ECU remapping of Custom Engines

[M5pilot]

Hi Guys,

we are presently being bombarded with questions from owners who have done engine upgrades like 2.7's, 2.8's and M50/52's with reference to ECU remapping.

Everyone seems to think that there is some generic chip file for these upgrades.

Example:

M20 with M52 crank and Schrick or CAT Cams camshaft.

I would like to make it very clear tha there is NO generic chip file for this engine. You can make an educated guess but it's never ever going to be right. It might be ok under full throttle between 2-3000rpm under 50% load or part load between 5000-6000rpm but the rest will less than ideal ignition and fuelling. It is simply not possible to make a file while the car is not running. Rough guesses can be made but it's simply not going to give you results.

To have this type of 'custom' setup mapped properly you need a properly live mapped custom map.

A 2.7 chip is nowhere near optimimal as many of you'd hope for.

This is the very reason people go with standa;one and Uni-Q upgrades where full control over the ignition and fuelling maps is given and live mapping on road or dyno can be done. Ideally you need to use a dyno (that's an entire seperate subject).

You will spend a few thouand pounds building your engine and having it installed but the most important area is skimped on.

The mapping is the MOST important part of the engine setup. I have seen, on our dyno now, well setup properly mapped 2.5's produce the same power as 2.7's which have not been mapped.

Removing the airflow meter in the process and going alphaN realeases further power and torque with much improved throttle response. This can be done with standalones like the megasquirt (A-Tech) or the Uni-Q (Evolve and Circuit Motors).

Please understand this very basic concept, even the best specification of engine will not produce good power or give good driveability without correct mapping.

Your typical expense on engine management is between £800-£1000 depending on which setup you choose to go for.

If mapped correctly on a dyno and road, this final and most important step will give you the following -

- Best fuelling and therefore fuel economy, no running lean or rich anywhere, safe running under full throttle.

- Minimum best ignition timing accross all load points - results being best power and economy and no pinking ever!

- Best driveability without any flatspots

Using the standard ECU with Uni-Q or going standalone will give you the same results in terms of power and torque. Both give complete control.

For those that want factory like driveability and want to remain NA then the Uni-Q is ideal. For those that have FI in mind then standalone is probably the best route.

I hope this helps those of you who are thinking that a £75 chip will give the best power/torque/fuel economy possible.

For further information please contact the following:

A-Tech - megasquirt standalone
Circuit Motors - Uni-Q
Gunni - VEMS Standalone
Evolve - Uni-Q / MBE / AEM

[Gunni]

It can´t be said enough that power doesn't only come from the engine´s breathing but proper timing and fuel as well.

Especially on a system using a afm system where WOT driving does not get measured from the AFM.
There are tricks around that though ;)

[GrindCulture]

Think this should be stickied?

[BadDave]

With a bit of luck, people will actually read it before asking the usual silly questions.

FFs No two stroker/turbo or supercharged engines are built the same, so why on earth would there be a generic chip available

Standalone all the way IMO
Costs about the same as a piggy back unit anyway, and has more potential for further engine upgrades
I.E. boost control/leccy fan control etc etc

[E30BeemerLad]

so are them ebay chips no good then?

[GrindCulture]

Only in an iS

[StuBeeDoo]

It's the old, old story though, isn't it?.....
Can just about afford a £75 chip, but don't even think about spending a penny more.

[reggid]

IMO try and use the factory ECU as much as possible as while standalone is good in that it can operate almost anything this is only if setup correctly which can be very time consuming. There is a lot more to it than simply 'making' a Load vs RPM map for fuel and timing when the engine has reach operating temp.

In my experience with mine i wouldn't recommend Alpha N on an M20 running a stock manifold use MAP for load leave for the S engines or dbilas ITB setups.

[Gunni]

Why not use alpha-n?

There is nothing wrong with using it no matter the engine and intake setup

[M5pilot]

There is nothing wrong with Alpha N.

So many E30 M3's are running well with this setup.

