• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

b230+t LH2.4 running rich, ECU learning backwards

yeah the learned O2 long term trim is definitely applied at WOT / open loop. tho if your car is sorted (mapped correctly, no dumb mods) they shouldn't deviate from zero correction much if at all. technically short term trim is always applied too, but that is reset to the middle as closed loop exits.
 
Ok. Let's do it this way. I realize I don't have the evidence to support my claim... party foul on my part. I gave up on LH2.4 20 years ago and the data I accrued was stored five laptops ago. I also don't have a dog in this fight... I don't own an lh2.4 equipped turbo Volvo (nor do I plan to in the future).

So what kind of evidence could you provide to support your claim that LH2.4 is solid on a sorted car? I can handle being shown to be incorrect in my thinking.

First, I don't expect you (the LH2.4 apologists) to have back to back dyno pull printouts with boost and afr... That's a hard rabbit to pull out of a hat...

What would suffice as sufficient evidence would be something like the datalogs (or plots) of three back to back WOT street pulls of an lh2.4 equipped turbo car... must have boost and afr logged as well to count. If the datalogs (or plots) show a constant (and sane) AFR sweep as a function of RPM and load I'd consider that to be good evidence to support your claim. I suppose plots of AFR vs RPM would work provided you also tell me the boost the car is running.

Granted... this is assuming you log your lh2.4 data and/or use a wideband at all as part of your data collection system instead of relying on looking at what the flashy lights on the afr gauge say as you blast down the street in traffic.... which is kindof another way of saying you have zero data of what your car is doing or whether it's even the slightest bit repeatable.

Alternatively, provided you don't have the data, and I get that not everyone has the data...(see flashy lights above) What's the one thing the OP could do (provided he doesn't have any mechanical issues, boost leaks, etc) to fix his problem (namely AFR going massively rich during boost)? without requiring him to:
1. change has boost and or wastegate settings (MBC, etc)
2. swap injectors,
3. change out an air mass meter,
4. pull out his aftermarket chips, blow them off then reinsert,
5. unplug the battery for 10 minutes.
these by the way being the standard "fixes" for lh2.4 turbo issues.... chipped or unchipped.

What's the one thing he could do to fix his issue? Surely on a 'well sorted' lh2.4 system you could suggest the one crucial change he could make to rectify the issue once and for all?

I await the evidence to support your claim. Convince me that a well-sorted moderately boosted (read more than 7psi) lh2.4 system exists.


If I cared to continue this debate I'd pull my data logs and show you some dyno charts on my 300hp 20+psi 945.

It sounds like a vacuum/boost leak to me. It'll run in closed loop at idle and be stoich stoich but the LTFT is learning to add fuel which makes it far too rich at higher load.

You could eliminate this variable with a chip emulator and disabling STFT and LTFT.
 
If I cared to continue this debate I'd pull my data logs and show you some dyno charts on my 300hp 20+psi 945.

It sounds like a vacuum/boost leak to me. It'll run in closed loop at idle and be stoich stoich but the LTFT is learning to add fuel which makes it far too rich at higher load.

You could eliminate this variable with a chip emulator and disabling STFT and LTFT.
Ahh yes... the 'I did 300 hp on a dyno with lh2.4' as proof of my claim... BT, DT, GTT. Was it 2004 or 2005 that we did that? Can't remember.... old age getting to me.

Would still love to see the datalogs.
 
Im really pretty sure there's no boost leak I've taken it all apart and put it back together a few times. There is a VERY small leak in the intercooler but it holds pressure for a few minutes so that's not the issue. The 02 sensor is just over a year old. And no codes.
Ok so this part right here we need more details on.

What do you mean the iC has a small leak ? Any leak on LH will cause issues. Without a pressure test confirming air leaks hard to say for sure why it would get worse.

This is one of the major issues with these cars as at this age I have seen intake gaskets seep, throttle body shaft seals leak, TPS seals all causing issues that take ages to try down and get to seal correctly for a mod'd LH car to run consistent which is one of the hardest things for most users who overlook those little things.






