot.

Since your fuel is different, I would have the basemap start with reasonable but conservative timing, low boost, and work on getting the AFRs in line. After that is well calibrated you can work on timing and boost. While you can do a lot with road tuning, it is difficult to really get a perfect calibration as you can't see how small changes effects torque... and of course a loading dyno makes that much better.

Jeff

Thanks Jeff, much appreciated

[sponaugle]

I think I want a standalone ECU. One thought is to put an EGT sensor on all cylinders. I believe standalones have the potential for many inputs?

I use EFI Logics in Bethel, CT for my dyno tuning. They are one of the best. I believe Mikey can tune using any major system. I am not opposed to working with someone to supply me with maps based on my data. OTOH, I do want to put the time into gaining the knowledge necessary to evaluate the data and eventually do on-track adjustments myself. I actually will have weeks while at my cabin to read and research more fully the self-tuning aspect of this. Can't fish 24/7. Well... I guess I could, but I think I'd tire of it.
Thanks for your input. I'm hoping you can narrow my options for me on components or give me some pros and cons.

Ok that helps to narrow things down a bit. If you are doing a standalone here are some options and pro/cons to consider. There are almost certainly other options out there that I have not personally worked with, and some of them might be a fantastic fit. Here are a few of the ones I would put on my list:

1: Motec - PRO: Very high quality, respected, lots and lots of feature, good software, lots of calibrators have experience with it, awesome logging. CON: Expensive.. really expensive and features cost even more. Race teams use this stuff for a reason.

2: Vipec - A rebranded Link G4 PRO: Has lots of motorsports features included, software is pretty good but not perfect, has good Subaru support, Maps are easy to build, Seems pretty bug free at least in my experience. A LOT cheaper than Motec.; CONS: Not as feature rich as the Motec, Software can be slow sometimes. Motec logging is better.

3: AEM EMS - PRO: Pretty inexpensive, support for plug and play Subaru. Software interface is pretty straightforward. CONS: In my experience so far, AEM has been the buggiest. Lots of little weird things even in different versions. Just doesn't feel as rock solid as it should.

4: Pectel - Like Motec in terms of features.. lots of them. Good professional support and software. Lots of analog inputs. CONS: Like Motec, expensive!

Others in the list: Autronic, Haltech, EFI, Gems/Omex.

Most if not all of these have a good number of analog inputs, which you can use for a lot of things. EGT, extra AFR channels, Air temp, etc. All support the AIM dash, which is a good dash. I think the three I like best would be the Motec, Vipec, and Pectel. The Vipec is certainly more aimed at a user to tune with, while the Motec and Pectel are a bit more professional. I suspect if you price things out the Vipec would be the cheapest as well. It might be worthwhile to check and see what EFI logics has to say. I knew Chris at EFI (RIP), and they are good guys and are Cobb Protuners as well. I think a MOTEC with all the features can hit 10k, while the Vipic would be more like 2.5k. The Pectel is in the middle around 5k... I know... expensive no matter how you look at it!

If your interested I can give you a list of some good books to read on calibration?

Jeff

[Hindsight]

Hi Jeff, I'm Jeff from Portland as well. I grew up in Portland and lived there until 2005. Sure do miss it. Great car scene there.

I've been working on a problem with another vehicle of mine for around a year and can't figure it out. Since you seem to have a really solid knowledge of EFI systems and components, I thought maybe I'd ask and see if you had any thoughts. I haven't been able to obtain help anywhere else.

The vehicle is a stock 1993 VW. It has a 5cylinder gas engine, naturally aspirated. The problem is that under acceleration, it isn't smooth and feels a lot like someone is turning on, and off the air conditioning every second or two (that feeling of losing and gaining 10hp or so). It doesn't feel or sound like a misfire (no sound associated with it at all). The heavier the load on the vehicle, the more pronounced the symptom is.

Here is the ONLY clue I have to go on so far and it's a pretty big one: If I unplug the o2 sensor, the issue goes away and the vehicle runs smooth as butter.

Vehicle is pre-OBD but does have limited scanning abilities. No codes found. Just passed emissions with flying colors.
It is a stock speed density system - just a MAP sensor, hall sensor, and air temp sensor.

These parts have been replaced within the last 9 months with OEM Bosch components: Hall sensor, cap, rotor, wires, plugs, coil, fuel injectors, o2 sensor, fuel pressure regulator.

I've tested nearly everything including wiring (grounds are all cleaned and tested, etc), fuel delivery rates and pressures, ignition timing (its set perfectly). I even have a spare ECU and swapped it in - no difference.

I feel like it can't be ignition related, or unplugging the o2 wouldn't fix the issue. So that leaves fuel system. I checked the pump, pressure, delivery. The o2 sensor and closed loop system is working: If I disconnect the vac line to the fuel pressure regulator, the o2 voltage pegs, then slowly comes back down and returns to a normal fast cycle between .2 and .8 volts.

No exhaust leaks (tested with engine cold by spraying exhaust manifold and down pipe with soapy water, then duct-taped shop-vac exhaust to the tail pipe), no vac leaks (smoke tested). I built a new o2 sensor wiring harness to ensure there wasn't a short in the harness wiring (didn't help).

The only other info I can add is that I ran a test by monitoring the o2 sensor voltage with it connected to the ECU, and disconnected from the ECU.
With the o2 connected to the ECU, the voltage cycled between 200 and 600mv until about 3800RPM when it went to 800mv and stayed there to redline.
With the o2 disconnected from the ECU, the voltage read only 30mv (yes 0.03v) with no change, until 3800RPM when it went to 800mv and stayed there to redline.

So this makes me assume that the ECU's in these vehicles go into open loop mode above 3800RPMs (regardless of throttle pedal position). I also did some WOT runs and found that the ECU doesn't go open loop during WOT.
The above makes it look like the ECU is running the van very lean in open loop mode, though with a narrowband o2 it's obviously impossible to know how lean, but you'd think open loop would run a little rich, if anything. So I'm asking myself, why does it run lean in open loop mode, and why does it only run well in open loop mode?

I bought an Innovate Wideband for the 818 project, but I intend to put it in the VW first to help get a better picture of what's happening with the A/F ratios in order to better pinpoint this issue.

Any input you could provide would be very much appreciated.

[Boog]

3: AEM EMS - PRO: Pretty inexpensive, support for plug and play Subaru. Software interface is pretty straightforward. CONS: In my experience so far, AEM has been the buggiest. Lots of little weird things even in different versions. Just doesn't feel as rock solid as it should.
Jeff

Do you mean AEM's stuff in general? or specifically the series 2 EMS or EMS-4? I keep seeing advertisements for their new "Infinity" line and was just curious how it compares to the other big names.

[sponaugle]

Do you mean AEM's stuff in general? or specifically the series 2 EMS or EMS-4? I keep seeing advertisements for their new "Infinity" line and was just curious how it compares to the other big names.

No, just the older Series 2 EMS. I have not used the new Infinity system, and it does have some very nice features. Since it is pretty new they are including all of the optional features for free. (for a limited time I suspect). The infinity line is clearly going after the top dogs. They have some nice engine protection features (sensor based), from what I have seen one of the better boost control methods (speed, gear, load, temp comp, and a few more modes), and fast logging. I love the AEM sensors, and use them on a few of my cars with other standalones. Since I have not used one, it is hard to comment on how well it performs.

I do like how they list injector on-time accuracy at 1/10 uS.. which just really represents the accuracy to which they calculate injector on-time inside the ECU. Your injector and your engine will not notice a variance of 1us, much less 1/10us. Of course as an engineer I can appreciate the nuance

Jeff.

