Fuel Computers - Your Choice

[Alex]

Hi All,

 

Whats your take on the Apex'i Super AFC and the HKS AFR Fuel Tuner. I know the AFC modifies the fuel map, but is that what the HKS AFR does aswell? Just checking this is a fair comparison, or should it bebetween the F-ConV and the AFC?

 

If I was to buy an AFR or an AFC would I also need a HKS Adjustable Fuel Pressure Regulator or is that the basic version of the AFR. Or is that the same as an FSE Fuel Pressure Booster.

 

Cheers,

 

 

(Edited by Alex Holdroyd at 10:42 pm on Nov. 24, 2001)

[Ring Master]

 

Alex,

 

I think the days of add on kludges are numbered,

save your money, wait for the mappable ECU.

It wont be long now.

 

Keep boost down to 1.1 , enjoy 360ish bhp, and when the

new ECU is available, with a FMIC, cams and 550cc injectors

and your own map you should seee 400bhp+ ultra reliably

 

8-)

 

Its all exciting stuff, a motec fitted to an Evo gives over 30bhp

with no boost increase at all !!!!!!!!!!!!!!!! I guess by running an

ignition map which is perfect for 98ron fuel.

 

 

 

[Alex]

Point taken, but will they take a MINES ECU as part exchange!!!!

[Guest Ash]

Quote: from Alex Holdroyd on 9:58 pm on Nov. 24, 2001[br]Hi All,

 

 

If I was to buy an AFR or an AFC would I also need a HKS Adjustable Fuel Pressure Regulator or is that the basic version of the AFR. Or is that the same as an FSE Fuel Pressure Booster.

 

Cheers,

 

 

(Edited by Alex Holdroyd at 10:42 pm on Nov. 24, 2001)

 

 

Fitting an AFC by itself is useless. Unless you happen to like the look of it and/or want to use the device for "monitor mode" features only. In which case, I'd advise the Apexi as it has a better display.

 

The primary reason for fitting an AFC is to provide the facility of a fuelling map (albeit a crude one) that extends beyond the capability of the stock ECU.

 

This can ONLY be done by using an AFC in conjunction with either: 1) an increase in fuel line pressure, or 2) by increasing the size of the injectors.    

 

Take the case where, say, 20% larger injectors have been fitted. Obviously, the engine ECU doesn't know this so it will continue to feed the injectors with the same signals as before. Except under light cruise conditions - where the ECU runs in closed-loop mode - the ECU will correct the fuelling so as to run at 14.7 A/F. So we need not bother about light cruise conditions.

 

However, under open loop conditions, the ECU will not make any fuelling corrections as the output from the oxygen sensor is ignored. Which means the engine will run 20% richer than normal simply because the injectors are 20% bigger and the ECU has no way of knowing it.

 

Increasing the fuel line pressure has the same effect.

 

Take a fuel injector that is rated to supply 440cc of fuel per minute at a line-pressure of 36psi. In theory, if you were to increase the line pressure by a factor of four, thereby raising it to 144psi, that same injector would flow 880cc of fuel per minute.  

 

So if you were to raise the line pressure such that the fuel injectors flowed 20% more fuel, under open-loop conditions, the engine will run 20% richer than normal because the fuel line pressure has been increased and the ECU has no way of knowing it.

 

Now you need to consider that the one device on the car that primarily determines how much fuel to squirt into the motor (on the J-spec that is) is the manifold pressure sensor. Forget conditions such as the warm-up cycle; closed-loop under light cruise; or fuel shut-off on the overun, etc. When you are *driving* the car, as opposed to just poodling along, it is the output signal from the manifold pressure sensor that tells the ECU how much fuel the injectors should squirt.

 

The manifold pressure-signal output, is a simple DC voltage that varies between zero and +5 volts. The sensor is rated at 2 Bar Absolute pressure. So zero volts refers to zero pressure and +5 volts refers to 1 Bar boost pressure. Pressure values between these two extremes give a corresponding voltage output.

