RomRaider

Open Source ECU Tools
 FAQ •  Register •  Login 

RomRaider

Documentation

Community

Developers

It is currently Sat Oct 25, 2014 2:09 pm

All times are UTC - 5 hours [ DST ]





Post new topic Reply to topic  [ 36 posts ]  Go to page 1, 2, 3  Next
Author Message
 Post subject: Speed Density on OEM 05 STI ECU :)
PostPosted: Sat Jun 23, 2007 8:51 pm 
Offline
Experienced

Joined: Thu Oct 19, 2006 5:21 pm
Posts: 134
Location: Roy, WA
There is a discussion, maybe in the development area about having the ECU rewritten for Speed Density. Well, you don't need to go that far. The OEM ECU already has speed density built into it as a failsafe, the trick is to use that system for the total operation of the car. :)

A friend of mine brought me his car with several issues, bad idle, tip-in hesitation, hesitation through out the powerband, and was unhappy with the power (330whp) he was getting from his GT30R 05 STI engine swap.

I attempted to fix all these problems and was successful other than the tip-in hesitation. At the time I was only touching up another tuners tune, but the original tuner tuned the car for modded injectors with only the tip-in and MAF scale which caused most of the problems. Other problems were caused by the Perrin Blow-thru MAF, dips and spikes in the voltage were not good for me or the car.

Mike got some information from the ECUTek forums on how to switch the car to speed density 8) By pulling only the center pin out of the MAF connector, you still get a Intake Air temperature reference that the ECU uses for correction. Since this car was already blow-thru, this worked out perfectly.

Next problem was, our goal for boost was outside the range of the stock MAP sensor. So we installed a TurboXS Map sensor, after about 13 reflashes we had it configured properly other than than a CEL from low input caused by high vacuum during deceleration. This bit me in the ass when we went to speed density because when this fault is present speed density stops working as well. (Thank you Merchgod for providing the logger definition so I could tune boost above 37 psia)

So, I bet you guys are asking how you tune the fuel/timing for Speed density once the MAF signal is removed? You have to log Engine Load directly, this almost makes tuning much easier. Just tune the fuel map to hit your target AFRs and timing. The car runs prefect in speed density if you get your injector scale and latency settings correct.

All the problems of the previous tune have been solved. All hesitation is gone, car idles perfectly, AFR is nice and steady since there is no MAF problem caused by blow-thru. The tune is not complete yet though, AFRs are still a bit on the rich side (dipping down into the 9s) but the car still makes amazing torque. That 30R can move some air that is for sure.

I have not verified if this works with the WRX ECU yet. It appears that it would, but I tried it on my g/fs 04 WRX automatic. The Automatic ECU uses MAF for something to do with the transmission and it gave me a ton of codes along with a blinking oil temp light.

Also, blocking the CEL for MAF sensor low input caused the car to not run. I guess the only real tweak that needs to be made to the code is to block the CEL without disabling the Speed density mode that the CEL causes.

*Note* To anyone who wants to attempt this, DO NOT hack up your harness. The pin in the connector can be easily removed with a small flat head screw driver. I'll have pictures and instructions on how to do this once I get some batteries for the camera

Attached are logs from ECU Edit. I know criticism is coming because of the really rich AFR in the speed density tune, but I still have more work to do. Spending 8 hours of the day in a car tuning has drained me and I'll start up again later. Also I have a copy of the ROM before and after.


Attachments:
File comment: ROM after touching up the previous tuners map.
Mike 30R - MasterTune Rev 13.hex [1 MiB]
Downloaded 104 times
File comment: Speed Density Rom
Mike 30R - MasterTune Rev 13 - New MAP Sensor82-SD17.hex [1 MiB]
Downloaded 123 times
File comment: A run through the gears after speed density tuning. Still very rich, but the owner is happy with the way the car drives. Just need to stop dumping so much fuel, haha
ecuEdit_AJ930-8791_2007-06-23_14-43-47.csv [9.38 KiB]
Downloaded 73 times
File comment: Before Speed density or the MAP sensor. Watch the MAF voltage, dips caused some small knock at this phase even with the rich AFRs which I set up because of the dips so it wouldn't Run so lean.
ecuEdit_AJ930-8791_2007-06-13_21-04-17.csv [3.84 KiB]
Downloaded 57 times
Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 3:22 am 
Offline
Newbie
User avatar