[sponaugle]

Hi Jeff, great to see you here in the forum.
I have an 818R on order, due next December. In the meantime I'm forcing myself to learn the Subaru world by racing the original Subaru factory road race team Legacy GT Wagon across the Pacific NW. I just finished a race at PIR, nxt race at ORP July 25-27. Here's a thread about the car: http://legacygt.com/forums/showthread.php/cool-article-05-scca-wagons-floating-around-still-73859.html Cobb/Surgeline did a lot of the work and the tuning to get this car back on the track. And they are still helping me everytime I have a question. They are great folks there.

My question for you: Does opening a duct from a high pressure area such as thru a headlight or somewhere else in the front of the car and routing that cooler air to an SRI that is not sealed, but just the ducted air blowing on/around the filter, have beneficial effect? And what if the duct is actually sealed up all way so the higher pressure cooler air is actually forced thru the filter and into the intake tract past the MAF and on to the turbo inlet. My top speed at Spokane was 156 mph, so I expect there could be quite a bit of pressure on the air entering the intake tract.

Thanks and hope to see you at ORP or PIR. We are racing at PIR the same weekend as the Big NW Subaru Meet, August 10th, it should be fun to be racing a Legacy GT at the same time thousands of Subaru gearheads are out there. http://www.bignorthwest.com/ Will you be there? if so please come by the paddock and say hello!
31140

Nice to meet you. I will try to make it out that weekend, as it would be cool to see your car on the track. Congratulations on getting the 818. It is a very cool project to build, and there are so many different variables you can tweak. A dream for any engineer.

Those are two great questions, and something I have played a little with. The first question was about routing air (with some significant flow) to the area around the filter. I tried this many years ago on my Supercharged Acura Integra that I raced at PIR. I went as far as installing a temperature logging system to see if it made a difference. I had a duct from the front of the car that pushed air up into the area of the filter, which was a ‘shorty intake’ tube. My primary reason for doing the modification was to get some colder air around the intake as the filter location in the Integra is close to the exhaust header.

Interestingly enough, it made a bigger difference around town. I suspect the extra air flow at low speeds provided some cooling, and driving around town would typically include a lot of time stopped and idling. Without the tube installed the AITs took longer to fall. At the track however, it really did not make any difference. I suspect that once at speed there was enough airflow thru/over/and under the radiator to keep ambient air temperature near the filter at all times. I did not do much more investigation and I left the tube in place. It certainly did not hurt anything, and it did not seem to have any effect on the resulting tune. However keep in mind this was a speed density car, so no MAF sensor.

The second question has to do with putting the air feed from said pipe directly into the intake track. I did try this once on my WRX, and it had some weird effects. This was back in 2003 and I did not at that time have access to the factory ECU tuning tools we have now. What I remember is that it caused so very unusual MAF readings, especially at high speed. At the time I was guessing that with a significant amount of positive pressure coming into the intake that there would be some reversion around the MAF sensor. I’m not convinced that is the case now, but I would have to do some testing.

There is a great article in Sport Bike magazine where they tested a number of different ‘Ram Air’ intake systems. All of these systems did exactly as you described. They put a direct air feed from the front of the bike into the intake tract. The guys at the magazine measured the change in pressure at the intake point and recorded that change vs speed.

Here is a graph of one of the test:

http://www.sportrider.com/sites/sportrider.com/files/styles/medium_1x_/public/import/page_element_images/146-9910-ram-07-zoom.jpg

The black line is speed and the orange one is the pressure in the intake tract. This is a normally aspirated bike and the intake pressure would normally be very close to 0 mb. The spikes in pressure on the way up are from the engine pulling back a bit during shifts. You can see that speed makes a difference, and at 140mph the pressure was about 10mb which is about .145 psi. (Yes, point one four five, not one point four five). That is not a huge amount of pressure, especially compared to a turbocharged system.

To figure out if this makes a difference they ran the bikes on a dyno using a compressor to make the intake pressure the same. Sure enough there was a difference of something in the 5hp range.

So, how does this help you? Certainly in a turbocharged environment the reduced negative pressure in front of the compressor wheel makes a difference in reducing the pressure ratio, and that in turn increases the efficiency of the turbocharger. It is difficult to estimate how much it would help, as I would have to collect some data from your specific car and use a compressor map for your turbocharger. As with many things the best way would be to install pressure and temperature sensors pre and post turbo and measure the effect. My instinct says that while it will probably make a positive difference I would not expect it to be huge. It might move you to a more efficient part of the compressor map, but that isn’t going to give you 50hp.

There is one possible negative effect however that might need some investigation, and that is the effect on the MAF sensor. While the MAF sensor can work in a positive pressure environment it alters the calibration a bit. I'm not sure how consistent the change would be, or if there would be any turbulence effects. If you were to switch to speed density you would not have that problem. It might be fun to try out, but you would want to have some good logging equipment.

Interesting questions. I like it.

Cheers,

Jeff

[Sgt.Gator]

Thanks for the in depth reply. I've been thinking about this because virtually every turbo charged car from the factory now comes with an air intake located in the high pressure area just above and/or in front of the outside of the radiator. I've been studying race car aero and it's pretty interesting where the high pressure and low pressure areas are located, and often closer together than you would think. The engineers are clearly taking in colder air and slightly pressurized, or at least not in a negative pressure area. With an FMIC and engine radiator pretty well sealed up to the body, that leaves the typical aftermarket SRI in the hot zone, and depending on location a possible negative pressure zone.

I don't know if you follow Formula 1, but this year Mercedes is kicking ***. Part of their secret seems to be the way they split the air coming into their air scoop a couple of directions:
31174

Article: http://www.formula1.com/news/technical/2014/0/1179.html

And this longer article about air intakes in Racecar Engineering, however it has more to do with the NA V8 cars, not this year's turbo cars: "At the end of the straight at the Shanghai circuit where the car is travelling at close to 330kph with DRS open, the air pressure hitting the air filter will be close to 1,070 millibar – normal ambient air pressure is around 1,020 millibar. This means the engine produces around 5% more power – equivalent to 40 bhp – than at ‘walking speed’."
http://www.racecar-engineering.com/articles/f1/technology-explained-f1-airboxes-filters/

As for my car, I don't know if Lance tuned it with just the MAF, SD, or Hybrid. I hope Hybrid, but I'll have to find out. I've cut a hole thru my headlight and ducted a velocity stack on the lens to the Perrin air filter/intake. I'll see how that does on the track, I can tape the stack closed, log IATs, then open it up and log it to compare. I also have a pressure meter so I can record relative positive/negative air pressures all over the front of the car and inside the engine compartment. Step two will be trying to duct air directly into the intake, under pressure from the car moving, but I'll have to find out abouth the MAF/SD function first before taking a chance on that.

As for our 818s, I think the car is going to have problems with heat soak if we don't find a solution. The TMIC is not going to get enough air flow. Unless of course we solve all these issues with one of these in carbon fiber:
31175

I'm looking forward to meeting you. And Brandon says Hi!

[nuisance]

About 10 years ago I did a lot of work in this area. I was racing small airplanes, and of course looking for an advantage. I developed a ram air system that I licensed out and is still being sold.

The idea is that we are converting the kinetic energy of the incoming air to static pressure, to feed the intake. Works equally with N/A or turbo engines. The formula is 1/2 (air density) * (V)squared. Skipping out on the unit conversion, there is available a gain of .235 psi at 100 knots (115 mph). At 200 knots where I was flying the gain was .941 psi which was worthwhile. About the best anyone can do is to recover 80% of this energy, and that is where I finally finished after much experimentation with inlet shapes and expansion chambers. Note that an airplane has an intake duct about 12" long from inlet to throttle body.

