ScoobyNet.com - Subaru Enthusiast Forum

ScoobyNet.com - Subaru Enthusiast Forum (https://www.scoobynet.com/)
-   Engine Management and ECU Remapping (https://www.scoobynet.com/engine-management-and-ecu-remapping-453/)
-   -   ESL Fuel Trims - Gasket leak? (https://www.scoobynet.com/engine-management-and-ecu-remapping-453/1033024-esl-fuel-trims-gasket-leak.html)

The Rig 22 December 2015 04:16 PM

Jaysz , wonder why your fuel trim table is all green where all out others are red and green when an adjustment is made ??

Jaysz 22 December 2015 04:18 PM

Looks like ecu does a lot thats not shown

Jaysz 22 December 2015 04:22 PM


Originally Posted by The Rig (Post 11774140)
Jaysz , wonder why your fuel trim table is all green where all out others are red and green when an adjustment is made ??

Mines the only table that is green as no corrections on it not sure why some different colours

sean turbo2000 22 December 2015 04:37 PM

my table is green with no corrections too, even though ecu is clearly making adjustments when viewing the logs

The Rig 22 December 2015 04:43 PM

weird, if corrections are being made, why are the trims still non adjusted in both your maps ?

I see it as the fuel trims NEED to be adjusted, as we all have different setups, i want more fuel pumped in than say the standard 8S map as i have more boost at say a certain point so want more fuel added, hence the furl trims are showing this

thats how i see them,unless im totally mis understanding them

Jaysz 22 December 2015 04:55 PM

Could be lambda showing less than ideal fueling so trims add or remove fuel
Have u tried correcting it with fuel map

The Rig 02 January 2016 04:17 PM

Well, after a few weeks of learning, here are my fuel trims, car performs well, so im not convinced the cells should be all the same as in not adjusted by the ECU , My lambda is new as well .

https://cimg0.ibsrv.net/gimg/www.sco...f13d01ffac.jpg

ben.harris 02 January 2016 07:19 PM

Ok - just been out to the car and grabbed a screenshot of mine as they currently sit. I spent some time last weekend going round all the vacuum connections, spraying them with Easy Start to see if I could find a vacuum leak, with no success. I also tried attaching a smoke-machine to the inlet manifold and filling it with smoke to see if I could see leaking anywhere, also without any luck - although if I'm honest, I'm not sure I was doing it right! I then reset the ECU, as the Easy Start, and me pulling off random vacuum lines had really upset the fuel trims, so the values below are after a week of driving:

https://cimg6.ibsrv.net/gimg/www.sco...b022ba50eb.png

The trim adjustments still look a bit high for my liking. I've read that anything over 5% generally implies a fault/leak somewhere on a standard OEM engine/map (general advice for any car, not Scoob specific), and I'm not convinced that any of my modifications would be enough to warrant the corrections I'm seeing from the OEM map - although happy to be told otherwise, if others think that a decent panel filter, intercooler Y-pipe, and free-er flowing exhaust could make that much difference.

My current thinking is that it's not a vacuum leak - if it was, the lower trim values would be increased, but not the higher ones. I've got two other possible theories:

Firstly, there might be a boost leak somewhere that only becomes evident once the car is making boost (i.e. not at vacuum). I'm thinking this might be the case since I've not been conviced that the turbo is spooling as quickly as it used to, although my logs show that it is still getting to full boost. Possibly the OEM recirc valve, or the gaskets around the turbo/downpipe. I've got replacement gaskets and another (2nd hand) recirc valve that I plan to fit when the weather warms up a bit.

My other theory is that the fuel pump might not be supplying enough pressure any more and the ECU is compensating by opening the injectors longer to make up for it. The car still has it's original 18-year old OEM fuel pump, so this could very well be a contender and I'm considering replacing this anyway, just as a matter of course.

Also, as an aside, I had the car MOT'ed a couple of days ago, and whilst it failed on a split CV joint boot, it passed the emissions test easily, so the fuel trim compensations appear to be doing a good job for the current state of the engine - implying that they are indeed necessary, and not the result of something like a faulty lambda sensor.