 

What an AFC does is to tinker with the value of this pressure signal.

 

Okay, so we have a motor that runs 20% richer than normal. We then connect an AFC and set it so that the pressure signal, fed to the ECU, is divided by a factor of 20%.

 

So to recap: if we were to increase injector size by 20% thus making the engine run 20% richer (or increase the fuel line pressure to achieve that same end); and if we then connected an AFC, and programmed it to subtract 20% off the manifold pressure signal, before it is fed to the ECU: in theory, the fueling would remain exactly the same as before.

 

However, one important thing has changed. And that is, we have divided the manifold pressure signal output by a factor of 20%.

 

The level of the manifold pressure signal which is fed to the ECU, at 1 Bar boost, with a 20% correction factor applied by the AFC, will be only 4 volts. As opposed to the maximum of 5 volts as it would be at 1 Bar boost under stock conditions.

 

Under stock conditions, driving the ECU with a pressure signal voltage of only 4 volts, at 1 Bar boost, would cause the engine to run lean. But the fitment of 20% larger injectors (or a corresponding increase in fuel line pressure) allows the engine to fuel normally.

 

So what has been created, is a 1 volt fuelling-headroom. (Though, in practice, it is slightly less because of fuel cut that occurs around 4.85 volts.)

 

There is still the problem that the stock MAP sensor cannot detect boost pressures above 1 Bar. But this is easily got around by fitting an additional, higher-reading sensor that normally comes with the AFC unit. The Apexi-supplied pressure sensor, for example, has the ability to read up to 2 Bar boost pressure. Which is ample.  

 

Now the engine should be able to be fuel-mapped to 1 Bar plus 20%. Which equals roughly 17.5 psi.

 

The maximum correction factor that can be applied by the Apexi AFC is 50%. So, in theory, up to 50% bigger injectors could be catered for. Thus giving a fuelling headroom corresponding to a maximum boost pressure of 1 Bar plus 50%. Which equals 21.75 psi.

 

However, I do strenuously point out that these figures are purely theoretical. I would not recommend that this be done in practice, as fitment of an AFC has one MAJOR flaw.

 

The AFC takes no account whatever of one CRUCIAL factor, a factor that is *especially* important in a high-power forced induction engine, which is: ignition timing.

 

The fact of which places AFC devices firmly and squarely in the Cludge Department. Which basically means that such a device IMO could not seriously be recommended for fitment to a forced-induction engine making any kind of real power.  

 

But they do look pretty, and can have some useful monitor functions - especially the Apexi unit. I suppose for fuelling corrections in the order of, say, 10% coupled with a small increase in fuel pressure they would be okay.

 

But the true answer, as Justin has pointed out, is a properly mapped ECU.

 

Yours,

J

 

 

[Alex]

I've got 540cc injectors (MINES) and that MINES ECU unit.

 

So that's about a 20% difference over stock...if my mental maths is working at all...possibly not as its a Sunday.

 

Should any "tuning" centre be able to work out if the engine is running lean/rich....and therefore then be able to tell me if I should get my fuelling re-mapped? Is this recorded by the diagnostics??

[Guest Ash]

 

 

I think Chris Wilson offers that service. I'm sure I read a short while ago that he had sourced a portable UEGO sensor. All you need do is hook it to the tailpipe, find a long stretch of road, and give a blast. Not a very technical way of doing it. But it should give you a good idea of what's what.

 

Fuel mapping is not recorded on the OBDI car. The OBDII diagnostics are designed to detect lean conditions, under some circumstances, and misfires (amongst other things).

 

Yours,

J

[Chris Wilson]

Quote: from Ash on 1:32 pm on Nov. 25, 2001[br]

 

 

But they do look pretty, and can have some useful monitor functions - especially the Apexi unit. I suppose for fuelling corrections in the order of, say, 10% coupled with a small increase in fuel pressure they would be okay.

 

But the true answer, as Justin has pointed out, is a properly mapped ECU.

 

Yours,

J

 

 

 

So that's 2 people at least now in agreement with what I have said for years...

