Newbie ECU/Tuning Questions

[DSLeach]

Sorry if these questions have been asked/answered before but I've spent the last couple days reading posts on this forum and I haven't found really clear answers.

I'm fairly experienced with ECU operation and tuning but not for Lotus or Toyota systems. I built (SC, CAI, exhaust, fuel system, ported intake, etc.) a 2001 Cobra. Used a piggy back system to dyno tune up to 423 ft-lbs RW torque. Frankly some of the stuff I read on this forum doesn't make sense based on what I think I know about Ford ECU operation and tuning. Ford is not Toyota I know but engine management software is engine management software.

So, to the questions all applying to an 07 Exige S:

1. Will a slightly smaller SC pulley require tuning to be effective? Won't the MAF signal reflect the increased airflow and the ECU adapt by increasing fuel delivery? Is the MAF capacity or the injectors near max capacity for the stock configuration?

2. Has anyone tried porting the blower? Maybe the newest versions of the M62 already include the porting that Magnuson was offering a few years back but when I had my blower ported it was worth about 10 degress lower IAT (improved blower efficiency) at full boost/load.

3. Does the ECU really compensate for the mods like improved intake air flow or exhaust flow? I've read several threads indicating that the ECU eventually learns to cancel out such mods for open loop operation. This is counter intuitive to me as all the ECU strategies I'm familiar with are intended to allow the engine to benefit from better breathing. Some engine management systems are limited in their ability to take advantage of improved air/exhaust flow due to the capacity of the MAF, injectors, head flow characteristics, A/F ECU maps, etc. But, I've never come across an ECU with a strategy intended to counter improved airflow.

4. Is the ECU tuning out a mod more of a problem for closed loop operation vs open loop operation?

4. In one thread, I read that the Exige S ECU uses a constant for fuel pressure rather than picking up a value off the injector rail. Is this true?

5. Are there tuning tools available to "tweak" the stock ECU to allow changes to A/F in open loop operation, ignition timing, etc. short of complete replacement of the control system (e.g. EFI)?

I don't have a Exige S, yet. I will have either an Exige S or a Cayman S soon. I like to tinker with my cars and I'm trying to understand how hard or easy it is to "tinker" with the Exige S.

[GTsRasta]

1) The ECU is not dynamic enough to take increased MAF signals from increased supercharger airflow and make a perfect new tune for more boosted airflow. If that was the case, we wouldn't need engine management systems. A smaller belt won't just create more air, it'll create more heat as well from the supercharger working harder, thus more hotter air goes into the engine, which makes the engine more prone to detonation. Way different than just taking out your stock airbox, and putting in a cold air intake system, especially at different RPM's and loads on your engine after extended periods of use. So yes, a custom tune would be needed.

3) Lotus offers a Stage II exhaust, so that's a guaranteed power adder that doesn't need to be debated. Headers have been debated here a lot, but PPE and Weapon R both offer dyno proven race headers that won't revert back to the power found in the stock header, no matter how long the header is on. And intakes...are just noisemakers apparently. They have been dyno'd with more power than without, but many will argue it's only temporary.

You can either go with a prepackaged modification and ECU tune upgrade that charlie or forcedfed offers, or you can just upgrade your parts and have a higher state of power delivery with a custom tune on the EFI standalone ECU. Your choice.

[charliex]

1. Will a slightly smaller SC pulley require tuning to be effective? Won't the MAF signal reflect the increased airflow and the ECU adapt by increasing fuel delivery? Is the MAF capacity or the injectors near max capacity for the stock configuration?

No, the math would have to be identical or it starts to drift, a couple of people have tried it but i don't know of anything thats confirmed as a success, you'd need a way to be sure it was following the right breakpoints, if you could do that then the MAF tube is feasible, but it wouldn't change that with a smaller pulley you're increasing fuel pressure across the injectors. However if you did make the breakpoints follow the same mapping, i don't really see the point, since the spark/AF would be the same as stock, more air would require more fuel, unless you're just going off the 'lean is mean' theory.


2. Has anyone tried porting the blower? Maybe the newest versions of the M62 already include the porting that Magnuson was offering a few years back but when I had my blower ported it was worth about 10 degress lower IAT (improved blower efficiency) at full boost/load.

i've heard of people doing it, but i have no experience with it, i've heard mixed results.