In the literature I found that power produced is roughly linear to manifold pressure, so in the case of the airplane, we could get a 5% increase in power. Unfortunately, airspeed goes with the cube root of power, so I got about 1.6% on speed, or about 3 knots. Still, in a 100 mile race, it was a nice advantage. All the above numbers are at sea level by the way.

So for an 818, I am not sure chasing ram air is worth it. Especially at lower speeds, at 62 mph we would get 1/4 of the .235 psi. More rewarding though, if we can supply cooler air to the intake, we will see some real gains. The accepted number there (many refs) is a 1% gain in power for every 10 degrees F reduction in intake temp.

john

[Jaime]

Don't forget that the car will almost certainly be partially on bypass/blowoff at WOT, so any gains due to increased intake pressure will simply be vented. The discussion about efficiency still applies, but you won't see any increase in cylinder pressure due to the intake location. You might not even get the lower temperatures that you theoretically should unless you are able to use a smaller turbo.

Racing is all about getting the most performance while staying inside the boundaries that the rules set. On the power side, those rules are often designed to limit air flow. That makes race design all about "how do I produce the most power from this air stream" and it turns into a game of reducing losses and increasing charge density.

Most street tunes limit boost to the point where more boost would cause detonation. So, we put a boost target in the map and the ECU controls the bypass to hit the target. High ambient air pressure, low altitude, and ram air don't change the target.

[Boog]

My understanding may be way off here.
Isn't the comparison not between [0.25 psi] and [0 psi], but between [atmospheric + 0.25 psi] and [atmospheric - intake vacuum]? I always thought the benefit of ram air was to keep the engine from doing extra work pulling a vacuum in the intake tract.

[Jaime]

My understanding may be way off here.
Isn't the comparison not between [0.25 psi] and [0 psi], but between [atmospheric + 0.25 psi] and [atmospheric - intake vacuum]?
You might be right. I know a good bit about engine tuning theory, but I'm new to Subarus, so I could be wrong about how this system works. However, if Subaru does simply track boost compared to inlet pressure rather than something that approximates absolute charge pressure, then ram air could push the engine into a portion of the operational envelope where the tuner doesn't want to go.

Also, if that's true, then you could get almost all of the effects of ram air by simply dialing up the target boost number. The main difference would be that air compressed by the turbo would be a bit hotter than air compressed before it enters the intake tract. Cooling is important, but you could probably get better temperature reduction by blowing that air through the intercooler rather than into the inlet. That's why a WRX has a bigger intercooler scoop than its intake scoop.

I always thought the benefit of ram air was to keep the engine from doing extra work pulling a vacuum in the intake tract.Whether you think of it as "reduce ambient vacuum and make it easier for the engine to draw in air" or "stuff air in cylinder", the result is still increased cylinder pressure. Both models result in the same outcome, just the signs of certain terms go negative at different times.

[sponaugle]

Thanks for the in depth reply. I've been thinking about this because virtually every turbo charged car from the factory now comes with an air intake located in the high pressure area just above and/or in front of the outside of the radiator. I've been studying race car aero and it's pretty interesting where the high pressure and low pressure areas are located, and often closer together than you would think. The engineers are clearly taking in colder air and slightly pressurized, or at least not in a negative pressure area. With an FMIC and engine radiator pretty well sealed up to the body, that leaves the typical aftermarket SRI in the hot zone, and depending on location a possible negative pressure zone.

And this longer article about air intakes in Racecar Engineering, however it has more to do with the NA V8 cars, not this year's turbo cars: "At the end of the straight at the Shanghai circuit where the car is travelling at close to 330kph with DRS open, the air pressure hitting the air filter will be close to 1,070 millibar – normal ambient air pressure is around 1,020 millibar. This means the engine produces around 5% more power – equivalent to 40 bhp – than at ‘walking speed’."

As for my car, I don't know if Lance tuned it with just the MAF, SD, or Hybrid. I hope Hybrid, but I'll have to find out. I've cut a hole thru my headlight and ducted a velocity stack on the lens to the Perrin air filter/intake. I'll see how that does on the track, I can tape the stack closed, log IATs, then open it up and log it to compare. I also have a pressure meter so I can record relative positive/negative air pressures all over the front of the car and inside the engine compartment. Step two will be trying to duct air directly into the intake, under pressure from the car moving, but I'll have to find out abouth the MAF/SD function first before taking a chance on that.

As for our 818s, I think the car is going to have problems with heat soak if we don't find a solution. The TMIC is not going to get enough air flow. Unless of course we solve all these issues with one of these in carbon fiber:
I'm looking forward to meeting you. And Brandon says Hi!

That is an interesting approach (the F1 manifold). In the Racecar Engineering testing they were getting about 50mbs of above ambient pressure, which is in the range of what the bike guys were seeing at 180+mph. I would assume larger frontal collection area can make a big difference. Either way, that is still only something in the 1psi range.

I would suspect you car is tuned for SD, not hybrid. The hybrid works, but really is not needed. Both of my other Subarus run SD only. I don't even have a MAF sensor installed. If you have a pressure sensor installed in the intake tract close to the inlet of the turbocharger you should be able to get some good data. I mentioned in another thread that Lance did some testing of just changing an intake to see the difference in pre-turbo pressure:

http://forums.nasioc.com/forums/showthread.php?t=2643088

The interesting graph is:

https://cobb.box.com/shared/static/jm55ehb9zvqbch8r8nh5.png

This is done at wastegate pressure, without any active boost control feedback. The reduction in inlet restriction allowed the turbo to make more outlet pressure which did not exceed the wastegate release threshold. In this case, without any changes to the software or the boost control map the car made more power ( almost entirely because it was running more boost). There would also be some gain due to the reduced adiabatic heating because of a reduced pressure ratio, but in this case there really wasn't a huge change in PR (since the turbo produced more output pressure).

By this metric, if you were using wastegate boost control, I would expect there to be a power gain with a ram air intake system; However most of that power gain would be due to running more manifold pressure, and you would need all of the supporting details for that additional power (fuel, spark tuning, sensor range, etc).


About 10 years ago I did a lot of work in this area. I was racing small airplanes, and of course looking for an advantage. I developed a ram air system that I licensed out and is still being sold.
The idea is that we are converting the kinetic energy of the incoming air to static pressure, to feed the intake. Works equally with N/A or turbo engines. The formula is 1/2 (air density) * (V)squared. Skipping out on the unit conversion, there is available a gain of .235 psi at 100 knots (115 mph). At 200 knots where I was flying the gain was .941 psi which was worthwhile. About the best anyone can do is to recover 80% of this energy, and that is where I finally finished after much experimentation with inlet shapes and expansion chambers. Note that an airplane has an intake duct about 12" long from inlet to throttle body.

In the literature I found that power produced is roughly linear to manifold pressure, so in the case of the airplane, we could get a 5% increase in power. Unfortunately, airspeed goes with the cube root of power, so I got about 1.6% on speed, or about 3 knots. Still, in a 100 mile race, it was a nice advantage. All the above numbers are at sea level by the way.

So for an 818, I am not sure chasing ram air is worth it. Especially at lower speeds, at 62 mph we would get 1/4 of the .235 psi. More rewarding though, if we can supply cooler air to the intake, we will see some real gains. The accepted number there (many refs) is a 1% gain in power for every 10 degrees F reduction in intake temp.

john

I think your end statement sums it up. The gain would be small because the amount of pressure is not that large at car speeds. Keep in mind the references for change in power from air temp are typically based on NA behaviour. There are huge gains to be had with lower AITs in a turbocharged car, but a lot of it is programmatic ( more timing or more boost at lower AITs ) or intercooler based. Obviously denser air makes a huge difference, however careful data analysis will reveal that in many of those cases ( a cold morning drive ) the manifold pressure itself is higher and the intercooler efficiency higher. (which is a good thing!)