And without getting to much into the LH thing, once the car is sorted my two LH cars were fairly consistent, my 744 has the 301/302/301 dynos and at the 1/4 mile it always was consistently running the same times and trap speeds for what is expected. I never had back to back runs where for no reason it would be slower in ET or Trap speed. And I have 100+ 1/4 mile passes on LH. 945 was also rock solid consistent As a DD running 22psi. But at the same time I have come across many other LH cars that just never seem to run right but they are not mine. My kids current 945T is stockish (cam, 15g, exhaust,injectors, 11psi) but that is pretty much stock and runs the same AFRs. Then again these are my cars vs your average LH user.

But again my 744 hasn't run LH for 10+ years. I am 100% for ditching LH if you want proper power as the age of these ECUs and sensors can just cause so many issues. We ditched LH on a friends S/C 928 for MS and it made night and day difference in running before we took it to the tuner.

So again stock car rock LH, mod'd car with actual $$ Into upgrades spend the $ for something better ECU wise as at this age the sensors are just not the same as new. I have seen just swapping AMMs and relearning cause 2pt afr difference which is huge where one runs perfect at 11.8 while another runs 13.8 under load while both idle the same, I just do not trust these 30year old parts. Going MS / speeduino costs less than most people spend on a turbo or manifold these days.

Fun thread.
 
As soon as that TPS switch activates at WOT, that O2 data gets ignored for the pre-programmed fuel maps. A 30+ year old 8-bit (?) ECU is NOT going to be able to do closed loop WOT...especially not with a 0-1v sensor...
May be I'm wrong, but I remembrer to had read somewhere that on the turbo models the TPS switch for WOT is not used by the LH (this doesn't mean that on turbos there is no closed loop but that are used other logics to decide when to go in open or in closed loop).

I cannot say how it works on a chipped ECU, but on a stock LH2.4 you can disconnect the O2 sensor and the car simply follow the map. I had a faulty OS2 sensor on my fathers 940 that it was too slow in the response (also if it oscillated between 0,15V and 0,85V as it should do) and this had the consequence of a too rich mixture, so rich that on cold starting the engine made a poisonous gray smoke cloud from the tailpipe.
By simply disconnecting the O2 sensor connector I solved the problem temporary until I replaced it whit a new one and permanently solved the problem.
(Volvo 940 B200FT MY97)

Be aware that too rich mixture covers the OS2 sensor of carbon deposit and then it works bad also if it has been changed recently

P.S. it is normal that learning process takes so long? For what I've seen a few km are enough when the engine is ok (on stock engine)
 
Listen, I don't expect anyone to post actual data... this is turbobricks! I expect ad hominem... not evidence.;)

Firstly and mostly because it's not something people do as part of their normal drive. Hell I was logging data on my lh2.2+T small rod experience/hooptie with the aid of a techedge 2A0 back in the day (had to wait forever to get the darn thing from Down Under)... I had access to afr, boost, temp, rpm. (btw lh2.2 gave a nasty lean spike similar to what lh2.4 did and "chipped lh2.4" was always supposed to eliminate but never did.) People just put together their turbo-ems package one way or another following a recipe laid down ages ago... we're all standing on the shoulders of giants. I just got tired of the guesswork... why the F is the afr so rich this time? Back to back to back WOT pulls, be in on the dyno or street gave completely different AFR profiles without any other changes... It's running great today, why the F is it so lean today? Tuning lh2.anything back then was like learning Fortran from someone speaking Farci, I immediately saw the utility of megasquirt to KNOW what was going on inside the engine and left the lh world behind... forever.

Second, most people just don't care about the data. As long as their combination of AMM/injectors/boost/chips what the internet-wisdom-lh2.4 lore recipe tells them to do and they're not throwing bits of piston out the tailpipe or putting holes in their block at the end of a WOT run, who cares about what the AFR gauge is telling them??.... assuming they even have the WBO2 and are looking at a gauge.

Third, most, and I do mean most people haven't gotten to the ultimate step of attempting to tune lh2.4 (self included). I have invested zero time in it. I mean I looked at it (having already spent a few years in the roll-your-own $120 megasquirt v2.2 encampment), and simply wasn't willing to invest the time learning how to do it well.