3. Does the ECU really compensate for the mods like improved intake air flow or exhaust flow? I've read several threads indicating that the ECU eventually learns to cancel out such mods for open loop operation. This is counter intuitive to me as all the ECU strategies I'm familiar with are intended to allow the engine to benefit from better breathing. Some engine management systems are limited in their ability to take advantage of improved air/exhaust flow due to the capacity of the MAF, injectors, head flow characteristics, A/F ECU maps, etc. But, I've never come across an ECU with a strategy intended to counter improved airflow.

i wrote up an explanation of why it happens, and included pictures of the loop as to what goes on, though i'm not sure where i posted it, but its on one of the hundred or so threads with the same question.

its a feedback loop, garbage in, garbage out , and with an adaption capability, it'll drift.

It's not designed to counter it, there just aren't enough sensors or feedback information for it to be able to correctly retune for it and get the data it needs to adapt properly, its all based on fixed fuel pressures, particular air flow etc. cars that can retune for it sucessfully would have things like widebands instead of narrows, fuel pressure sensors etc.

if the MAF reads too much or too little air, it puts it on the wrong breakpoint row, the fuel mix is wrong, the Base AFR could be wrong (its a smaller resolution map) so in turn it attempts to adapt for the wrong settings.

You also have to be able to define 'improved' in the context of the engine itself, its really only going to get better if its suffering any to start off with, or if you think running leaner would add power.


4. Is the ECU tuning out a mod more of a problem for closed loop operation vs open loop operation?

not in my experience, so far i only know one person that says he's been able to sucessfully use a piggyback, lots have tried, part of the problem is turning off the adaption or adapting to it.


4. In one thread, I read that the Exige S ECU uses a constant for fuel pressure rather than picking up a value off the injector rail. Is this true?

yes, not just the S, all of them, there is no fuel pressure sensor,so thats the first problem.

5. Are there tuning tools available to "tweak" the stock ECU to allow changes to A/F in open loop operation, ignition timing, etc. short of complete replacement of the control system (e.g. EFI)?

i have mine, but its not on general release, i know of no other outside of the OEM.

[DSLeach]

Originally Posted by charliex

1. Will a slightly smaller SC pulley require tuning to be effective? Won't the MAF signal reflect the increased airflow and the ECU adapt by increasing fuel delivery? Is the MAF capacity or the injectors near max capacity for the stock configuration?

No, the math would have to be identical or it starts to drift, a couple of people have tried it but i don't know of anything thats confirmed as a success, you'd need a way to be sure it was following the right breakpoints, if you could do that then the MAF tube is feasible, but it wouldn't change that with a smaller pulley you're increasing fuel pressure across the injectors. However if you did make the breakpoints follow the same mapping, i don't really see the point, since the spark/AF would be the same as stock, more air would require more fuel, unless you're just going off the 'lean is mean' theory.


2. Has anyone tried porting the blower? Maybe the newest versions of the M62 already include the porting that Magnuson was offering a few years back but when I had my blower ported it was worth about 10 degress lower IAT (improved blower efficiency) at full boost/load.

i've heard of people doing it, but i have no experience with it, i've heard mixed results.

3. Does the ECU really compensate for the mods like improved intake air flow or exhaust flow? I've read several threads indicating that the ECU eventually learns to cancel out such mods for open loop operation. This is counter intuitive to me as all the ECU strategies I'm familiar with are intended to allow the engine to benefit from better breathing. Some engine management systems are limited in their ability to take advantage of improved air/exhaust flow due to the capacity of the MAF, injectors, head flow characteristics, A/F ECU maps, etc. But, I've never come across an ECU with a strategy intended to counter improved airflow.

i wrote up an explanation of why it happens, and included pictures of the loop as to what goes on, though i'm not sure where i posted it, but its on one of the hundred or so threads with the same question.

its a feedback loop, garbage in, garbage out , and with an adaption capability, it'll drift.

It's not designed to counter it, there just aren't enough sensors or feedback information for it to be able to correctly retune for it and get the data it needs to adapt properly, its all based on fixed fuel pressures, particular air flow etc. cars that can retune for it sucessfully would have things like widebands instead of narrows, fuel pressure sensors etc.

if the MAF reads too much or too little air, it puts it on the wrong breakpoint row, the fuel mix is wrong, the Base AFR could be wrong (its a smaller resolution map) so in turn it attempts to adapt for the wrong settings.

You also have to be able to define 'improved' in the context of the engine itself, its really only going to get better if its suffering any to start off with, or if you think running leaner would add power.


4. Is the ECU tuning out a mod more of a problem for closed loop operation vs open loop operation?

not in my experience, so far i only know one person that says he's been able to sucessfully use a piggyback, lots have tried, part of the problem is turning off the adaption or adapting to it.