Joined: Tue Mar 13, 2007 3:05 pm
Posts: 43
wow, that all seems to make sense. but it also seems to too good to be true.

also i bet the reason it wouldn't run when you disabled the CEL for low MAF input was because the ECU relys on that CEL to trigger Speed density, otherwise it keeps trying to run normaly useing MAF but since its not reading any air flow its not running.

_________________
2004 WRX STi (WR Blue)
Organic Racing Clutch
TurboBack Exhaust + Up-Pipe
Stage 2 Ecu Reflash @ 18 psi


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 4:47 am 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
The CEL turn off actually leaves some bits turned on in the background.

Turning of P0420, for instance, does not let you leave your rear O2 sensor out of the exhaust. You will still get the rich idle (~13-13.5:1) with the CEL in either position.

I think there are other certain failsafes you may be triggering when you kill the MAF signal.

I'm very surprised your fuel map isn't a complete mess. There is definitely NO VE map on that fail safe, it just uses some simple multiplication from manifold pressure.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 7:53 am 
Offline
Senior Member

Joined: Thu Aug 03, 2006 10:40 am
Posts: 1430
very interesting.

having tuned my utec in speed density mode for a while, i can tell you that it's really quite a simple map. ie, the variation from max to min ve, at least in the higher load zones (where you would be running open loop and not closed loop) is not that great.

therefore making an assumption that the engine's VE is a constant percentage regardless of rpm or load isn't a deal breaker, as long as you have some way of altering the ultimate fuel target. it appears that the fuel maps are still "working," so that provides all the input you need.

does this affect closed loop operation and fuel trims in any way you can see?

ken


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 9:38 am 
Offline
Administrator
User avatar

Joined: Wed Mar 29, 2006 10:38 pm
Posts: 5340
Read the SS that Jon posted in the other thread. The MAF failure mode impacts knock correction.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 12:09 pm 
Offline
Experienced

Joined: Thu Oct 19, 2006 5:21 pm
Posts: 134
Location: Roy, WA
All the systems seemed to work just fine. Closed Loop and fuel trims worked, knock correction worked ok. IAM fell and rose albeit slower than usual. The hardest thing is getting injector scale/latency correct so the car will start and idle correctly, it is a bit different from MAF tuning.

I think the biggest part of using this method is it leaves the IAT sensor in place and working, the ecu had to be making some kind of compensation because the AFR difference between a heat soaked 120* and a cruising IAT of 70* was about .1 AFR.


We will be doing more testing, colder weather, different altitudes, etc. And see how the system reacts. Other than that, the car drives smooth and the throttle response is better than the Blow-thru. Although I think the blow-thru setup could have worked a little better if I had tuned it from scratch.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 5:17 pm 
Offline
Newbie
User avatar

Joined: Tue Mar 13, 2007 3:05 pm
Posts: 43
you should try putting an air temp probe in the intake manifold, or atleast after the intercooler, and use that instead of the IAT.

_________________
2004 WRX STi (WR Blue)
Organic Racing Clutch
TurboBack Exhaust + Up-Pipe
Stage 2 Ecu Reflash @ 18 psi


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 5:58 pm 
Offline
Experienced

Joined: Wed Jul 26, 2006 3:19 pm
Posts: 650
Location: Connecticut, USA
LilJohn wrote:
you should try putting an air temp probe in the intake manifold, or atleast after the intercooler, and use that instead of the IAT.
I think he mentioned the car in question uses a "blow-thru MAF" which usually means the MAF/IAT sensors are located downstream from the turbo and intercooler.


Top
 Profile  
 
 Post subject:
PostPosted: Sun Jun 24, 2007 7:14 pm 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
The car uses IAT for a lot of other things. I know at least on the DBW cars IAM won't advance when IAT is not within 0C to 45C or so.