EDIT - I should also add that I don't have my wideband sensor hooked up yet. This is another job on my 'to-do' list, and I'm not going to change from the standard OEM map until I do. I am a bit concerned that I might be running massively lean when it goes open loop at the moment, as if it's having to compensate by adding up to 10% fuel in closed loop, maybe it should be doing the same throughout the range. The wideband looks like it's my next priority to get fitted.

sean turbo2000 02 January 2016 08:08 PM

You need to get it mapped, the basemap that comes with the esl is just that, a basemap to get you going, mine ran great on the basemap, but soon as i fitted a wideband you could see fueling was far from optimal, and closed loop was just compensating for this.

ben.harris 02 January 2016 08:28 PM


Originally Posted by sean turbo2000 (Post 11777476)
You need to get it mapped, the basemap that comes with the esl is just that, a basemap to get you going, mine ran great on the basemap, but soon as i fitted a wideband you could see fueling was far from optimal, and closed loop was just compensating for this.

Andy provided me with the 8S map, so I was under the impression that the map I'm currently running should be exactly the same as the OEM ECU was running for my car - I just had no way of seeing the fuel trim adjustments before having the ESL! I'll be attending the ESL training course later on this month, so after that, and once I've fitted my wideband, I'll be (slowly!) self-mapping it.

Interesting that you found closed loop was compensating for fueling on yours too though. What ECU/map was it running beforehand? Maybe the standard OEM map hasn't 'aged well' as the engines have got older and tolerances changed - if that makes sense!?

The Rig 04 January 2016 11:58 AM

Ben , what does your car drive like if you reset ecu and clear the fuel trims ?

Mine doesn't drive 100% until the fuel trims have been adjusted , once they have car drives supurb

Cheers

ben.harris 04 January 2016 12:27 PM

Seems to drive OK, even after a reset with the fuel trims cleared. They don't jump up straight away, either - they gradually climb. What I need to try is taking it for a drive, and then clearing them and driving it straight away again afterwards so I've got a direct comparison.

Also, I think I know why the '0' engine load cells might be high - I've noticed within the last week that I can sometimes hear my boost solenoid clicking, even when at idle and a bit of Scoobynet research suggests that (although I've never touched it), my TPS might be slightly out of calibration so the ECU isn't realising that I've taken foot off the accelerator and it should be controlling overrun/idle.

I know I really need to get my wideband connected so I can get a better idea what's going on.

bludgod 04 January 2016 01:05 PM

it could be that or it could be someone has set the BCS to be always on in the map. Whats your min/max duty like at the lowest load/rpm site - if it's not zero then the BCS will always be chirping away even it idle.

The Rig 04 January 2016 01:27 PM

Yeah , get your wide and fitted , until you see what AFR is like it's hard to tell if it's running lean/ rich unless majorly lean then you will know lol

You may need to tweak the injector scale to get the wideband to calibrate close to what your AFR table is etc. Unless mafless

Bludgod , my duty cycle even at 0 load is 100 , up to when spool starts then I obviously reduce duty cycle to level out boost , no point having waste gate open when it's not needed to be is how I look at it. :thumb:

ben.harris 04 January 2016 01:28 PM


Originally Posted by bludgod (Post 11778265)
it could be that or it could be someone has set the BCS to be always on in the map. Whats your min/max duty like at the lowest load/rpm site - if it's not zero then the BCS will always be chirping away even it idle.

It is something I've only noticed since fitting the ESL ECU, so that also sounds feasible. Here's a screenshot of all the current boost settings. It doesn't look like it's set to zero at any of the low load/rpm cells (assuming I'm reading it right) ...

https://cimg0.ibsrv.net/gimg/www.sco...cf37414ead.jpg

ben.harris 04 January 2016 01:37 PM

Here's a log from a couple of days ago. Looks like the wastegate duty cycle is at running at 100% all the time until it builds boost - which I get the impression it shouldn't be doing, at idle.

https://cimg9.ibsrv.net/gimg/www.sco...c2b3834a61.jpg

Notice how the wastegate duty stays at 100% for a bit, even after switching the engine off - which was when I first noticed it doing it...