Is it possible to obtain a 'well-sorted' lh2.4 turbo car? Of course it is. Maybe it even runs well. Maybe the version of chips you have in your ride match exactly whatever specific scenario they were tuned for the the first place... (in other words, you're using the exact same AMM/injectors/boost/chips, etc, etc, yada yada of the actual car that was on the dyno spitting out the chips).

All seriousness aside, if you do have the data, and are willing to post and/or share it with me, I'd love to see it. 😜
 
Last edited:
Third, most, and I do mean most people haven't gotten to the ultimate step of attempting to tune lh2.4 (self included). I have invested zero time in it.

So you're saying LH 2.4 is worthless and you haven't even tuned it? Of course it needs tuning, like any modified car.

How well does MS work without a tune?
 
So you're saying LH 2.4 is worthless and you haven't even tuned it? Of course it needs tuning, like any modified car.

How well does MS work without a tune?
Like I said... this is turbobricks... I expect ad hominem, not evidence.

I said no such thing. Please calmly demount your lh2.4 high horse and refrain from switching the topic.

And btw, please do me the kind favor of defining what "well sorted" lh2.4 car actually means.... especially when considered that the vast majority of all lh2.4 users have never, nor will ever tune their own shit.... but simply copy a recipe put forward at this point as internet lore.

You know the answer to your 2nd question just as well as any other person on this board.... but that's just it... no one expects to have to "tune" anything with lh2.4... it's just run these injectors and this amm and these magical unicorn chips from such and such manufacturer or "tuner" and everything will be "well sorted". "tuning to an lh2.4 turbobricker is swapping out the injectors or AMM when their hooptie runs like crap....Which one of you is paying $150-$500 per hour to park your car on a dyno with your chip burner and tune your lh2.4? The shops in this area wouldn't even take your money if you showed up with lh2.4.... it's not worth their time. To each his own and provologne.

Still love to see your datalogs... remember to include rpm, boost, and afr.... that is if you cared to continue this debate.
 
Last edited:
740atl: When you want 3 back-to-back WOT pulls, can I do them in somewhat different stretches of road, and how much cooldown, between pulls? Is 0 to 55 (or so) OK? I'd be running my B21FT with original non-chipped 937 LH2.4 / 148 EZK. (My logger is currently out of the car for upgrades, but I should have it back together in a week or two).

HELLOW: I was able to fix some small leaks along my intercooler seams by re-crimping them with a pair of vice grips:
1) remove the intercooler
2) cap the inlet/outlet with link
3) poke a small hole in one cap and inflate with an air compressor blow gun (set compressor to appropriate PSI)
4) pour water over the seams and look for bubbles
5) adjust a set of vice grips to lightly squeeze the seam, squeeze around the bubbling section
6) tighten vice grips just a bit, and repeat until not bubbling
7) if you have time to kill, you can walk around the whole seam a vice-grip-width at a time
 
740atl: When you want 3 back-to-back WOT pulls, can I do them in somewhat different stretches of road, and how much cooldown, between pulls? Is 0 to 55 (or so) OK? I'd be running my B21FT with original non-chipped 937 LH2.4 / 148 EZK. (My logger is currently out of the car for upgrades, but I should have it back together in a week or two).
Hi Bob, sure... if you can do them on the same stretch of road that would be better. Whenever I'm doing street tuning, I try to drive the same route so I can more consistently compare rpm/sec or torque variables to get a sense as to whether the change I made, made any difference in output.... but yes, I'd love to see your logs.
 
740atl: When you want 3 back-to-back WOT pulls, can I do them in somewhat different stretches of road, and how much cooldown, between pulls? Is 0 to 55 (or so) OK? I'd be running my B21FT with original non-chipped 937 LH2.4 / 148 EZK. (My logger is currently out of the car for upgrades, but I should have it back together in a week or two).

HELLOW: I was able to fix some small leaks along my intercooler seams by re-crimping them with a pair of vice grips:
1) remove the intercooler
2) cap the inlet/outlet with link
3) poke a small hole in one cap and inflate with an air compressor blow gun (set compressor to appropriate PSI)
4) pour water over the seams and look for bubbles
5) adjust a set of vice grips to lightly squeeze the seam, squeeze around the bubbling section
6) tighten vice grips just a bit, and repeat until not bubbling
7) if you have time to kill, you can walk around the whole seam a vice-grip-width at a time
Im gonna buy a air compressor and just test the whole intake system this weekend. I tested the intercooler itself with a bike pump and it held 20 psi for a few minutes. It's possible something like the throttle plate poles are leaking or the brake booster.
 