The skill is in the mapping to make it run properly and safely.

[ShepsEvo3]

I echo everything Sal has posted up here.

I have just converted another E30 M3 with AFM delete and Uni-Q upgrade with CF airbox. Standard 195bhp engine which made 189bhp before and 210 after with. 40 YES 40BHP increase at 6400 rpm.

Its all posted up on the Bavarian board, but as I can't post pics here I'll leave it out. Its a good example of what good mapping can do and the owner has said he now sufferes from foot ache as the car cruises on the motorway with much less throttle

See http://www.bavarian-board.co.uk/forum_p ... 42681&PN=1

FWIW, the Uni-Q can control boost solenoids to keep boost pressure in check (even on after market turbo conversions) and turn on cooling fans when you want, including intercooler ones if you have them. It can also operate water spray onto the intercooler if temps get too high.... although I havn't mastered this yet, it can be done. Cost wise - that depends like everything else.

Oh, although I am a Unichip dealer, I will map all other engine management systems too Thanks for the plug Sal

[eko]

Paul could you work magic on a 56 plate Seat Leon 2lt diesel?
Standard is 140 brake I believe?

[ShepsEvo3]

Paul could you work magic on a 56 plate Seat Leon 2lt diesel?
Standard is 140 brake I believe?

I believe so yes Not done one yet, but Sal maybe the guy for you on that one. I am sure he will have generic maps for that. If not, I can Unichip it

[eko]

Paul could you work magic on a 56 plate Seat Leon 2lt diesel?
Standard is 140 brake I believe?

I believe so yes Not done one yet, but Sal maybe the guy for you on that one. I am sure he will have generic maps for that. If not, I can Unichip it

Its actually a good friends car who is serious about having it done.
I phoned Alan Jeffries in Plymouth last week for him and was quoted £500 for a `chip` supplied and fitted.Do this sound reasonable in your opinion?

[ShepsEvo3]

Paul could you work magic on a 56 plate Seat Leon 2lt diesel?
Standard is 140 brake I believe?

I believe so yes Not done one yet, but Sal maybe the guy for you on that one. I am sure he will have generic maps for that. If not, I can Unichip it

Its actually a good friends car who is serious about having it done.
I phoned Alan Jeffries in Plymouth last week for him and was quoted £500 for a `chip` supplied and fitted.Do this sound reasonable in your opinion?

Its a little off topic buddy, so I'll pm you

[reggid]

Why not use alpha-n?

There is nothing wrong with using it no matter the engine and intake setup

because airflow and hence load does not always correlate with TPS and therefore its often not as economical as something like MAP regardless of how its mapped, its fine if your only after performance though.

Sal there are lots of Sx engines running AlphaN but these have something in common which makes them suitable for it and or not suitable for MAP.

[M5pilot]

I have an M20 2.7 engine running AlphaN. Once mapped correctly it runs well and consistently.

Same goes for the S14 engines.

[reggid]

I have an M20 2.7 engine running AlphaN. Once mapped correctly it runs well and consistently.

Same goes for the S14 engines.

mine ran well and consitantly just not as economical or good as it does now.......there is no reason to run AlphaN on M20 there are no advantages over a MAP setup, its just as responsive with MAP (TPS is used for accel enrichment anyway). Alpha N is the most primitive and least precise usually used when the other methods don't produce good results.

[Gunni]

fyi
F1 engines run alpha-n with map multiplication,
That is the best way of doing things for any circumstances apart from OEM production runs

[ShepsEvo3]

I see advantages of both. Either way, I find its alot better mapping against MAP than by MAF or AFM.

I don't feel the need for MAP on a road/race/track or dodgem car unless you have a turbo car.... for that application, the MAP sensor is crutial.

I also much prefer mapping against TPS as that doesn't move while you alter fuel and timing - unless you get a tired right foot With MAP, I find you have to keep an eye on keeping to the site you want to tune while steady state tuning on the dyno as it appears adjusting the ignition and fuel can have an effect on MAP.

