What I want to know is whether or not the UTEC can for the stock ecu into switching over into its own OLF mode. I am aware that the 04's have some problems with this but Im not talking about that, I am mainly focused on the 02/03 models. I have a UTEC and am aware that for the longest time it relied on the stock ecu fuel map and just piggy-backed off of it. Now they have their own "Open Loop" mode which has its own fuel map but lets ignore that for a moment as well.

This next paragraph is under the assumption that the UTEC's new "Open Loop Fueling" is disabled and not being used.

When the UTEC switches over from its CLF to OLF, its really just transitioning from its 0% column to its 10-100% column. Well what about the stock ecu? Perhaps it has not made its own internal transition from its TRUE CLF(realtime feedback from front-02) to its OLF mode. Here's my real question: Does the UTEC "FORCE" the stock ecu into switching to its own OLF at the same time that the UTEC does or are they not necessarily in sync? I do not know if the UTEC has the power to do that to the stock ecu or not, it is never really explained in that much detail. There is no reason why the UTEC cant make the change over to the 10-100 while the stock ecu is still humming away in its own CLF trying to maintain 14.7 or what is in its own internal fueling table. If these do not happen in sync you could have serious problems.

There is a chance that the UTEC cannot force OLF switching upon the stock ecu and that is why the old recommendation (Before the utecs new built-in olf fuel map) for CLF->OLF TPS value was set at 60%; because that is where the stock ecu switches I think (80% sure). I think they did that so that they would be in sync, due to the fact that the UTEC cannot force the stock ecu to do a darn thing (im guessing).

I have read the manual and used a UTEC for about 3 years now so please no flaming. I think this is a valid question that a lot of people probably just made an assumption on and dont truely know the answer to.

Thanks in advance,
Jeff

The UTEC doesn't do anything directly to the stock ECU. It only fudges the signals the ECU sees. In the older closed loop fueling (i.e. piggyback fueling), it literally adds or removes voltage from your MAF signal, but the stock ECU is still running the injectors. This has no bearing on any OLF/CLF transition delay on the ECU itself. I believe the OL/CL transistion on the stock ECU is most controlled by TPS and its internal logic, so there's no way to really do much about it besides a reflash.

This raises a big problem. When the UTEC "turns on" and you're running anything but the 0% column, you can, say, tune out the overly rich tune the stock ECU has by putting negative numbers in the map. Then you'll fix up a 10:1 area to run more like 11:1. But if you happen to hit that same load site but the ECU is still in open loop and targeting 14.5:1, you're suddenly running 15.5:1 or something.

And then there's another problem. The stock ECU is constantly trimming its fuel system in closed loop, so your tune will fade in and out as the stock ECU adjusts its long term fuel trim. If you modify fueling too much on the light load sites when the ECU is often in closed loop, it will try to "adjust out" your fueling adjustments. The stock ECU will run the injectors in a way that it expects to see 14.7:1 on its own front wideband, but if you fudge the fuel map and it sees 14.1:1, it will try to reach 14.7:1.

When in open loop, the stock ECU ignores its own wideband and just runs according to the fuel trims it has been adjusting under closed loop, and a few other maps in the ECU.

This is why you want to use UTEC open loop fueling. Run the 4.2c or newer firmware and turn it on. Once the UTEC takes over, the stock ECU's learned fuel trims have no bearing on fueling. The UTEC is not fudging the MAF and allowing the stock ECU to run the injectors. Instead it runs the injectors directly so you get much more consistent results. You really want to use OLF on a UTEC.

......jeff.....from what i have read, and I may be wrong, but the first thing any 'reflash' does is revert the '04+ OL/CL back to the '02-'03 table....so if ya got an '02-'03, yer fine as far as the stock ECU goes

Freon - you explain things well for those here who don’t know how the UTEC works, but I already knew everything you said. You did re-iterate a point I was hinting at, which is if you have the UTEC cross over into its OLF but the stock computer has not yet made the switch, then the stock ecu will be fighting the UTEC's fuel mods and still try to get 14.7, which could/would throw off your ST/LT fuel trims.