Most street tunes limit boost to the point where more boost would cause detonation. So, we put a boost target in the map and the ECU controls the bypass to hit the target. High ambient air pressure, low altitude, and ram air don't change the target.

I think in many situations this would be the theoretical outcome, however as Lance demonstrated it depends a lot on what the boost mapping is, and the limits of the turbocharger.

Take a look at this:

https://cobb.box.com/shared/static/p0uhzlk5c3qu2m2ebgmi.png

These were runs with the same boost control settings, yet different results came out. The primary reason is the boost control was set about the capability of the turbocharger in both cases, so the additional pressure range in the aftermarket intake case just allowed the turbo to get closer to the setpoint. If the setpoint had been 12 psi, and the wastegate curves adjusted to get the exact same boost on both cases, the power output would be similar ( but perhaps not the same due to the greater efficiency in the better intake case ). I actually asked Lance if he would consider doing that test.


My understanding may be way off here.
Isn't the comparison not between [0.25 psi] and [0 psi], but between [atmospheric + 0.25 psi] and [atmospheric - intake vacuum]? I always thought the benefit of ram air was to keep the engine from doing extra work pulling a vacuum in the intake tract.

That would be absolute (psia) vs relative or gauge(psig). Most of the references we are talking about here are all relative to atmospheric pressure. In the bike test 0mb was atm pressure, so the intake would normally be negative.


Jeff

[sponaugle]

You might be right. I know a good bit about engine tuning theory, but I'm new to Subarus, so I could be wrong about how this system works. However, if Subaru does simply track boost compared to inlet pressure rather than something that approximates absolute charge pressure, then ram air could push the engine into a portion of the operational envelope where the tuner doesn't want to go.

Also, if that's true, then you could get almost all of the effects of ram air by simply dialing up the target boost number. The main difference would be that air compressed by the turbo would be a bit hotter than air compressed before it enters the intake tract. Cooling is important, but you could probably get better temperature reduction by blowing that air through the intercooler rather than into the inlet. That's why a WRX has a bigger intercooler scoop than its intake scoop.
Whether you think of it as "reduce ambient vacuum and make it easier for the engine to draw in air" or "stuff air in cylinder", the result is still increased cylinder pressure. Both models result in the same outcome, just the signs of certain terms go negative at different times.

The subaru (and most car systems) run off absolute pressure sensors sometimes combined with a barometric sensor. Boost control is based off the primary absolute pressure sensor, and is not directly biased by the inlet pressure. If you have a good well calibrated boost control system and dial in 12psi, you should get 12psi regardless of the inlet pressure. In practice however this is not true. Most of the boost control algorithms are based of a starting waste gate duty cycle followed by a limited PID control algorithm (usually less I, more PD). While you would expect these to have good control they are usually overdamped to prevent boost spiking and sometimes will not maintain a precise output pressure with varying input conditions.

Equally true if your targets are higher then the turbo can reach (WGDC is >90%), things will change when inputs change.

In general, you can always make more power the fastest with more manifold pressure, assuming you have a turbocharger system that can produce the required airflow, fuel quality and quantity, and hardware that can handle the power. That would lead to your last conclusion that better intercooling would be a better way to spend your air budget, and I would agree. If you are running a 'stock' sized turbo at the very right edge of it's compressor map, invest in intercooling first. That is especially true in a hot racing environment.

Jeff

[waruaki]

Jeff,
Could you recommend a turner in the Portland or better yet the Salem area for a final dyno tune on our 818S. Our donor was a 02 WRX. Its had the codes deleted (TGV, secondary O2) and a updated map.

[Sgt.Gator]

Great discussion.

I did some more internet searching and found this video. It's an interesting test because the best and really only power gain occurs at the very end when they move the filter outside the car. I'm not sure if they were to re-tune the ecu if they might get even more. From what I can see they keep making changes to the intake but don't change the tune.



https://www.youtube.com/watch?v=gCi2yo4UqPI


Thanks for the replies. On a street car this would be meaningless, but when racing 6 Hp across the RPM range is a big deal. You can spend a lot of money to get 6 HP.

For our 818R I'd like to see a good CAI that actually draws in outside air and a good intercooler solution. I may have to go with a water/alcohol injection setup to cool things down. I ran one on my Mazadspeed 6 turbo setup and it worked great. Back in those days Christian Cobb actually did my tuning by emailing maps and datalogging. We got a good increase in power with nothing but a Mazdaspeed CAI and Water/Alcohol injection. I still have the system, I sold the Speed6 without it. A water/alcohol injection discussion should be it's own thread though, not trying to hijack this one!

[Sgt.Gator]

I had my tune done at Cobb Surgeline, the place Jeff started. I have nothing but good things to say about the crew there!

[sponaugle]

Hi Jeff, I'm Jeff from Portland as well. I grew up in Portland and lived there until 2005. Sure do miss it. Great car scene there.
I've been working on a problem with another vehicle of mine for around a year and can't figure it out. Since you seem to have a really solid knowledge of EFI systems and components, I thought maybe I'd ask and see if you had any thoughts. I haven't been able to obtain help anywhere else.
The vehicle is a stock 1993 VW. It has a 5cylinder gas engine, naturally aspirated. The problem is that under acceleration, it isn't smooth and feels a lot like someone is turning on, and off the air conditioning every second or two (that feeling of losing and gaining 10hp or so). It doesn't feel or sound like a misfire (no sound associated with it at all). The heavier the load on the vehicle, the more pronounced the symptom is.
Here is the ONLY clue I have to go on so far and it's a pretty big one: If I unplug the o2 sensor, the issue goes away and the vehicle runs smooth as butter.
Vehicle is pre-OBD but does have limited scanning abilities. No codes found. Just passed emissions with flying colors.
It is a stock speed density system - just a MAP sensor, hall sensor, and air temp sensor.

These parts have been replaced within the last 9 months with OEM Bosch components: Hall sensor, cap, rotor, wires, plugs, coil, fuel injectors, o2 sensor, fuel pressure regulator.
I've tested nearly everything including wiring (grounds are all cleaned and tested, etc), fuel delivery rates and pressures, ignition timing (its set perfectly). I even have a spare ECU and swapped it in - no difference.
I feel like it can't be ignition related, or unplugging the o2 wouldn't fix the issue. So that leaves fuel system. I checked the pump, pressure, delivery. The o2 sensor and closed loop system is working: If I disconnect the vac line to the fuel pressure regulator, the o2 voltage pegs, then slowly comes back down and returns to a normal fast cycle between .2 and .8 volts.
No exhaust leaks (tested with engine cold by spraying exhaust manifold and down pipe with soapy water, then duct-taped shop-vac exhaust to the tail pipe), no vac leaks (smoke tested). I built a new o2 sensor wiring harness to ensure there wasn't a short in the harness wiring (didn't help).

The only other info I can add is that I ran a test by monitoring the o2 sensor voltage with it connected to the ECU, and disconnected from the ECU.
With the o2 connected to the ECU, the voltage cycled between 200 and 600mv until about 3800RPM when it went to 800mv and stayed there to redline.
With the o2 disconnected from the ECU, the voltage read only 30mv (yes 0.03v) with no change, until 3800RPM when it went to 800mv and stayed there to redline.

So this makes me assume that the ECU's in these vehicles go into open loop mode above 3800RPMs (regardless of throttle pedal position). I also did some WOT runs and found that the ECU doesn't go open loop during WOT.
The above makes it look like the ECU is running the van very lean in open loop mode, though with a narrowband o2 it's obviously impossible to know how lean, but you'd think open loop would run a little rich, if anything. So I'm asking myself, why does it run lean in open loop mode, and why does it only run well in open loop mode?