 

Maybe in another year or so people will then agree that to set up a "proper" ECU you need the engine on an engine dyno, to control all the temps and hold steady state levels of torque, RPM, and boost whilst setting the parameters. A rolling road, or the road itself (shudder) just won't cut it. For sure on road trimming of very light load running and cold start and acceleration enrichment are fine, but not trying to map full torque, full boost on a 500 BHP engine.

 

As an aside I now believe this ECU people are currently all agog about is a re vamp or maybe re box of GEMS own unit. I downloaded some info and demo software (thanks for the link Ash) and the .dll file for the demo is the same as that for GEMS demo software, and the box the ECU is in looks very GEMS like. I will try and speak to someone in authority at Gems this week as I want some Skyline ECU's altering by them.

 

Nice post and explanation on the limitations of fuel cludges Ash, if I may say so?

[Gareth Davies]

Ha ha, I doubt anyone actually thought (believed?) an add on controller would truly replace a fully mappable ECU.

 

So about how much (cost / time) would a proper engine out, dyno tune, engine in cost for a supra? For those not technically minded enough to already know

 

So anyone going to arrange a group buy for these things?

[Chris Wilson]

Assuming that the mods were already made, i'd say to pull the engine, re loom for say a Motec M4, make engine to dyno bed adaptor plates and hook up hoses, make flywheel adaptor, et cetera, a 2 day mapping session, including pump fuel and oil would cast about 5000 pounds. The NEXT engine would be cheaper as mounts would exist then.. This EXCLUDES the ecu itself and wiring in any "unnecessary" bits like the air con idle speed increase or PAS idle speed increase. Dyno time is running around 1000 a day for a decent dyno cell, including operator / mapper fees. Engine mods could be any price of course, depends what someone wants.

[Ring Master]

 

Chris,

 

Tthere are hundred of evo's running around with mappable ECU's

and none of them (as far as im aware) have had an "engine out jobbie" to get a decent mapped ECU.

 

Surely all we need is for one engine out, and then use that map

for all future jobs, trimming it to that particular cars spec on the

road ?

 

I do hope so anyway.....

 

 

 

[Chris Wilson]

Quote: from Ring Master on 8:00 am on Nov. 26, 2001[br]

Chris,

 

Tthere are hundred of evo's running around with mappable ECU's

and none of them (as far as im aware) have had an "engine out jobbie" to get a decent mapped ECU.

 

Surely all we need is for one engine out, and then use that map

for all future jobs, trimming it to that particular cars spec on the

road ?

 

I do hope so anyway.....

 

 

 

If you did one proper mapped engine on an engine dyno, then cloned others to that EXACT spec, and ran the same fuel pump, exhaust et cetera you could just download the original map to other (the same) ECUs. Mapping a turbo car on a rollining road will NOT get a very good map. It's not for the good of my health I am now tearing out my rotary engine and paying Swindons some God awful sum to make engine adptors for it... It is the only way to do i correctly though.

[Guest Ash]

 

 

I hear all that you say, Chris, and have always agreed that these add-on devices have serious limitations. Unfortunately, I only discovered that after spending a load of money and started to try and fit them to my car. But that's another story.

 

I have *long* since come to the conclusion that the only way forward is a properly mapped ECU. But when I use the word "ECU" I don't necessarily mean an aftermarket unit such as the MOTEC. I just mean an electronic system of engine control that is *properly* mapped for the application.

 

For my car I have a design, on paper, for a piggy-back ECU controlling 6 extra injectors that would have been fitted just ahead of the stock injectors. The piggy-back device also has the ability to modify the ignition timing. The injectors would have been fed from an additional fuel pump and fuel line operating in parallel with the stock pump.

 

The theory behind it all was, the stock ECU is perfect for controlling the engine and gearbox up to about 12 psi boost pressure. As it starts running out of steam, I simply bring on another ECU. That way I wouldn't have to bother about trying to find an aftermarket ECU to control the gearbox, or the basics of the engine. (Something which would have engaged me in a LOT of time and expense.) I would simply use the stock ECU for all that, and use my piggy-back ECU to supply extra fuel and/or ignition timing modifications (or whatever modifications) as they were needed.