4. In one thread, I read that the Exige S ECU uses a constant for fuel pressure rather than picking up a value off the injector rail. Is this true?

yes, not just the S, all of them, there is no fuel pressure sensor,so thats the first problem.

5. Are there tuning tools available to "tweak" the stock ECU to allow changes to A/F in open loop operation, ignition timing, etc. short of complete replacement of the control system (e.g. EFI)?

i have mine, but its not on general release, i know of no other outside of the OEM.

Thanks for the answers. I'm not of the school that "leaner is meaner" but assumed the Toyota ECU was at least as capable as Ford's ECUs. The Ford ECUs at least have the capability to compensate for small increases in airflow (e.g. moving one step down in SC pully) by adding fuel to the increased airflow. Its not a perfect tune but sufficient and a bit on the rich side. This works reasonably well until you get to the point of significant cumulative changes that challenge the range of the AF maping or IAT or the ability of the fuel system to deliver sufficient fuel (fuel pump or injectors).

I'm surprised the Toyota ECU can't accomodate small increases in air flow.

[qball]

Originally Posted by DSLeach

Thanks for the answers. I'm not of the school that "leaner is meaner" but assumed the Toyota ECU was at least as capable as Ford's ECUs. The Ford ECUs at least have the capability to compensate for small increases in airflow (e.g. moving one step down in SC pully) by adding fuel to the increased airflow. Its not a perfect tune but sufficient and a bit on the rich side. This works reasonably well until you get to the point of significant cumulative changes that challenge the range of the AF maping or IAT or the ability of the fuel system to deliver sufficient fuel (fuel pump or injectors).

I'm surprised the Toyota ECU can't accomodate small increases in air flow.

It's not a Toyota ECU... it's a special ECU designed for the Lotus.

[jmiesse]

Read the forum. especially charlie x's posts. You will find that the powerplant is the only thing that is the same.

[Kris from DRS]

The Lotus cars, unlike the Toyota, come with a EFI Technology Italy ECU. Depending on the model, these ECUs are basically modified versions of the Italian firms generic ECUs so that the end result is a proprietary product for Lotus. And the federalized versions are slightly different than the Euro market versions. Charlie is the expert with offering remapping of the OEM ECU.

If you are looking for an ECU that can be fully controlled by the end user I would suggest the stand-alone kit we offer. This product is made by EFI Technology USA (different company than the above mentioned, but the same name makes it confusing sometimes). This stand-alone system is designed to replace the OEM control unit. It will offer some features that the OEM system does not and supports some inputs that the OEM system does not, however it will also lack in some areas compared to the OEM system.

There are many stand-alone systems to choose from. Ours is unique in the sense that it is preconifgured for the Lotus and is packaged with a wiring harness for ease of install. There are different levels of these systems as well.

In short, if you want a good product for a set up that is within the realm of what most people are looking for (i.e. BWR kit and Katana) Charlie has a good solution.

If you are looking for a product that is something you want to tune yourself for your specific needs (i.e. custom parts, require internal boost control, wideband Lambda, custom engine specs, or race conditions that require specific adjustments for different tracks) then the EFI stand-alone is a good solution. You can visit our web sites to learn more about these systems at Welcome to DRS or Welcome to DRS

Kris

[Ryephile]

So it sounds like the Lotus ECU is basically a pain to work with. Since I've found no info from searching, I'll ask: "how does the Lotus T4e function?" Fuel and ignition tables as a function of MAF vs. RPM vs. IAT vs. PSIambient? How does it use its knock sensor; in a "ping and retard" method or off a base map with knock retard in the background just in case? Does closed loop operation occur only below a certain MAF, or MAP, or ??? Subsequently, does that fuel trim get applied to open loop fuel values? Basically, how does the Lotus ECU "learn"?

charliex's explainations about how

Originally Posted by charliex

the math would have to be identical or it starts to drift, ...., you'd need a way to be sure it was following the right breakpoints

would be appreciated to be expounded upon. What do you mean by breakpoints? Datapoints within tables, or other variables? I'm sure he has to know what he's doing or there would be lots of Katana's out there going ka-blam-o. I see charlie making lots of comments about everyone reading AFR wrong, yet on his Gotham site he shows two dyno charts with AFR's, both of them unwisely lean for the vast majority of boosted PFI setups I have experience with to date [assuming the AFR plots are cars running E10 and the WBO2 measurement is pre-cat]. I'm not attacking charlie, I'm just looking for clarification to improve continuity. I have seen independent dyno plots of his ECU tunes with some very nice AFR traces, so I know the goodness is out there, it's just not cool to have bad examples of your work on your main site.