Having IAT in the downstream doesn't mean it recalculates fuel injection based on air temp. With SD, the colder the inlet temperature (at the throttle body or in the intake manifold) the more fuel you need to inject. The stock ECU cannot do this, there is no map for it. So your tune for a warm day will run lean on a cold day. It doesn't matter where you put the IAT.

There are some things that need to be hacked in before it will really work right. I'm glad it is working for you, but I wouldn't load that on to someone else's car as a guinea pig.


Top
 Profile  
 
 Post subject:
PostPosted: Mon Jun 25, 2007 9:15 pm 
Offline
Newbie

Joined: Wed Jul 26, 2006 12:23 pm
Posts: 21
in log 14-43-47 how come the throttle opening angle never changes?

how do you do the fix to tune above 37psia? how high will it let you log?


Top
 Profile  
 
 Post subject:
PostPosted: Mon Jun 25, 2007 10:07 pm 
Offline
Experienced

Joined: Thu Oct 19, 2006 5:21 pm
Posts: 134
Location: Roy, WA
Freon wrote:
The car uses IAT for a lot of other things. I know at least on the DBW cars IAM won't advance when IAT is not within 0C to 45C or so.

Having IAT in the downstream doesn't mean it recalculates fuel injection based on air temp. With SD, the colder the inlet temperature (at the throttle body or in the intake manifold) the more fuel you need to inject. The stock ECU cannot do this, there is no map for it. So your tune for a warm day will run lean on a cold day. It doesn't matter where you put the IAT.

There are some things that need to be hacked in before it will really work right. I'm glad it is working for you, but I wouldn't load that on to someone else's car as a guinea pig.


I got the information from the folks over at ECUTek. They said it should compensate based on IAT, whether that is true or not remains to be seen. I don't think anyone else here has tried it this way specifically so we are in uncharted waters.

How do you know that there is no map for it? There are still large portions of the ECU we have no figured out yet. It could be a fail safe only map, we just don't know.

The owner of the car is willing to be the guinea pig on this one, he brought me the idea and since it appeared like it would fix the problems with his car, we tried and it works pretty well. I will be doing additional checks and tuning on this car, this wasn't a set it and forget it tune.


beavis820 wrote:
in log 14-43-47 how come the throttle opening angle never changes?

how do you do the fix to tune above 37psia? how high will it let you log?


I am not sure why in that log it shows the throttle angle at the same value. Other logs I took were based off of full throttle and it worked fine, the other logs also show the throttle position correctly. I guess that was a bad log to show and didn't even realize it had gotten stuck at a certain value.

Merchgod provided me with a definition for the RomRaider logger that allow me to read what the ecu said for psia. I was able to read all the way up to 44.5 psia at one point when I was tuning the boost for 24 psi (that was one hell of a spike). I believe he is in the process of adding that definition for other roms as well and it may show up in the next logger update. He has been a great help in that regard since I needed that for my speed density experiment.


If anything, at least this sparked some debate. If in fact the ECU does not compensate by IAT, then that could be the only major code change they need to make for Speed density to be a viable solution.

*I really should have picked up batteries for my camera tonight..*


Top
 Profile  
 
 Post subject:
PostPosted: Mon Jun 25, 2007 11:20 pm 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
Master2192 wrote:
How do you know that there is no map for it? There are still large portions of the ECU we have no figured out yet. It could be a fail safe only map, we just don't know.

I have looked through every 2D and 3D map that exists. All 2D and 3D maps are defined in the ROM, so it is very easy to look through them if you have an XML that shows them all. None of them appear to be a VE map.

Quote:
If in fact the ECU does not compensate by IAT, then that could be the only major code change they need to make for Speed density to be a viable solution.

... but why not just write it from scratch then? If you're inserting any code it is pretty easy to then just rewrite the whole darn subroutine for MAF scaling.

And this still doesn't get around the fact that you're triggering other failsafes, as Merchgod posted.

I think it would be a lot easier just to write it from scratch.