Other than that, I have no other preferances

[Gunni]

MAF tuning is SO easy if your tuning the ecu directly.
It´s ridiculous really

[reggid]

fyi
F1 engines run alpha-n with map multiplication,
That is the best way of doing things for any circumstances apart from OEM production runs

what do you mean by MAP multiplication?

[Turbo-Brown]

There really shouldn't be any difference between Alpha-N and Speed-Density in terms of the accuracy of the load signal.

For a given crank speed and throttle opening, the manifold depression should be exactly the same (perhaps ignoring air temp but that's factored in for seperately)

With either method though, you're not actually measuring air flow, just assuming that for a given depression or throttle angle and crank speed, you have the same amount of air flow every time.

[Jhonno]

With either method though, you're not actually measuring air flow, just assuming that for a given depression or throttle angle and crank speed, you have the same amount of air flow every time.

So affected by changed in altitude..?

[JazzMan]

Ant did a M20B28 chip for me, works well. I know its not going to be 100% (for all the reasons mentioned in the first post) but it does the job until I can sort out a really good solution with a RR map.

[Turbo-Brown]

There will be changes brought about by large altitude changes, but these should be compensated for by a barometric sensors.

[ShepsEvo3]

MAF tuning is SO easy if your tuning the ecu directly.
It´s ridiculous really

What do you do? Just alter numbers in the ecu program and power run to see what effect the changes have made?

[Andyboy]

First of all - yes, a custom engine build needs a custom map. Even the best educated guess won't be 100%.

Secondly - there is NOTHING WRONG with the stock Motronic system as fitted to a 325i, nothing at all. There's not a lot wrong with the standard air flow meter. Sure, it's not the most advanced thing in the world but by the same token it's not like having a carburettor either. It works okay and I have yet to drive an E30 with Standalone Alphasquirt N that drives as well as that silver Hartge H26 which clearly had a lot of detailed ECU work by Hartge. An M20 is a 30 year old 12v engine, nothing more. M3's are a different story because it can ingest and exhaust a lot more air per cylinder.
To get the best from Motronic, you need to increase the fuelling AND alter the timing characteristics. Many '2.7 chips' just chuck some extra fuelling in and hope for the best - and they are shite. They don't take into account the fact that the piston on a 2.7 will arrive at BTDC faster than on a 2.5 so the ignition timing is wrong. And it's pretty fundamental that if the spark doesn't happen at exactly the right time, you will lose a lot of power.
As I've said before ad infinitum, it's no longer considered rocket science to take a Motronic ECU and play with the timing and fuelling and get it pretty much bang on. It is now ancient technology which can still be made to work very well. Starting on the button, excellent driveability, reliaibility and very good power can all be had with what a 325i has as standard.

But, if you really need that extra 7.6 bhp...............

I'm not slagging off standalone by any means, but feel that Motronic and the good old airflow meter gets a bad press. Anyone reading these threads would think that the standard system is hopeless!

[M5pilot]

Anyone reading these threads would think that the standard system is hopeless!

That's generally the impression of the standard motronic.

Many people believe that adding a standalone or uni-q adds power which is completely untrue.

It's the mapping that does the trick and the mapping needs to be done live on a dyno with good cooling with some mapping on the road.

If the AFM can be replaced and the car runs 'exactly' as before then there is no harm.

Gunni's standard 2.5 running just an Evolve chip makes 178bhp on our dyno. Ant's old 2.5 with standard engine running MS without AFM was running 182 bhp on the same dyno.

So around 4-5 bhp is gained from removing the afm on a 2.5

My tests on a 2.7 showed a consistent 8-10bhp with and without an AFM.

You could most definately feel the difference on the road above 4000 rpm.

While some may not think 10 bhp is alot but a 'REAL' 10 bhp is substantial.

As far as I am concerned, the AFM is not as highly restrictive as it's made out to be. The biggest advantage in removing it is the gain you get in response.