I already have UTEC version 5.0 and currently run their new OLF. I do not like the poor transitioning that I am getting when it changes from running the stock ecu's closed loop to the utecs own internal open loop map. The AFR will either jump to lean or too rich enough to make the car run poorly for a second or so in some cases. I have tried just about everything to tune this out and nothing has worked.

The reason your car has fuel trims is a good one. If it feels as though you need to add small amounts of fuel that is for a reason. TurboXS made their own fuel table because people wanted a way to scale the injectors without having to put crap like "-20" in their fuel map just to accommodate their 700cc/850cc injectors. If people had an easy way to modify the stock ecu's programming for injector size then TurboXS would have no reason and no benefit to their own OLF map.

I happen to have the ability to re-flash my stock ecu with a Tactrix cable, so I have no need for TurboXS' new OLF map anymore because I can just adjust the ecu in the first place without beating around the bush via huge MAF voltage mods. The utec does continue to modify your MAF voltages the entire time your car is running, whether or not you are in OLF. It only makes a real difference when you are in CLF and the ecu is running 14.7. Here is something to think about: when you are in the UTEC's new OLF past lets say 30%TPS but not yet into the stock ecu's own OLF, the stock ecu is adjusting its fuel trim to try to hit 14.7 or whatever it is targeting for and your trims are getting thrown off. So when you go back into CLF after your WOT, your trims can/will be incorrect.

Fuel trims are designed to compensate for conditions that would throw off your Air/Fuel ratio. Wouldn’t it make sense for these to also apply (only LT) while going WOT? Whatever is affecting your ratios during cruise is probably (definitely) going to throw it off during WOT. Therefore, it would make sense to assume that utilizing the stock fuel trims would only make your tune MORE ACCURATE, not less. If your car is tuned properly in the first place then your trims should be minimal anyways, so what’s the big deal?!?!

Do not take this post/thread as an insult to TurboXS in any way, I love the UTEC and have praised it for years. I also like their other products such as their cast-iron up-pipe that I have on my car right now, its great. I also like their tuner-pro, although it is a bit pricey.

My final point is this: If you can re-flash your stock ecu to accommodate larger injectors and get your trims dialed in to near-zero 99% of the time, then you truly have no need for TurboXS’s new OLF.

If anybody in the Jacksonville area would like their Ecu re-flashed to compensate for bigger injectors, higher redline, or anything else, I will do it for FREE, just send me a PM.

Uncle Scotty - Yea I know im fine in regards to the transition delay, I wanted to make sure that it was not a part of this discussion though, for the sake of confusion or anything.

Jeff

If anybody in the Jacksonville area would like their Ecu re-flashed to compensate for bigger injectors, higher redline, or anything else, I will do it for FREE, just send me a PM.

....but not the '06, yet....right????

....but not the '06, yet....right????

If I can get a hold of one and download the rom, then it would probably take no more than 1-3 days to decode all of the maps, including Fueling, Ignition Timing, Target Boost, Injector size, etc. This is assuming the ECUflash software works with the 06's.

Jeff

It doesn't. I'm working on that right now. I should have DBW ecu support at the end of the month.

Colby

Awesome, well I’m sure the entire ECU flashing community is behind you all the way on this one. As soon as its done then the 06's will be fair game and definitions will be flying left and right. I'll be right there waiting!

Jeff

It doesn't. I'm working on that right now. I should have DBW ecu support at the end of the month.

Colby


....so Uncle no have to buy AP for his '06 :alien:

:banana:

...so....other than a laptop with a usb...what will I need to 'talk' to my '06 WRX 5MT ECU and log/flash it????

The 06 wrx is using an ECU similar to the 04/05 STi. It may take a bit of decoding to get all of the maps out of there because it's a different processor and then, the logic is different on top of that.

Not impossible, but just a bit of work.