 

Naturally, I felt my approach bore merit. And it was an avenue held open to me due to the fact that I have a good understanding of electronics. However, I can see that for someone whose knowledge is not so inclined, they are basically stuck with either the Cludge-route, or the aftermarket ECU approach.  

 

Now that AEM are bringing out a fully programmable "plug and play" ECU, complete with base map for both the engine and gearbox for just $2,000 there is now no way my piggy-back ECU design will ever become a reality. Which is a bit disappointing as I put in a lot of research.

 

Oh, talking of AEM, they make no secret of the fact that the base software IS licensed from GEMS.

 

Yours,

J

[mark brown]

Chris & Ash

 

The issue here is that anyone buying an AEM Ecu is going to be faced with the problem that the spec of their engine is unlikely to match any of the maps that the suppliers provide as we all seem to be running different specs and that is the problem with Justin's suggestion of only one engine needs to be put on a dyno.

 

I for one cannot afford to take the chance that if I buy this Ecu that the map provided will be ok for my spec and I can't warrant spending the sort of money required to get it mapped properly.

 

I've currently got a set up without any cludges apart from an AIC that works ok, I know its not perfect, but the sort of money it is likely to cost to have the mapping undertaken properly is unrealistic.

 

Mark Brown

[Alex]

I hear you Mark....

 

My question at the start of all this was which is best out of these two...not, right lets discount these fuel computers and find out which mappable ecu will work best for me!!

 

Of the real answers I've had including a reply to a private e-mail I sent to Nathan at TDi (ooopppsss did I mention TDi damn I'm gunna get it now!! :o :biggrin: ) on this matter is that the S-AFC is the one to have but I still wonder if it is only because the user interface is more "pretty".

 

I would dearly like the AEM/GEMS ECU but I can't wait that long not knowing if I'm running lean/rich....not being able to make sure that I look after my baby when I want to go for a blitz.....The AEM ECU isn't here yet....I don't have 2k lying about ready to buy it I don'y have 1k lying about to get it mapped up properly....but I might be able to afford a second hand S-AFC and a session at a tuning centre...or with a skilled independant.

 

Perfection doesn't exist....hate to have to break it to you so bluntly. I'm doing the best job I can with the resources I have. So if anyone asks a question would it be too much to ask to just answer the question without going off on a tangent about the really expensive options...its not as if I didn't know about the AEM ECU since I started its bl**dy thread on this BBS.

 

Laters All,

 

With a bit of sleep I'll go back to my normal non-confrontational mood    

 

 

[Chris Wilson]

Quote: from mark brown on 9:43 pm on Nov. 26, 2001[br]Chris & Ash

 

The issue here is that anyone buying an AEM Ecu is going to be faced with the problem that the spec of their engine is unlikely to match any of the maps that the suppliers provide as we all seem to be running different specs and that is the problem with Justin's suggestion of only one engine needs to be put on a dyno.

 

I for one cannot afford to take the chance that if I buy this Ecu that the map provided will be ok for my spec and I can't warrant spending the sort of money required to get it mapped properly.

 

I've currently got a set up without any cludges apart from an AIC that works ok, I know its not perfect, but the sort of money it is likely to cost to have the mapping undertaken properly is unrealistic.

 

Mark Brown

 

I understand EXACTLY what you are saying, my best advice, and I TRULY wish someone had told me this years ago, and that the technology existed, is if you are pushing the boundaries and are not 100% sure whether you are lean or rich, is to either buy or hire a wide band Lambda sensor, and find out just what th mixture really is. For sure you need to take ignition timing into account, but you really the controlled environment of an engine dyno cell to get a grip of that. Mixture is something that can be measured on the road quite easily with the right gear, so spend some dough on a wide band or with someone who can measure it for you.