I bought an Innovate Wideband for the 818 project, but I intend to put it in the VW first to help get a better picture of what's happening with the A/F ratios in order to better pinpoint this issue.

Any input you could provide would be very much appreciated.

Well, I would probably start at the same place; getting a wideband installed. With the wideband you can see how the fueling is changing as it switched modes. The data you collected from having the sensor unplugged is interesting, as it would suggest that there is some internal behaviour that occurs at 3800 rpm. Narrowband sensors are very inaccurate outside the single transition at stoichiometric, so the actual values once rich or lean are hard to make a determination from.

One thing you didn't mention is what RPM does the surging start at? Before or after the 3800 RPM open loop transition. It seems that narrow band readings are similar in both the connected and unconnected states above 3800 rpm, but a wideband can help confirm that. It is possible that in one configuration the mixture is much richer. I assume this O2 is before a cat?

You could also measure timing (using an oscilliscope on the ECU output) to see if that is changing as part of the O2 feedback.

[sponaugle]

Jeff,
Could you recommend a turner in the Portland or better yet the Salem area for a final dyno tune on our 818S. Our donor was a 02 WRX. Its had the codes deleted (TGV, secondary O2) and a updated map.

As Sgt. Gator mentioned, Cobb Tuning has a facility here in Portland (Cobb-Surgeline). They are on 72nd Ave in Tigard/Portland right off I5 and they have a Mustang AWD. Either Lance or Tim can take care of you. There are of course other calibrators around; PRE, PSI, and certainly others. One question is are you planning on using an Accessport, or the Opensource tuning tools?


Great discussion.
I did some more internet searching and found this video. It's an interesting test because the best and really only power gain occurs at the very end when they move the filter outside the car. I'm not sure if they were to re-tune the ecu if they might get even more. From what I can see they keep making changes to the intake but don't change the tune.
Thanks for the replies. On a street car this would be meaningless, but when racing 6 Hp across the RPM range is a big deal. You can spend a lot of money to get 6 HP.
For our 818R I'd like to see a good CAI that actually draws in outside air and a good intercooler solution. I may have to go with a water/alcohol injection setup to cool things down. I ran one on my Mazadspeed 6 turbo setup and it worked great. Back in those days Christian Cobb actually did my tuning by emailing maps and datalogging. We got a good increase in power with nothing but a Mazdaspeed CAI and Water/Alcohol injection. I still have the system, I sold the Speed6 without it. A water/alcohol injection discussion should be it's own thread though, not trying to hijack this one!

I am familiar with the video and those guys. They have done a number of videos on different car related myths. Unfortunately they tend to skip over any analysis of cause and effect. It might be true that in the particular case they tested there were no power gains, but that is of course only part of the story. As Lance from Cobb demonstrated in the 15 STI, just the single intake change made a surprising difference in power. Of course that same power gain could have been made with tuning alone. None the less it is always interesting to see real world testing. If they would only dig a bit deeper into the ECU operation and resulting outputs during these tests they would really have something.

I too have used methanol and water injection. I made a post about the installation on my GTO here: http://www.ls1gto.com/forums/showthread.php?t=385931. Jeff Perrin made a great write up about his H6 STI at Sema that had a very trick Aquamist system: http://blog.perrinperformance.com/perrin-h6-build-up/. [and as fate has it, that exact car is my donor car for my 818] I was a dealer for Aquamist for a few years, so I know their stuff pretty well. Indeed you can do a lot with methanol injection, however after years of doing it myself and with customers, I have generally moved away from it. Most people are just not diligent enough to keep the system working, and unless you do it just right a system failure can be bad news. The aquamist stuff is very good on this front, but it does require some maintenance. I prefer to just use better fuel if I can. I run 4 of my cars 100% on E85, and that gives me a lot of headroom without the failure potential. If you are willing to make sure to keep thing working however, it can make a big difference.

Cheers,

Jeff

[Hindsight]

Well, I would probably start at the same place; getting a wideband installed. With the wideband you can see how the fueling is changing as it switched modes. The data you collected from having the sensor unplugged is interesting, as it would suggest that there is some internal behaviour that occurs at 3800 rpm. Narrowband sensors are very inaccurate outside the single transition at stoichiometric, so the actual values once rich or lean are hard to make a determination from.

One thing you didn't mention is what RPM does the surging start at? Before or after the 3800 RPM open loop transition. It seems that narrow band readings are similar in both the connected and unconnected states above 3800 rpm, but a wideband can help confirm that. It is possible that in one configuration the mixture is much richer. I assume this O2 is before a cat?

You could also measure timing (using an oscilliscope on the ECU output) to see if that is changing as part of the O2 feedback.

Jeff, thanks for the reply. The surging starts from 2000 and continues until 3800. So basically it seems that any time the ECU is running in closed loop mode, the hesitation occurs. Force it to open loop, or let it get to open loop on it's own by going over 3800RPMs and the issue goes away. It's so strange.

I don't have an oscilliscope to measure the timing, but I do have a VCDS (which emulates a VW-factory scan tool) which allows me to check timing while driving and according to it, nothing odd is happening.

Thanks again; I'll see what the wideband says. The only issue there is that if the wideband shows that the engine is lean in closed loop, or even rich for that matter, I'm not sure what I can do about it. No tuning devices or software are available for this ECU that I'm aware of. My only thought was to maybe put a potentiometer on the intake air temp sensor and try to use that to alter the IAT signal to fool the ECU into adding or removing a little fuel.

Oh and yes, the o2 is pre-cat.

[sponaugle]

Thanks again; I'll see what the wideband says. The only issue there is that if the wideband shows that the engine is lean in closed loop, or even rich for that matter, I'm not sure what I can do about it. No tuning devices or software are available for this ECU that I'm aware of. My only thought was to maybe put a potentiometer on the intake air temp sensor and try to use that to alter the IAT signal to fool the ECU into adding or removing a little fuel.
.

That is exactly what I was thinking. There are a couple of alternatives to consider:

(1) you could make a circuit that leans out the narrowband sensor sooner/a bit more..., although it sounds like the narrow band is already reading very lean (30mv).

(2) you could build a circuit to make the IAT colder either all the time, or up until the 3800 rpm cut off. You are right that a trick like that will usually make the car run richer.

Also, have you checked fuel pressure, and made sure the injectors are clean? If you unplug to vacuum reference to the fuel pressure regulator does it make a difference (should be a bit richer)

Can you see STFT from that tool? Also can you see what the ECU sees for IAT and Coolant temp?

Jeff

[waruaki]

One question is are you planning on using an Accessport, or the Opensource tuning tools?

At this point personally neither as I don't have the skills for the open source and want a more tailored map for our car than what I believe the Accessport will give. When we were ready for our fist start a few weeks ago we paid to have someone come and delete the permanent CEL codes caused by our modifications. Now I need to find a shop for the tune.

[sponaugle]

At this point personally neither as I don't have the skills for the open source and want a more tailored map for our car than what I believe the Accessport will give. When we were ready for our fist start a few weeks ago we paid to have someone come and delete the permanent CEL codes caused by our modifications. Now I need to find a shop for the tune.

Hmm. Your post puzzles me. If you had someone 'delete the CEL codes' that is done by using one of the three major tuning/flashing systems out there. I would suspect that the person you asked used the open source ECUFlash tools plus something like RomRaider or EcuEdit to disable the internal DTC checks. There is no real conceptual difference between disabling the CEL codes and 'tuning', at least from a 'what you are doing to the FLASH memory in the ECU.

I'm not sure what you are referring to about when you said 'more then what I believe the Accessport will give'. Perhaps you are thinking about the OTS maps provided free with the AP?