Mike

So, back to the original post... what's the throttle position for CL/OL switchover on the non-04s? :D

It's not as simple as that- the logic is the same as the 04/05 cars- just because you cross a threshold, it doesn't mean you instantly go into OL fueling.

From what I've found the RPM switch is 4300 RPM, the TPS swtich is 46% up until 4400 RPM when it goes to 0, and LOAD is 88 until 4400 RPM until it goes to 0.

There are still delay length values, and depending on which triggers are triggered, a different delay value will be used. They are:

366 2075 1098 122
122 976 488 366
366 2075 1098 366
122 732 366 3662

I was able to find the chunk of data in the 02/03 ecus since the values are pretty similar to the 04/05 ECUs. I'm not positive on this since I haven't really examined it in IDA, so take it for what it's worth.

Thansk!
Mike

Do you suppose that subaru programed a delay for a reason? It must be there for a reason but what that is who knows. I am pretty sure it has been found in the rom for 04/05's but I dont know about 02's.

Jeff

Emissions.

Do you suppose that subaru programed a delay for a reason? It must be there for a reason but what that is who knows. I am pretty sure it has been found in the rom for 04/05's but I dont know about 02's.

Jeff

I can remove the delay in the 04/05 ROMs. I think I can modify it in the 02/03 ROMs, but don't have acar to test on.


Mike

Sort of resurrecting this thread....

I started tweaking the CL/OL delay periods on my car (02 WRX) yesterday... the car is much more sensitive to transitioning out of closed loop. At local road speeds and normal throttle inputs, it's perfect... but at highway speeds in 5th gear, it'll come out of closed loop at very low throttle (15%) and moderate load (1.7 on ecuExplorer).

Anyone know the specifics of which CL/OL delay number is being used for each set of conditions?

Sort of resurrecting this thread....

I started tweaking the CL/OL delay periods on my car (02 WRX) yesterday... the car is much more sensitive to transitioning out of closed loop. At local road speeds and normal throttle inputs, it's perfect... but at highway speeds in 5th gear, it'll come out of closed loop at very low throttle (15%) and moderate load (1.7 on ecuExplorer).

Anyone know the specifics of which CL/OL delay number is being used for each set of conditions?

While trying to decode the 04/05 maps, I wrote a logger that logged all of the memory addresses that the programmed CL/OL values were being compared to. This was for the 04/05 ECU, but I think you could do the same for the 02/03. For example, look in the ROM to see what value in RAM the modified load is being compared to, and then log while driving around for a bit. If you log enough of the variable, you should be able to see which triggers are causing the switchover to happen.

What did you change in the CL/OL settings? Just the 16 magic values?


Mike

While trying to decode the 04/05 maps, I wrote a logger that logged all of the memory addresses that the programmed CL/OL values were being compared to. This was for the 04/05 ECU, but I think you could do the same for the 02/03. For example, look in the ROM to see what value in RAM the modified load is being compared to, and then log while driving around for a bit. If you log enough of the variable, you should be able to see which triggers are causing the switchover to happen.

What did you change in the CL/OL settings? Just the 16 magic values?


Smart idea... I'll dig around in the disassembler and see what I can find.

I just changed the 16 magic values to ones similar to what *obb lists as the STi v7 values in their ST manual. It's funny... the car runs great and feels better than ever... just watching the wideband makes me cringe at what my fuel mileage is going to be. :D

If it's any consolation, my car pretty much goes into OL with minor throttle inputs- on the highway, as soon as I hit boost, I am running about 13.5:1. I still get around 22 mpg.

There is hope :)

Mike

If it's any consolation, my car pretty much goes into OL with minor throttle inputs- on the highway, as soon as I hit boost, I am running about 13.5:1. I still get around 22 mpg.

There is hope :)


Good to hear... I just richened the hell out of my low-load, low-RPM sites back when I thought that it'd work even when the car was in closed loop... I think I'll just let it go OL and tune the lower sites to be closer to stoich to keep my mileage in check. :D