https://cimg0.ibsrv.net/gimg/www.sco...272578f2db.jpg

[Hopefully I'm showing you the right things here! - Still trying to work it all out!]

ben.harris 04 January 2016 01:41 PM


Originally Posted by The Rig (Post 11778280)
Yeah , get your wide and fitted , until you see what AFR is like it's hard to tell if it's running lean/ rich unless majorly lean then you will know lol

You may need to tweak the injector scale to get the wideband to calibrate close to what your AFR table is etc. Unless mafless

Bludgod , my duty cycle even at 0 load is 100 , up to when spool starts then I obviously reduce duty cycle to level out boost , no point having waste gate open when it's not needed to be is how I look at it. :thumb:

Presumably, injector scaling is an adjustment (correction factor) that's applied to the AFR's across the board? If that's the case, could it simply be that the scaling isn't quite right for my (aging) injectors? I know .... wideband, wideband, WIDEBAND! :lol1: - oh, and read the manual! :) [EDIT - that comment's aimed, at me, not you!]

bludgod 04 January 2016 02:09 PM

yeh see your 0/0 for min and max wastegate duty - stick 0 in there so when your off throttle your not using any wastegate duty.

ben.harris 04 January 2016 02:17 PM


Originally Posted by bludgod (Post 11778301)
yeh see your 0/0 for min and max wastegate duty - stick 0 in there so when your off throttle your not using any wastegate duty.

Ah - gotcha. I'll give that a go this evening.
Cheers! :)

bludgod 04 January 2016 02:27 PM

good stuph, for what it's worth I would usually have 0 across the board for 0 load/throttle and idle rpm - no point pounding on the solenoid if the engine isn't in any state to be making boost.

Looks a little like this:

https://cimg9.ibsrv.net/gimg/www.sco...098f4663cc.png

The Rig 04 January 2016 02:54 PM

Yeah your injector scaling is as you describe Ben , I had to lower mine yo 0.85 to get accurate wideband readings , at 1 I was 19 on the wideband , very lean , if yours is still at 1 it could be running lean as an example .

Bludgod , hear what your saying about the solenoid but for faster response I thought it best to not have such a dramatic jump from 0 to suddenly say 100 or 70 duty cycle .

Each to their own I guess bud or up for discussion lol

bludgod 04 January 2016 03:15 PM

yup I get ya - was more for an example but take into account the scaling for X and Y axis as well though, when you plot that as a 3d map and ECU starts interpolating the values it will become a little smoother in operation. Plus at any time you go to 100% throttle you'll be wanting maximum duty from the solenoid so there will have to be some jumps/steps in the table until you get closer to the point where your turbo can actually spool and make boost (test it in 5th/6th gear to see at what RPM you need to start tapering the duty). This way you can avoid surging when climbing a hill in higher gears or just squirting the pedal a little to overtake without downshifting.

The Rig 04 January 2016 11:00 PM

Yeah, ive spent a wee while maximising the spool up point and duty cycle point, think i have it spot on now, this is all on the road tho, on a rolling road who knows what that would tell me compared to what i think is happening ha

https://cimg9.ibsrv.net/gimg/www.sco...7ecd5b4957.jpg

ben.harris 04 January 2016 11:30 PM

Interestingly, on my journey home this evening, there was no sound from the boost solenoid at all when idling. Also, no 3 seconds of 'clicking' after I switched the engine off either. I haven't been near it with the laptop yet, so nothing has changed map wise, so I'm still thinking back to the TPS theory for this one. I've also checked both of the base maps that Andy supplied with with, and whilst they have slightly different boost duty maps, neither of them are at 0 for the idle/no-load areas.

I reckon the ECU might have some different logic, not exposed by ESL in it for the boost solenoid when it knows the TPS is signaling an idle condition, regardless of the boost duty map - otherwise I can't explain the intermittent behaviour of my boost solenoid.

Looking at several of the logs, my TPS is usually sitting around 9% when my foot's not on the accelerator. However, one of the logs where the solenoid started behaving strangely, I can see it drop down considerably lower to 4.5% momentarily when in vacuum. I reckon the ECU's re-interpreting the TPS 'foot-off' value when this happens, and then when it returns to 9% (as shown in the log, literally 10 seconds later), it's thinking that it shouldn't be idling, and is then using the boost duty min/max maps, which would then explain why it's clicking away.