--->Charlie; what are your AFR targets [versus RPM @ WOT] for boosted applications for E10 91 [R+M]/2? Being that the Lotus market is very limited, and thus the knowledge base getting any and all the info they can, it's easy for people to grasp onto "theory" they've heard without learning the hard way. With any PFI setup running our American E10, peak torque ranges from 12.3-12.7:1 AFR pre-cat [which is a half point off the traditional 12.8-13.2:1 with 100% gasoline, of course]. Now, running a boosted setup, especially one at or over 20:1 dynamic compression ratio, demands at least a half number of richness for chemical quench, with a full number the norm. This means real AFR targets for boosted PFI is 11.3-11.7:1 AFR with E10. Of course, for the few people that run 100% gasoline, we can lean that out a half number because the lambda is changed accordingly. I see no reason the 2ZZ-GE is a special case to run AFR's any differently, unless you have knock data to support running leaner during durability testing.

I have experience with tuning many boosted applications, from SRT-4 to STI to Evo to 1st and 2nd gen MINI to carburated American V-8's with BDS 6-71 and 8-71 blowers. Everything from Alpha-minus to closed loop WBO2 MAF/MAP hybrids and K03 to GT35R, so please don't take my questions as a knock or diss, and more of a request to be on the same page.

Finally, I would like to discover how to effectively control the 2ZZ-GE in the Elise/Exige through different hardware configurations. Is running something like an AEM ECU running closed loop WBO2 controlling fuel and spark while the T4e sends CAN to the dashboard a reasonable solution? Does anyone else have an cracked Lotus ECU product, or perhaps charliex is working on such a product for custom applications? Hardware-wise I see nobody working on a twin-screw supercharger, which is tragic since they are so much more efficient than the piggy roots blower.

Thanks and Peace,
Ryan

[charliex]

Thats a lot of questions!, i've answered all of them before too.

Firstly, there are at least 3 twin screw projects that i know of, all of them up and running, two of them are planned to be commercial and in the final stages, the third i don't know the plans for.

Its pretty common for people with experience in other cars to come along and say the cars are runing lean, i use the same target AFR's as the factory does

With over 260 cars running in 10+ countries, i'm fine with where there are.

Also you're looking at dyno sheets with exhaust sniffers, do me a favour and before you start saying stuff isn't correct, or i'm not showing the best examples, find out whats going on first, my tuning software connects directly to a wideband going through the post 02 sensor, and the dyno is usually on a tail pipe, so unless we're decated or running a spare bung with the wideband in it, you're looking at the wrong stuff, some of the sheets are mixed between pre cat and sniffer, thats because i didn't do the 'marketing' runs,

There are base AFR maps in the ecu to work from, so i don't have to guess what the factory is set for, plus the AFR targets achieved are from 3 years of hard work.

Every so often someone comes along with the same things to say though.

I also do document that the traces on the site are done with an exhaust sniffer, again its all there to read.

cheers

[Ryephile]

Thanks for the reply, though it doesn't really help in mapping the strategy of Lotus ECU. With only two pages of threads in this sub-forum, there isn't much info to pick through.

Originally Posted by charliex

i use the same target AFR's as the factory does
..........the AFR targets achieved are from 3 years of hard work.

Your quote is a conflict. 3 years of hard work I'm not doubting. Your ability to crack the ECU, understand the code, and successfully rewrite it cannot go without significant applause. That said, it's a good idea to not confuse your software success with holistic engine management wisdom. Just because Lotus chose data "X" doesn't make it relevant. Now, that said, it's probably not unwise to assume Lotus sets the AFRs for their factory boosted applications at a safe and reasonable level; I'm just not one to take it for granted. Based on how many Loti you've flashed and how many track their cars, it makes a good case that the 2ZZ-GE indeed has good mechanical quench and can operate close to peak torque AFR as a boosted application. You also have to admit the BMEP is generally pretty low compared to other boosted 4-cylinders out there, so that helps in not needing safe-rich AFRs.

Originally Posted by charliex

do me a favour and before you start saying stuff isn't correct, or i'm not showing the best examples, find out whats going on first

Ok, let's calm down here. I commented in my initial question here in this thread that I was assuming pre-cat [like all proper AFR data is gathered]. Chassis dyno facilities that use tailpipe sniffers are just providing a poor estimation. The fact is your website examples isn't providing accurate information regarding your AFR targets. Surely you have some better representations of your work.