There are three major options when it comes to 'tuning' your Subaru WRX ECU. You can use the open sources tools, the Cobb Protools, or the EcuTek Tools. The open source tools have the advantage that they are free, however in most cases they are simply map editors and do not make significant 'code changes' in the ECU logic. This can work fine for many cases, but it leaves you lacking on a few fronts.

The Accesport/Cobb ProTuner tools are somewhat different. The Cobb system has significant code changes to the actual code running in the ECU. This enables a lot of advanced features such as Launch Control, Real Time Tuning, Speed Density, etc. As a customer using an Accessport you have a few options. You can have your car 'protuned' at a Cobb dealer (such as Cobb Portland). When this happens you car will be tuned on a dyno, and the resulting tunes (often 3 or 4 of them) will be stored on your Accessport. You can then load these maps as you need them. For example you could have a map for each different fuel type, etc.

Since the Accessport has significant code changes it can also support running your car in Speed Density which eliminates the MAF sensor. This is a fantastic upgrade and I do it on two of my Subarus. Of course flat foot shifting and launch control are just extra goodness. In terms of the Factory ECU the Accessport+Pro Tuner can do pretty much anything from a 250hp wrx to a 1000whp drag race machine. You can also purchase Street Tuner for the Accessport which gives you a copy of the tuning software for your particular car.. this software is virtually the same as the Protune software.

The EcuTek solution is similar except there is not standalone module, just a dongle for your computer that you use to flash, and EcuTek doesn't sell a user tunable version for your car. (no StreetTuner equivilent) EcuTek like Cobb makes actual code changes to enable features like launch control, etc. I have not used the latest EcuTek stuff so there are probably other features I'm not mentioning.

From a tuning perspective, one major advantage of the Cobb stuff is the realtime tuning. You can makes changes to many of the critical maps in real time without having to relfash the ECU. That saves a lot of time on the dyno, and can be useful if you are doing some kind of street tuning as well.

Lots of options... but really only three. ;)

Jeff

[waruaki]

He used a laptop with a usb to obd adapter. I was under the impression the Acessport had preprogrammed maps that you picked from depending on the modification done to your motor? As I mentioned this is an area that I need to become more familiar with. Thanks for the clarification.

[sponaugle]

He used a laptop with a usb to obd adapter. I was under the impression the Acessport had preprogrammed maps that you picked from depending on the modification done to your motor? As I mentioned this is an area that I need to become more familiar with. Thanks for the clarification.

Ah yes, he almost certainly used the opensource tools with a Tatrix cable (OBD to USB). My apologies if my post seems cross. (unintended!)

You are correct that if you purchase an Accessport it does come with a selection of 'preprogrammed' maps for a few common modification levels, but most people would typically not use those but instead have a custom map created on the dyno. As an added benefit once you get that custom map made, you can use the StreetTuner software to edit the maps further if you find things are not just right. This is very helpful if over time you find a few problem areas. As well you can send the map off to someone to edit and send back to you.

Jeff

[waruaki]

My apologies if my post seems cross. (unintended!)

None taken. I appreciate the education.

[Evan78]

You are correct that if you purchase an Accessport it does come with a selection of 'preprogrammed' maps for a few common modification levels, but most people would typically not use those but instead have a custom map created on the dyno. As an added benefit once you get that custom map made, you can use the StreetTuner software to edit the maps further if you find things are not just right. This is very helpful if over time you find a few problem areas. As well you can send the map off to someone to edit and send back to you.

JeffI'd like to add that the Cobb Protuner software has the option to lock the maps so that StreetTuner users can't edit them. This is a point to discuss before paying for pro tuning services if you want the ability to modify the maps yourself later. I haven't followed developments for the last several years, but this was the case at one point. Please correct me if I'm wrong.

[Sgt.Gator]

I'd like to add that the Cobb Protuner software has the option to lock the maps so that StreetTuner users can't edit them. This is a point to discuss before paying for pro tuning services if you want the ability to modify the maps yourself later. I haven't followed developments for the last several years, but this was the case at one point. Please correct me if I'm wrong.

I discovered this. My maps are locked. I downloaded Accesstuner Race which is free if you purchase an Accessport, but was unable to open the custom maps Cobb Surgeline created. I will be discussing that with them....

There are two tuning software packages from Cobb, Accesstuner Race and Accesstuner Pro. Pro is only available to professional tuning shops. Race anyone can get , but they lock it so it will ONLY work with your Accessport and no one else's. So you can't use Race to tune someone else's car.

Lots of info here: http://www.cobbtuning.com/Accesstuner-s/70463.htm

[Hindsight]

Also, have you checked fuel pressure, and made sure the injectors are clean? If you unplug to vacuum reference to the fuel pressure regulator does it make a difference (should be a bit richer)
Can you see STFT from that tool? Also can you see what the ECU sees for IAT and Coolant temp?


Yes, fuel pressure and delivery are good and injectors are brand new Bosch stock (same part number as the ones I removed). When I unplug vac to the FPR at idle and watch the o2 signal, it quickly goes up to 800mv and stays there for 5 seconds or so, then comes back down and starts cycling again. So it tells me the ECU/o2 circuit is working. FPR is new as well. I can't see STFT with the tool, but I can see IAT and ECT from it yes. The values for the two look correct and when the car has sat overnight, the value for both is exactly the same so they do seem to work properly.

It might be a couple of weeks before I can get to it but I'll report back after I've had a chance to weld the bung onto my downpipe for the wideband and then do some 4th gear pulls to see what's really happening with the mixture.

Thanks again for your help!

Jeff

[sponaugle]

I discovered this. My maps are locked. I downloaded Accesstuner Race which is free if you purchase an Accessport, but was unable to open the custom maps Cobb Surgeline created. I will be discussing that with them....

There are two tuning software packages from Cobb, Accesstuner Race and Accesstuner Pro. Pro is only available to professional tuning shops. Race anyone can get , but they lock it so it will ONLY work with your Accessport and no one else's. So you can't use Race to tune someone else's car.

Lots of info here: http://www.cobbtuning.com/Accesstuner-s/70463.htm

Yep, by default when maps are saved in the Pro software, they are Pro only maps. It takes a different click to save them as racetuner maps. If you contact Cobb they will convert the maps for you without a problem. I (and I suspect Cobb) is surprised at how few people use AcccessTuner Race. I have always assumed that people would want to learn about the changes in the ECU and make tweaks.. but no so much!

And Evan78 you are correct... best to tell the Protuner first that you want racetuner maps. If a tuner says no to that, you should run, not walk away. Tuning is a lot like Crypto algorythms, if someone says they have a secret method they invented themselves be very afraid. :)

Jeff

[Wayne Presley]

And Evan78 you are correct... best to tell the Protuner first that you want racetuner maps. If a tuner says no to that, you should run, not walk away. Tuning is a lot like Crypto algorythms, if someone says they have a secret method they invented themselves be very afraid. :)

Jeff

So true, you mean physics doesn't change depending on the tuner?

[Goldwing]

Hi Jeff,

I wanted to ask about the basic safety of the stock tune with the changes we make to install the wrx engine in the 818 chassis. I am planning to get a COBB Accessport and have it tuned at a local shop (probably Smokey's in Akron or ProTuning in Toledo, OH). Until I get that far, with a COBB intake tube wearing a K&N, bigger intercooler unknown brand, stock injectors, bigger fuel pump in Boyd tank, and a modified COBB downpipe, on an 07 wrx engine is the stock tune safe to drive (lightly or aggressively)? More on the downpipe: I put the cat right after the bellmouth discarding the rest, then fitted the exhaust through a 2.5" donut to a C6 Corvette muffler. I obviously don't want to damage anything before a proper tune can be installed. What are your thoughts on the safety of the stock tune, or commonly available COBB tunes to keep me safe until I can get the pro tune?