A well-sorted engine of any kind is one where the components can be expected to work together to contribute to providing the desired result. You don't run an 8.5:1 406 inch small Chevy with a 260 @ 0.050 cam with a Performer intake and inch and a half headers, and a TH350 and stock converter. You can't expect to be able to tune anything into this combo. Your comp ratio, camshaft, cylinder head, turbocharger, injector sizing, etc all have to come together in the same rpm range to provide best result.

The dyno we use gets excited when we come with our LH stuff. He makes his living on 600hp and up LS and whatever Ford stuff can get you there, but if you wanna talk about cookie cutter forum combos, there you go. For him, it's fun to see some idiots doing something totally different and having various levels of success. We raced in LeMons on a tuned LH, and no, it wasn't a turbo car, but that doesn't matter. We got more power and used less fuel on our combo after it was tuned, and we won races and consistently ran in the top 10 and often the top 5. My dyno pics for that engine are up here, and we'd do maybe 15 pulls back to back and the only deviations we'd get were the ones we'd commanded. I'm glad that you have success with MS or whatever you use now, and most importantly I hope that it's still fun for you, but to be honest you sound like a guy who forgot this is supposed to be fun.
 
Tuning lh2.anything back then was like learning Fortran from someone speaking Farci,
:lol: Fortran 66 or Fortran 77?

Somehow, this entire thread reminds me of attempting to "tune" a 1 Bbl carburetor with an O2 sensor. (It would be a few more years before a wide-band unit became available... but even the narrow band unit was eye-opening.)

Gosh, to have today's tools, back then... we could have learned so much... or maybe we would have been tossing barbs about Carters vs. Rochesters vs. Holleys, who knows.

Listen, I don't expect [...]
A wee bit incendiary, to start a sentence that way? But hey, I don't expect anyone to change how they address people... ;)

I tested the intercooler itself with a bike pump and it held 20 psi for a few minutes.

Doesn't sound like a leak to me, at least not in the portion you isolated. Heck, that's very similar to what a building inspector would want to see for a rough plumbing test...
 
A well-sorted engine of any kind is one where the components can be expected to work together to contribute to providing the desired result. You don't run an 8.5:1 406 inch small Chevy with a 260 @ 0.050 cam with a Performer intake and inch and a half headers, and a TH350 and stock converter. You can't expect to be able to tune anything into this combo. Your comp ratio, camshaft, cylinder head, turbocharger, injector sizing, etc all have to come together in the same rpm range to provide best result.

The dyno we use gets excited when we come with our LH stuff. He makes his living on 600hp and up LS and whatever Ford stuff can get you there, but if you wanna talk about cookie cutter forum combos, there you go. For him, it's fun to see some idiots doing something totally different and having various levels of success. We raced in LeMons on a tuned LH, and no, it wasn't a turbo car, but that doesn't matter. We got more power and used less fuel on our combo after it was tuned, and we won races and consistently ran in the top 10 and often the top 5. My dyno pics for that engine are up here, and we'd do maybe 15 pulls back to back and the only deviations we'd get were the ones we'd commanded. I'm glad that you have success with MS or whatever you use now, and most importantly I hope that it's still fun for you, but to be honest you sound like a guy who forgot this is supposed to be fun.

I know what a well-sorted engine is and I know that you know what a well sorted engine is.

The point that threewagons was making via an untuned MS system (or whatever aftermarket system) and the point that you are making via untuned SBC systems and the point I'm trying to make about most LH2.4 systems is that it's all going to run like shit... especially at the limits. They might be able to hang just tooling around, but the moment you mash the go-pedal, well... you know what I'm going to say.