Can anyone help in figuring out the Lotus ECU's operational strategy? We know it's MAF with a narrowband O2, but what other sensors are there? Knock, IAT, ECT, atmospheric pressure? Any help is appreciated!

[xtn]

Originally Posted by Ryephile

Can anyone help in figuring out the Lotus ECU's operational strategy? We know it's MAF with a narrowband O2, but what other sensors are there? Knock, IAT, ECT, atmospheric pressure? Any help is appreciated!

Yes. Charlie can.

If you alienate him you might have to start from scratch with the service manual.

xtn

[Ryephile]

I apologize, I'm not trying to anger anyone here, just trying to see how this ECU operates as it appears to vary from the experience I have.

I found the thread "Deciphering the ECU" that scottyb started back in '06. This is a very helpful thread, and it clearly shows how knowledgeable charliex and rob13572468 are with code. It doesn't help much with engine strategy, at least from this side of the monitor. If I were a code monkey I'm sure I wouldn't be asking any of these questions!

[ZAMMY1]

rye, you seem to know all about the questions, you obviously are well schooled on the technology.
we'd probably all like your input rather than a chastisement of the info that is already presented.
could you work with a dump of the ecu and then reverse engineer the code, maybe you could pick up something with a 2zz-ge engine in it and provide even more enlightenment?
we're all listening.
respectfully, sam

[Ryephile]

--->ZAMMY1: Thank you for the kind words. Part of my predicament is 1) I'm still shopping for a 111 to buy, so first hand development is still a ways off, and 2) I'm a mechanical engineer and software code development is very much not within my competent ability. I can develop hardware all day, and give me a GUI, WBO2, 'scope and a big datalogger and tuning the engine is cake.

I'll admit I was ribbing charliex about his AFRs. His customer base and track record however will clearly show he's competent and capable. Basically my point was from my perspective [a new forum member, but hardly new to car tuning] sees his website and wants to know what's up. That's all, not trying to be malevolent!

FWIW, I work for Freescale, so it's likely I can help everyone including charliex in terms of hardware tools. Does the '06+ ECU's still use the 68376 MCU, or did Lotus switch to a different part?

Cheers,
Ryan

[charliex]

First off, i'm quite calm thanks, as i said this happens all the time, people jump in without sorting out the facts first and we end up sorting out all the stuff thats tangential from that, i personally think that you catch more bees with honey than vinegar, but thats just me But i'm still quite thick skinned and i've gotten into way worse than someone just picking apart AFR's.

The negative side is a lot of people read the thread and pick up the wrong info from someone who posts a long resume and why they think it doesn't work properly, but the answer being the person questioning it didn't find out why its like that in the first place and just jumped in both feet, when its just a simple answer like its a tail pipe sniffer.

Like i said it happens all the time, and as i've said i have answered all the questions before, posted diagrams, part numbers, data sheets and so on, its all there for you already.

I currently don't offer the tuning software out to anyone but a few select tuning shops.

The 376's were EOL'd i believe but they weren't quite fast enough to handle the TC, so they're unfortunately PowerPC now.

[Ryephile]

Thanks Charlie. Looks like I'll get searching.

Any chance you have the MCU part number of the T4e 06+ ECU?

Also, have you thought of selling a tuning GUI for mechanical guys like me for $$$?

Cheers,
Ryan

[charliex]

It was the plan at one point, but i didn't go through with it because ultimately someones going to grenade their motor or worse someone elses, through their own mistakes and i'll get saddled with it, people told me that sort of thing wouldn't happen, then it did, so now i only deal with companies that have insurance etc and i haven't had any problems since then. Which is a shame because i spent a lot of time making a reflashable end user ecu that could recover itself, and yes i've been through all the 'what if's'.

563, unfortunately as i said its a PowerPC and it has a much more limited write lifetime, and its difficult to replace the chip if it does go bad, I've yet to hear of anyone reaching the 1000 cycle limit and it failing since its a common enough chip in other cars but they had to move all the long term fuel trims etc to the smaller longer life memory internally.

[Ryephile]

This PowerPC?

If the MPC563 is indeed the T4e MCU, it's a 388 pin BGA, which can't be removed and reinstalled by hand. At least it's a reasonably modern part.

[charliex]

Yes it can. (be reworked that is)

[Ryephile]

I didn't say it can't be reworked, I said it can't be reworked by hand. You need a hot air station. You obviously can't solder BGAs with a pencil iron, LOL. Like you said though, there's yet to be a part that has failed or ran out of rewrite cycles. If that does happen, I have a socketed BGA solution.