I'm also considering a VF48 turbo, looking to add power without sacrificing the bottom end where I think this car will shine the most. I throw that in there to ask the same question in that setup.

It seems a lot of guys are doing a similar setup, so this will likely grab the attention of others too. I appreciate you starting this thread, and the guidance you are providing. Thank you.

[AZPete]

Good question, Rich. I'm following this, too.

[305mouse]

There are a bunch of different maps you can load onto the ECU through your AP. Without a turbo, just the normal stage 1 map should be fine, stay out of boost. Changing turbos will have you looking at different maps than stage 1. It's been a while since I looked into it. I had a Cobb AP Ver 1, not sure which Version you have. Just make sure you try to stay out of boost before your tune.

[Scargo]

Agree with above. Also, you will leave a lot on the table and use too much gas if you go with canned COBB maps. Not that they're bad, but they are too conservative. I would get a proper dyno-tune if you do much of anything to the motor or learn to tune (using COBB's Street Tuner Race or similar) if you have that geeky side.

[Goldwing]

Just make sure you try to stay out of boost before your tune.

My wife is going to have to hold the keys for ransom.

[EODTech87]

Goldwing

Just as a heads up the VF48 interferes with the rear shock tower brace. You'll need to either modify the brace or switch to a external wastegate.

[Evan78]

I'm sure Jeff will chime in eventually, but I'm going to say the "Stage2 ACN91 Octane LWG" map (from the Cobb Map Database) (http://www.cobbtuning.com/Accessport-Off-The-Shelf-Maps-s/40100.htm#/v/accessport/maps/subaru/wrx/WRX-Maps-07-Stock-SF-Intake) is most appropriate off the shelf map for your situation Goldwing.

[wleehendrick]

Thats the map Im planning to start with, but Im not using a cat, and the Cobb website warns of boost creep in this case. Would the Stage 1 be 'safe'? Itll obviously leave a lot on the table, though.

[Goldwing]

Goldwing

Just as a heads up the VF48 interferes with the rear shock tower brace. You'll need to either modify the brace or switch to a external wastegate.

Thanks, I did see that. I'm working on my welding skills, lol. At this point I'm pretty good at blowing holes. :)

[Goldwing]

I'm sure Jeff will chime in eventually, but I'm going to say the "Stage2 ACN91 Octane LWG" map (from the Cobb Map Database) (http://www.cobbtuning.com/Accessport-Off-The-Shelf-Maps-s/40100.htm#/v/accessport/maps/subaru/wrx/WRX-Maps-07-Stock-SF-Intake) is most appropriate off the shelf map for your situation Goldwing.

Thanks! I will save that little gem and get to ordering my cobb AP. That's stock turbo, right?

[Evan78]

Yes, stock turbo. On the page (http://www.cobbtuning.com/Accessport-Off-The-Shelf-Maps-s/40100.htm#/v/accessport/maps/subaru/wrx/WRX-Maps-07-Stock-SF-Intake) that I linked to, click the links that say "[Manual Trans. Map Notes]" to access a PDF with details on what setups the maps are applicable to.

The 91 octane maps run less timing than the 93 maps (and maybe a little richer?). The ACN91 (Arizona, California, Nevada) maps are even more conservative. The LWG (low wastegate) variations are going to have a more conservative wastegate duty cycle (i.e. WG opens earlier). Your fuel pump is probably going to result in higher than oem fuel pressure, causing you to run richer than you would on the oem pump. I'm guessing all those added together should make things plenty safe for you to casually cruise to your shop for a good tune.

Wleehendrick - I'm guessing that a Stage 1 map may well be less safe than a Stage 2 map for your setup. Tons of people run catless without boost creep, although the much shorter exhaust might increase the chances in the 818. My guess is that Cobb is just being conservative. Regardless, you could run the LWG map and keep an eye on the boost gauge. I don't know what the maps are actually set at, but it seems that the off the shelf maps only produce full boost at full throttle. Backing off the throttle even a little seems to result in several PSI drop. Do you know who/where you plan to go to for a custom tune?

It'd be interesting to hear if Jeff has any recommendations for those of us in Southern California.

[Goldwing]

Ah yes, I got a chance to dig through there this morning. It makes a whole lot more sense now. Thanks for the link!

[wleehendrick]

Wleehendrick - I'm guessing that a Stage 1 map may well be less safe than a Stage 2 map for your setup. Tons of people run catless without boost creep, although the much shorter exhaust might increase the chances in the 818. My guess is that Cobb is just being conservative. Regardless, you could run the LWG map and keep an eye on the boost gauge. I don't know what the maps are actually set at, but it seems that the off the shelf maps only produce full boost at full throttle. Backing off the throttle even a little seems to result in several PSI drop. Do you know who/where you plan to go to for a custom tune?

It'd be interesting to hear if Jeff has any recommendations for those of us in Southern California.

Thanks for the info. My plan is to start with a stage 1 ot 2 OTS ACN91 tune and keep an eye on boost via the AP. Crawford Performane is close by, so Im thinking of getting tuned there.

[sponaugle]

Hi Jeff,

I wanted to ask about the basic safety of the stock tune with the changes we make to install the wrx engine in the 818 chassis. I am planning to get a COBB Accessport and have it tuned at a local shop (probably Smokey's in Akron or ProTuning in Toledo, OH). Until I get that far, with a COBB intake tube wearing a K&N, bigger intercooler unknown brand, stock injectors, bigger fuel pump in Boyd tank, and a modified COBB downpipe, on an 07 wrx engine is the stock tune safe to drive (lightly or aggressively)? More on the downpipe: I put the cat right after the bellmouth discarding the rest, then fitted the exhaust through a 2.5" donut to a C6 Corvette muffler. I obviously don't want to damage anything before a proper tune can be installed. What are your thoughts on the safety of the stock tune, or commonly available COBB tunes to keep me safe until I can get the pro tune?

I'm also considering a VF48 turbo, looking to add power without sacrificing the bottom end where I think this car will shine the most. I throw that in there to ask the same question in that setup.

It seems a lot of guys are doing a similar setup, so this will likely grab the attention of others too. I appreciate you starting this thread, and the guidance you are providing. Thank you.

Great question. To give you a direct answer first, I suspect the closest map will be the one Evan mentioned below. Starting with a Stage 2 map will be closer in terms of boost control, and the 91 octane map is pretty conservative. (for the stock turbo).

However… In general I would not expect any of the off the shelf maps to very close to spot on. Not only is there a significant difference in the exhaust path with the 818, but a significant difference in engine load due to the 2WD and lighter weight. The air intake path isn’t exactly the same, and the top mount intercooler will be working very differently (and sadly not as well) in the 818 as compared to a WRX. Given that you are building a kit car, you are clearly ok with getting your hands dirty. That is good, and that means you can do a lot before you even get to a protuner.

The question of safety is an interesting one. The factory ECUs have adequate knock detection (except stock 02-3 WRX), but are unfortunately lacking in the case of active full throttle AFR feedback. The factory AFR and O2 sensors are used for closed loop operation during cruise, but once you get into boost everything is open loop. (there is a bit more to this actually) While knock detection can (and will) save you in many cases, running high boost with a subpar fuel and very lean AFRs can do damage faster than the knock sensors can correct for.

Here is what I would recommend, and really this is for everyone doing a factory ECU on the 818.

(1) Make sure you install a wideband AFR sensor in the downpipe, after the turbo, and before the cat if possible.
(2) Make sure the factory knock sensor is installed on the block and wired up to the ECU.
(3) Make sure you have a good fast readable boost gauge mounted somewhere easy to see while you are driving.