Leave the AFM in place or not, what's important is that the mapping is done live giving minimum best timing accross as many load points as possible and good safe fuelling. Under part load you'll want around 14.7 AFR and around 13 AFR under WOT.

[Gunni]

MAF tuning is SO easy if your tuning the ecu directly.
It´s ridiculous really

What do you do? Just alter numbers in the ecu program and power run to see what effect the changes have made?

Inside the ECU there are

MAF curves, a "fuel table" which is actually just a lambda table or correction table from lambda 1.
ecu does all the actual fuel calculations behind the scenes.
So in effect all you have to adjust are the lambda values you want at a particular load , which is measured in kg´s/s.
and it will become that,

it get´s harder when you change the injectors and you change the MAF.
like when you do turbo or whatever. I.e when the injectors become a limiting factor or the MAF outflows.

It´s because MAF systems don´t care about engines volumetric efficiency.

[hoshy]

Seeming as we're in to free plugs:)

Shropshire Automotive Services for guys in the West midlands - UniQ dealer, 4wd dyno, live mapping of MS, Emerald, SMT, Omex, Motec etc.

[daimlerman]

As a DiY type of person myself,I can accept that my lightly tuned 2.7 will benifit from a live mapping session with an expert.My personal problem is that I enjoy playing with the thing myself and find it difficult to come up with the justification to spend the equivilant of two weeks hard earned in this way.If the lotto or the premium bonds were generous to me I would be on the phone straightaway,of course.

[ShepsEvo3]

My personal problem is that I enjoy playing with the thing myself and find it difficult to come up with the justification to spend the equivilant of two weeks hard earned in this way.If the lotto or the premium bonds were generous to me I would be on the phone straightaway,of course.

I understand that ;) Infact all I do here is operate the dyno while the owner plays.... but most of the time, I end up doing it for them as sometimes they get confused. In the long run, its cheaper to just have the car live mapped on the dyno. As long as all the conditions are the same as on the road, there is no reason to re-map or tweak. Unless you change a component like a camshaft for instance.

Gunni, I understand that mate, is it the same for ignition timing? I had a guy doing the same thing to a Fiat Coupe Turbo, he could get into the chip and set anything he liked.... but it was very hit and miss and took a full day dyno session to get right. Mind you we were tuning a 430bhp 20v turbo ;) So nothing standard there. You have pm too

Hoshy - you have pm also ;)

[Gunni]

Yes a timing curve is also based on "LOAD".

this is why piggybacks go massively wrong when you have changed injectors, maf and what not.
As you have to fool the ecu into running less fuel pulsewidth = less load.
less load = more advance

Ecu advances timing = can be bad as you know when load is indeed more then what the ecu sees.
But live mapping by retarding the ignition 10-15degres and then slowly moving up again to find MBT would be ideal.

That´s why I still haven´t seen how the uni-q runs higly boosted OEM cars in piggyback form ;)

[ShepsEvo3]

That´s why I still haven´t seen how the uni-q runs higly boosted OEM cars in piggyback form ;)

I have just come back from a course at Dastek South Africa, and this is one thing which is covered concerning the Uni-Q. It can be done and is pretty easy.... as the owner of Dastek has a Lambo Gallardo which he is converting to twin turbo and will be keeping the management standard with standard injectors and having the Uni-Q operate another set of injectors for when extra fuel is needed. Voltage and Signal clamps within the Uni-Q takes care of keeping the ECU happy, infact it won't even know turbo's and extra injectors were fitted. Its also operating a water injection system too.

Infact there is also a Unichip controlling the gearbox.

There are several Hondas and Suzuki's (in car form) with standard engines and management systems running upto 1bar of boost on the OLD Unichip with no problems so far. I know, I did the conversions. The Uni-Q makes even more things possible.

This is by no means suggesting that the Uni-Q is the definitive solution, as Standalone does have advantages. It all comes down to whats required, time and money. But I do say its a very worthy product for most solutions.