Top
 Profile  
 
 Post subject:
PostPosted: Tue Jun 26, 2007 9:06 am 
Offline
Senior Member

Joined: Thu Aug 03, 2006 10:40 am
Posts: 1430
Freon wrote:
Having IAT in the downstream doesn't mean it recalculates fuel injection based on air temp. With SD, the colder the inlet temperature (at the throttle body or in the intake manifold) the more fuel you need to inject. The stock ECU cannot do this, there is no map for it.


well, let's think about this a minute...


IF the stock ecu does not have a map for it (and i have no reason to doubt your claim that it does not) then that just means we can't ADJUST it.

it doesn't mean that there is no correction inherent in the rudimentary S-D calculation that's occuring in the background.

judging by the testimony that a wide swing in logged IAT did not result in a wide swing in AFR, i'd have to say that obviously there is SOME level of temperature compensation occuring, even if we cannot adjust its parameters.

this is similar to the way the utec's speed density fueling algorythim works:

1) MAP and IAT are used to perform a background "1st order" calculation,
2) the user-tunable 3-d VE map is applied against an inaccessible target AFR map, and then
3) a separate user-tunable 2-d IAT compensation map is applied.

with the oem ecu it does appear that we have 1.

2. is a bit different in that we CANNOT adjust the VE map, but we CAN adjust the target AFR map. in the long run, this is effectively the same thing...

3. is the thing that we do not have in adjustable/programmable map form

the background calculation is actually not complicated, and the relationship between gas pressure, temperature, volume, and mass is well defined. imo getting a good SD fuel map is 95% plugging and chugging (which can be easily done in the ecu) and 5% tweaking.

i can tell you that despite its very simplistic nature the utec SD fueling runs remarkably well with zeroed IAT compensations. i have been slowly working on dialling mine in but they are still relatively small, ie much less than 5% correction despite IATs ranging from below 32*F to above 100*F. i think it's important to note that these are the "2nd order" compensations, almost "error terms," instead of the fundamental terms of the initial density calculation.

very interesting thread... :) i might just have to try this. i GREATLY prefer the effectiveness and ease of tuning that comes with a SD based load.

ken


Top
 Profile  
 
 Post subject:
PostPosted: Tue Jun 26, 2007 9:49 pm 
Offline
Experienced

Joined: Thu Oct 19, 2006 5:21 pm
Posts: 134
Location: Roy, WA
Freon wrote:
Master2192 wrote:
How do you know that there is no map for it? There are still large portions of the ECU we have no figured out yet. It could be a fail safe only map, we just don't know.

I have looked through every 2D and 3D map that exists. All 2D and 3D maps are defined in the ROM, so it is very easy to look through them if you have an XML that shows them all. None of them appear to be a VE map.

Quote:
If in fact the ECU does not compensate by IAT, then that could be the only major code change they need to make for Speed density to be a viable solution.

... but why not just write it from scratch then? If you're inserting any code it is pretty easy to then just rewrite the whole darn subroutine for MAF scaling.

And this still doesn't get around the fact that you're triggering other failsafes, as Merchgod posted.

I think it would be a lot easier just to write it from scratch.


Yeah, but it may not even be a compensation table. It could just be an equation based on the ideal gas law.


But I did just get an idea on how to test if there is IAT compensation. When the MAF connector is unplugged it throws a IAT fault and the ecu defaults to -40 Fahrenheit. I'll tune it the way I have been, make sure AFRs are steady. Then unplug the MAF connector and retest, if the AFRs get way richer then I know for sure that it is compensating :)


Top
 Profile  
 
 Post subject:
PostPosted: Tue Jun 26, 2007 10:14 pm 
Offline
RomRaider Donator
User avatar

Joined: Sun Apr 09, 2006 12:05 pm
Posts: 866
Location: Indianapolis, IN
Color me skeptic, but I don't think that is supported by the evidence. I don't doubt it is possible.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 36 posts ]  Go to page 1, 2, 3  Next

All times are UTC - 5 hours [ DST ]


Who is online

Users browsing this forum: Bing [Bot] and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Style based on FI Subsilver by phpBBservice.nl