[Guest Ash]

 

Quote: from mark brown on 9:43 pm on Nov. 26, 2001[br]Chris & Ash

 

The issue here is that anyone buying an AEM Ecu is going to be faced with the problem that the spec of their engine is unlikely to match any of the maps that the suppliers provide as we all seem to be running different specs and that is the problem with Justin's suggestion of only one engine needs to be put on a dyno.

 

I for one cannot afford to take the chance that if I buy this Ecu that the map provided will be ok for my spec and I can't warrant spending the sort of money required to get it mapped properly.

 

I've currently got a set up without any cludges apart from an AIC that works ok, I know its not perfect, but the sort of money it is likely to cost to have the mapping undertaken properly is unrealistic.

 

Mark Brown

 

 

Mark, from what you say I think you are not realising there is MUCH more to mapping an engine other than setting the fuelling for maximum power at wide open throttle. Such a condition probably occurs on a road-going engine for maybe only 2 or 3 percent of the total time the engine is running.

 

The ECU comes with a base map that will run the engine under all normal conditions. The idea is, you just tweak the fuelling and/or ignition depending on the boost curve, or type of turbos, cams, etc. that are fitted to whatever particular engine.

 

GEMS have a similar system for the EVO. They swap the ECU circuit board thus retaining the original box and connecting plug (which is a requirement for some race series or other). The new ECU is programmable using a laptop. It comes with a base map that runs the car for the vast majority of the time when it's not at full throttle, maximum boost. Then all you do is tweak the fuelling and/or ignition to suit.

 

The GEMS ECU works very well and costs only about £1,000-odd. Such that, an ECU swap, is fairly high on the agendas of EVO owners wanting to explore the upper limits of their cars.  

 

I would also say that virtually all the MKIV owners on this list would have no need of an aftermarket ECU.

 

Add-on Cludges do have their place and seem to fullfill the needs of the majority of owners on this BBS who have fitted them. Though I have always maintained that people get away with it, in a technical sense, because no-one in the UK is making any real power. (The tiny few who tried all blew their engines.)

 

For more evidence of this, you only need turn to the Yankie list. Engines with big turbos, for example, destroy torque convertors and eat auto tranmissions at an *alarming* rate. Plus, the Yanks blow up engines all the time.    

 

Not only that, our whole consumer market is replete with goods that are packed in fancy boxes, coupled with cleverly worded advertising, that give the impression they can do the same thing as a more expensive item at a tenth (or whatever) the price.

 

Having been around for a while has taught me many lessons. One in particular has been the true realisation of the phrase: almost is not entirely.

 

Yours,

J

[Guest Ash]

Quote: from Alex Holdroyd on 10:48 pm on Nov. 26, 2001[br]I hear you Mark....

 

My question at the start of all this was which is best out of these two...not, right lets discount these fuel computers and find out which mappable ecu will work best for me!!

 

 

 

 

Alex..............

 

Your initial question had nothing whatever to do with which one is better.

 

First, you asked if the HKS AFC modified the fuel mapping like (as you already knew) the Apexi AFC did.

 

Second, you asked about a comparison between the F-Con. Which I didn't attempt to answer as I have no experience of such.

 

Third question was to ask regarding a situation where if you bought an AFR or an AFC, would you also need a HKS Adjustable Fuel Pressure Regulator or is that the basic version of the AFR. Or is that the same as an FSE Fuel Pressure Booster.

 

I *especially* went out of my way to help you understand how AFCs work. And to explain, in detail, why an AFC should be used in conjunction either fitment of larger injectors or an increase in fuel line pressure.

 

In the light of your recent comments, please let me assure you that I shall not make the same mistake again.

 

Yours,

J  

 

 

 

(Edited by Ash at 11:57 am on Nov. 27, 2001)

[Alex]

Ash,

 

Point taken....my initial question was confused....which shows you how confused about the whole situation I am.

 

What I was tryomg to say is that its all well and good talking about the ideal fix. I need to know how, on the cheap, I can get the car mapped to 1.2-1.3 bar using my 540cc injectors.