The problem is that 99.9% (could probably add a few more 9's) of all lh2.4 users are using cookie cutter combinations. Someone, somewhere, tuned their lh2.4 car... on a street, on a dyno..., but clearly on a very specific set of components, dumped their tune onto a chip, printed a lot of them and sold them... primarily to Tb members.. who knows. I lost count of the number of chip tuners. The chips get mailed out, planted on a decently (or not decently) running engine and sometimes it works, and other times it doesn't When it works well, we call it 'well-sorted'. When it doesn't, we tell the person to swap out his coolant sensor. Of course, actual tuning, with a wideband, on the street or on a dyno is mandatory for extracting maximum performance out of an engine. But who exactly is doing that? Apparently it's you guys and more power to you. I don't have a dog in this fight. I'm happy there's people tooling around with lh2.4. As stated previously, I just don't have the interest (or time) to invest in learning it at this stage of my life. I'm happy where I'm at with my hooptie-squirt. I'm too cheap to do anything more expensive at this point and I'm too old to go back and mesh with lh2.4.

So, going back to what I said the other day....

what kind of evidence could you provide to support your claim that LH2.4 is solid on a sorted car? I can handle being shown to be incorrect in my thinking.

So far only Bob has volunteered to show me data.

:lol: Fortran 66 or Fortran 77?

A wee bit incendiary, to start a sentence that way? But hey, I don't expect anyone to change how they address people... ;)
The fortran was in the 90's... too many dead brain cells have sloughed off in the time in between to remember that particular detail. 77 I think.

Listen... ;)I talk to people here... like I'm talking to you now, exactly as if was sitting across from you at a bar having a beer. You can choose to read negative intent into it if you'd like but you'd be wrong.
 
I smoke tested my 91 740 recently. I found the throttle shaft leaking and the PVC hoses leaking. There are oil stains at both locations.
 
Pressure tested entire intake system yesterday. Found a small leak at the throttle shaft pole but nothing else. Its a pretty small leak. Had to put my ear right up to it to hear it at 15 psi pressure. Is this my issue? Any known ways to fix it or should I just get a lower millage throttle body?
 
I have in the past used the valve hushers that go on the top of the spring to seal this leak on the TB. I think the official part might be NLA. And again at this age things wear out. But a leak at the throttle shaft can cause an issue. See about resolving it and then let LH relearn and go from there.

The white top injectors are pretty much the same size as stock injectors (336 to 350) so scaling wise a stock AMM and ECU should be able to handle those without issues. Are you running the stock CBV or an aftermarket BOV/CBV.

Have you checked for leaks of tears in this?

Sounds like idle and cruise AFRs are correct before and after LH learns and the issue is only under load after learning?
 
Its a 16T stock CBV. The vacuum line to CBV was hooked up during the pressure test and it was not leaking. When I installed the turbo like a year ago I checked and the rubber was in good shape.

Yeah idle and cruise are perfect its just under load after learning the AFRs just keep getting richer as it "learns"

Does the TPS input to the ECU have any impact on AFRs under load?
 
I have in the past used the valve hushers that go on the top of the spring to seal this leak on the TB. I think the official part might be NLA. And again at this age things wear out. But a leak at the throttle shaft can cause an issue. See about resolving it and then let LH relearn and go from there.

The white top injectors are pretty much the same size as stock injectors (336 to 350) so scaling wise a stock AMM and ECU should be able to handle those without issues. Are you running the stock CBV or an aftermarket BOV/CBV.

Have you checked for leaks of tears in this?

Sounds like idle and cruise AFRs are correct before and after LH learns and the issue is only under load after learning?
I think the seal is NLA or superceded to be the same as the husher.

IIRC the chips in this car are from me, for this setup, OP I think you're local to me?

(Not to stoke the fire but) I would take you up on the offer to log AFRs but I don't keep logging stuff in the car generally. Hasn't been needed.

I see your point 740atl. Lots of folks seem to use mystery chips or bodged solutions on LH that really doesn't help the situation. Particularly going "out of bounds" on stock maps or fabricobbling different injector and MAF setups.

For the record my 940 is my only car. I drive about 20k miles a year on tuned LH, some autocross, infrequent track time, and haven't touched the tune in a long while (actually I just put bigger injectors in yesterday so I'm a big liar).

Obviously it's been tuned on a dyno.
 
Back
Top