That kinda makes sense in my head as a theory anyway! Perhaps my TPS is dodgy, or the throttle plate is dirty and isn't quite closing properly. Looks like another thing for me to check out, anyway!

bludgod 05 January 2016 09:36 AM

rig - the dyno run would usually set you on a little higher load than you would see on the road so you might get a bit of overboost when doing a pull. If you check your wastegate duty in 5th (try and roll on from your spool point) it will give you an idea of how it might behave on the rollers.

A well setup dyno should be the same as the road, but most times it will be a little higher. There's a nice hill near me which does well in 4th so I try and use that combined with a flat run in 3rd to set the min/max duty (though looks like you have the more basic boost control method with just a MAX duty table and no minimum?). Advantage on a newage car is you can set additional wastegate duty per gear as well so once 3rd-5th is sorted you can add a little more 1st and 2nd to get things moving :)

Out of curiosity - what kind of load are you seeing on your motor at the moment and what's your setup?

The Rig 05 January 2016 12:07 PM

Ben , yeah , adjust your tps if you can , should be around 4% at idle .

Bludgod , yeah I have a 95 classic so don't have quite the adjustments for duty cycles you newer boys have the joys of lol , my setup is nothing special , bigger Tmic , yellow injectors , cone filter , 3" hks exhaust , sportscat , later spec map sensor, new age coils , tightened wastegate actuator , etc .

At full chat my load reading is 64

Cheers

ben.harris 24 January 2016 07:53 PM

Since my last post on here, my car's been off the road, parked up at my local garage for a couple of weeks, due to a CV boot that needed to be replaced for the MOT and a seized driveshaft/hub meaning it's taken the garage much longer to sort out and cost me far more than I'd have liked, all for a £5 rubber boot! But... I got her back a couple of days ago and as today's been nice weather I got around to adjusting the TPS. It was sitting on 9% previously.

For something that should be simple to adjust, the two screws looked like a complete nightmare to get at, so I opted for a slightly more agricultural method of adjustment - namely, a screwdriver held tightly against the top front of the TPS sensor, and a couple of taps from a rubber mallet on the screwdriver to reposition the sensor slightly. 30 seconds later, and the TPS is now sitting at 5% on idle which seems to have resolved the constantly clicking boost solenoid issue I was having previously.

I also had a chance to try out my new Linx 7 - Windows tablet, running the ESL software with the car. It seems to work really well, and for under £50, was an absolute bargain. It certainly gets my recommendation for anyone who wants something small and neat to keep in their glovebox for ESL logging.

Next job... Wideband/AFR gauge.

ossett2k2 24 January 2016 11:34 PM

Nice one ben,where did you get the tablet from?
I'm using an old Samsung Q1 with winXP but it's too bulky to mount correctly so am looking to upgrade.

ben.harris 25 January 2016 12:06 AM

I got it from "Game" a couple of weeks back - they had it on offer for £40 so I snapped one up. Looks like are doing it for £50 at the moment, back in stock February 3rd - look on the right-hand-side, where it says "Other Sellers on Amazon". I'm really impressed with it - it upgraded to Windows 10 straight away, and came with a year of free Microsoft Office license, too. I picked up a case from eBay for £3 and might also get a mini bluetooth keyboard/touchpad for it, but other than that, I'm really pleased with it. Are you coming on the ESL training course on Tuesday? If so, I can bring it along if you want a look?

ossett2k2 25 January 2016 12:14 AM

Cool cheers ben,I've been looking at this kind of table but I thought they were like a mobile version of window like the windows phone,didn't realize they could run ESL.

Not going on the course,I was bang up for it as well but gutted it fell on my week of nights and it's impossible to get cover for my machine :(
It would be great to get to know the ins and outs of ESL properly.
Have a good one pal and be sure to let us know how you get on please,I know there are a good few guys from the forum going so be good to see a thread on how thing were with the course.


All times are GMT +1. The time now is 07:12 AM.


© 2024 MH Sub I, LLC dba Internet Brands