 

Which "cludge" is the one to get?

 

Apologies for going off on one.....stress levels are high whilst I'm still out of work!

 

I think your first piece was superb ... guess I just got frustrated by my thread getting diverted on to a "in a perfect world" daydreaming thread.

[Guest Ash]

 

 

If I may say, your initial post told me that you were rather clueless about the way AFCs work in conjunction with creating what I call "fuelling headroom". Which is why I went out of my way to try and explain it all for you in some detail.

 

When I read your later post I thought, "Flipping heck, what a cheek." But yes, you are right, sometimes the simplest of things can be mighty confusing.

 

If you are asking which AFC is best, then there is no "best" as such. The Apexi and the HKS units are both well made and will do the job required of them very well.

 

I suppose it ultimately boils down to which front panel appeals most to your tastes.

 

As regards the actual fuelling issue, I'm sorry but I cannot help you with that. You need to take your car to some dyno place where they can read your current fuelling situation and find out what is necessary.  

 

Yours,

J

 

[mark brown]

Ash

 

I do appreciate that mapping an ECU is not just about full power situations, particularly as most time is not spent in this situation. What concerns me is the time and cost it will take to undertake the tweaking and particularly the ignition.

 

The base map the ECU will be provided with, will be based on experience in the USA presumably, how close to either J Spec requirements or UK spec requirements is that likely to be? Also if you have changed cams, turbos etc surely the adjustments to the maps are going to be a bit more than tweaking?

 

The fuelling shouldn't be so much of an issue, as Chris says using a wide band set up this should be relatively simple to set, but as far as the ignition is concerned, can this be done on the road? I guess the complexity is dependent upon how many points in the rev band you plot. Also if you start adjusting the ignition timing the mixture requirements will presumably change. Is accurate mapping on the road feasible?

 

Mark Brown

[Chris Wilson]

Quote: from mark brown on 10:12 pm on Nov. 27, 2001[br]Ash

 

I do appreciate that mapping an ECU is not just about full power situations, particularly as most time is not spent in this situation. What concerns me is the time and cost it will take to undertake the tweaking and particularly the ignition.

 

The base map the ECU will be provided with, will be based on experience in the USA presumably, how close to either J Spec requirements or UK spec requirements is that likely to be? Also if you have changed cams, turbos etc surely the adjustments to the maps are going to be a bit more than tweaking?

 

The fuelling shouldn't be so much of an issue, as Chris says using a wide band set up this should be relatively simple to set, but as far as the ignition is concerned, can this be done on the road? I guess the complexity is dependent upon how many points in the rev band you plot. Also if you start adjusting the ignition timing the mixture requirements will presumably change. Is accurate mapping on the road feasible?

 

Mark Brown

 

 

IMO no, mapping ignition on the road is doomed to failure. Why? It is impossible to load the engine up at a steady boost and rpm. On the dyno I would hold a set of rpm pints, say, for a road engine and given limited dyno time, at every 250 RPM. Then do ignition swings at various load sites. The way I do any dyno mapping, and if it's something very expensive I chicken out and let Swindon Race Engines guru do it, is to set a safe ignition timing figure at that load point, then optimise fuelling, then carefull advance whilst listening for det and watching power. Some engines hit det around the same timing they make max power, others are "nice and kind" and power falls off before it hits det. The former are the risky ones, it's nice to grab say 10 BHP or so more, but fo a road engine I play safe and pull some timing out. A race engine would get mapped with det maybe only a degree more advance away, and the risks are accepted. I listen for det with a piece of wire (stainless multi strand shark hook trace actually :-)) fastened to the block unde a convenient bolt somewhere, and the other end is fastened to the bottom of a baked bean can. Don't laugh, it works great, and Swindons themselves use this method. The human ear is a miraculous thing when trying to diferentiate det from the other noises, something a det sensor and associated software still hasn't got really reliable. Swindons sometimes use a knock sensor too, but really rely on the bean can to see if the electronics are really hearing det or just noise.