If you have those three things when you start out, here is what I would do next.

(1) Download the Cobb RaceTuner for your AP. It is free and requires only you enter the AP serial number. You will need this to edit the maps as well as datalog.
(2) On step 1 above I mentioned installing a Wideband AFR sensor. Make sure it is one that is supported in RaceTuner. I would recommend either the new PLX 4.9 setup (http://www.plxdevices.com/product_info.php?id=WDBDSMAFR_DM6G4) or the Innovative Motorsports LM-1/LC-1. There are other supported ones as well.. just pick one.. Connect said wideband up to your laptop along with the AP itself.
(3) Pick a starting map, say the map mentioned above. You will almost certainly need to turn off some engine codes due to some things not being installed and that is easy to do in the Cobb software. Once you have a map that flashes on and starts the car without any CELs, you are ready to do some basic tuning.

At this point there are a few steps I would take to make sure everything is working as you expect. I can go into more detail on these as needed, and if helpful I could even do a webinar on the basics.

(1) Using the RaceTuner connected to the car thru the AP use the data monitoring to look at a few keys variables. Start with Coolant temp, air temp, MAF, IAT, MAP, the wideband AFR sensor, DAM, Knock Learning , AF Learning #1, AF Correction #1, RPM, TPS, Ignition Advance.
a. While the car is idling, you first need to make sure the core sensors are working. Coolant should be rising as temps come up, and eventually the fans should kick on. MAF should be reading something 2-4g/sec at idle. It should move if you rev the engine. MAP should be ~-9-10psi, and AF Correction should be bouncing around (once the car is warm) perhaps as much as +/-10%, and eventually stabilize to a smaller fluctuation. AFRs should stabilize to ~14.5-14.7:1 via the AFR meter.
b. The goal in the above step is to just make sure everything is working. Do the numbers look right.

(2) I would do the same steps above after you have gone thru a few heat cycles and drive the car on the road (or driveway even) at light throttle with no boost. If this is not a new motor that you are breaking in I would do some light throttle runs to 80% of redline, datalogging the entire way.
a. For each step, the goal is the datalog, then study the data log. For the light throttle no boost driving I would be looking mostly at the AF Correction to see how much correction has to be applied. If there are spots that have consistent and lasting 20+% corrections, something is off. You should also see the DAM climb…. But that depends on the car. The older 02-05 WRXs have a DAM that starts at 8 and goes to 16, and the newer ones, if I remember correctly, start at 0.5 and climb to 1.0. The purpose of the DAM is to represent a global ‘knocking index’. It starts out in the middle and after a bit of driving it will go up, and if you get knock it will go down. This number is used in the ignition timing calculations to provide an overall global correction. The more of this I am writing the more I can see this would be easier to explain in a webinar or in person!
(3) For each phase forward, I would do the same path.. do some driving with a new set of parameters, datalog, then review the datalogs. You can even post the datalogs here and we can all help out.
i. For the first time getting into boost, I would set the wastegate duty cycle table to zero so the car runs wastegate boost. That should keep the boost lower as well as highlight any boost creep issues you have. You will be looking at AFRs, Timing, Boost, and Knock retard for each of these passes. If you are absolutely going to get a pro tune you are just looking to make things safe.. however you can get a very close tune with driving on the road if you are always datalogging and making small tweaks.

I admit this does require a basic understanding of how the ECU maps translate to operation in the ECM, but it isn’t actually that complicated. I suspect the basics could be conveyed in an hour or two. Keep in mind I am talking about making small tweaks to an already close map, not building a new map for an advanced or very non-standard configuration. As always the best advice is to have a professional tuner do the tuning, however even in that case there is so much value in learning the basics of engine calibration.


Agree with above. Also, you will leave a lot on the table and use too much gas if you go with canned COBB maps. Not that they're bad, but they are too conservative. I would get a proper dyno-tune if you do much of anything to the motor or learn to tune (using COBB's Street Tuner Race or similar) if you have that geeky side.

Yes, most of those OTS maps are conservative, but keep in mind things can go from conservative to very aggressive due to a simple small change like running the MAF in a larger intake pipe. No matter how conservative the tune starts out, an unplanned change in inputs can cause big problems.


I'm sure Jeff will chime in eventually, but I'm going to say the "Stage2 ACN91 Octane LWG" map (from the Cobb Map Database) (http://www.cobbtuning.com/Accessport-Off-The-Shelf-Maps-s/40100.htm#/v/accessport/maps/subaru/wrx/WRX-Maps-07-Stock-SF-Intake) is most appropriate off the shelf map for your situation Goldwing.

Yep, that is a great place to start. The lower octane maps are pretty rich, but that is easy to tune around.


Thats the map Im planning to start with, but Im not using a cat, and the Cobb website warns of boost creep in this case. Would the Stage 1 be 'safe'? Itll obviously leave a lot on the table, though.

I would stick to the stage 2 maps to start with, for the single reason that the stage 2 maps usually have less wastegate duty cycle due to the more free flowing exhaust. No matter what map, I would first run a map with zero wastegate duty cycle. That will allow you to see if you have boost creep, and you can see the basis state of the tune.


Jeff

[sponaugle]

Your fuel pump is probably going to result in higher than oem fuel pressure, causing you to run richer than you would on the oem pump. I'm guessing all those added together should make things plenty safe for you to casually cruise to your shop for a good tune.


Assuming you have the fuel return and the stock regulator hooked up correctly, you really should not see much difference in fuel pressure at higher rpms on boost. It is possible that overdriving the regulator at lower throttle and at idle can cause higher than standard fuel pressure.. but if that is the case, get a better/larger fuel regulator. This would be a different story if say you were running twin Aeromotive 340s or the like… with that kind of setup the stock regulator would indeed result in higher fuel pressure at idle and low load and perhaps even moderate load depending on your injector flow.


Thanks for the info. My plan is to start with a stage 1 ot 2 OTS ACN91 tune and keep an eye on boost via the AP. Crawford Performane is close by, so Im thinking of getting tuned there.

Yep, they can certainly do a good tune there. I think there is also a Cobb shop in LA if I remember correctly…. And I am sure there must be a ton of other places around there. I’ll ask Tim (he is the head of calibration at Cobb) what other places he would recommend.

One other note: Everything I said above could equally apply to someone using the open source tools. As long as you have a tool that exposes all of the critical maps and supports good datalogging you are good to go.

[wleehendrick]

Yep, they can certainly do a good tune there. I think there is also a Cobb shop in LA if I remember correctly…. And I am sure there must be a ton of other places around there. I’ll ask Tim (he is the head of calibration at Cobb) what other places he would recommend.

Thanks for all the info, Jeff! Any other recommendations for Cobb shops in San Diego are would be appreciated.

I have a Cobb SF intake, and I understand that's pretty consistent, mixture wise, with the stock intake and compatible with the OTS maps. Besides the standard intake mods (silicone inlet and IC hoses, TGV deletes) I have a bellmouth to 3" catless exhaust with a Flowmaster on my EJ255. I'm a Subie-newbie, but the reading I've done says (besides turning off ECU codes for the missing emissions equipment) it's the boost creep that I need to be careful about until getting a custom tune. Anything else about my set-up that would require attention? I have a new APv3 ready to go, but a lot of wiring work to do before I can use it!

[Goldwing]

Jeff, thank you for taking the time to write such a detailed response. This will give me a lot to chew on for a while. I'll add the wide and sensor to the shopping list and report back after I've followed your suggestions. I'm also waiting to get my exhaust welded up. Don't want to contaminate my clean joints running it now. Again, thanks for your time!