 

You can map fuel and timing at as many points or as few as you like, using a few points means extrapolating between them, and you may well miss power that amore thorough mapping could find, or even hit det for similar reasons. Production engines are often mapped every 25 RPM on automatic dynos, dozens of engines getting worn out in the process. Cosworth and Ricardo have these sorts of cells, mega money and doubtful if anyone outside F1 or manufacturers could afford their fees.  The dyno cell and dyno itself are important, but even more so finding someone at home with the software and the intricasies of the system in use.

 

Anyone wanting an insight into how an aftermarket ECU is fitted and tuned might want to read the instructions the Haltech (Oz, low end ECU that's basic but pretty user friendly). I have stuck them on my site as:

 

http://www.formula3.freeserve.co.uk/E6K_Manual.pdf

[Phil Wall]

Fantastic post, Chris.

 

No wonder it costs a bomb to set up properly.

 

I just love the  high tec bean can.

 

 

 

(Edited by Phil Wall at 10:58 pm on Nov. 27, 2001)

[Guest Martin F]

Chris

 

That is a good post and seems to be very logical.

 

From your experience is there any external conditions which can affect the mapping once the engine is back in the car with drivetrain and suspension added to the equation ??

 

 

[Chris Wilson]

Quote: from Martin F on 11:06 pm on Nov. 27, 2001[br]Chris

 

That is a good post and seems to be very logical.

 

 

From your experience is there any external conditions which can affect the mapping once the engine is back in the car with drivetrain and suspension added to the equation ??

 

 

Thanks. Logical is a good word to remember when mapping, it has to be done in a logical order, a basic map achieved, then refined, a bit of to and fro'ing is required until things are optimised

 

 

Once in the car things like cold start enrichment, acceleration enrichment and cruise mixture are best done by an in car O2 sensor and seat of the pants. Economy with smooth operation from cold and smooth acceleration are to be found like this. But make no mistake, you will NOT get things as good as the OE map on an OE engine, they literallys throw months at the mapping and wear out dozens of ngines getting it right. They have cells where the temps are dropped to minus whatever, then Death Valley air temps are tried. The likes of you and I cannot get close to this sophistication, but we can get a very good set up with patience and of course some money thrown at things

 

Curiously I have NEVER found a web site that details a professionals approach to setting up ECU's on an engine dyno, depite careful searching, nor any books on this. Maybe this is why good dyno operators are treated like Royalty, it remains a mystique.

[Guest Ash]

 

 

Top post, Chris. Keep 'em coming mate, absolutely brilliant stuff.

 

Speaking of Cosworth, did you see the program about them on the box a couple of months ago?

 

Came across it quite by chance just as it was about to start. They showed the dyno cells that you speak of. Absolutely fascinating... the sheer precision of the engineering, quality of the workmanship, all second to none.

 

Having the ability to apply whatever steady-state conditions, at one's fingertips, is a luxury usually only afforded by manufacturers, top race teams, and the like. Yes, I totally agree, it makes for a LOT of convenience, which ultimately must save a LOT of time. And which gives the kind of pin-point accuracy demanded by top race teams, car manufacturers, et al.

 

But for a road-going application (even a high power, forced induction one) it is not strictly necessary IMO.

 

My opinion is that it is possible, by using the right equipment, to get a good setup on the road. Obviously, when I say road, I don't mean by bombing up and down the local high street, or anything daft like that.

 

Naturally, as with anything, there are difficulties to be overcome.

 

A main difficulty concerns the fact that you are at the mercy of the elements. Therefore, one has no control over temperature, humidity and atmospheric pressure. But these factors can easily be measured and logged.

 

So too can all the other engine conditions, gearbox conditions, or whatever conditions. They can all be measured, with the same degree of accuracy, in exactly the same way as a company like Cosworth.

 

The technology to do this is readily available, relatively cheap, and the whole setup can be driven by just a basic laptop.

 

Okay, 20 years ago that was *not* the case. However, nowadays, even amateurs such as myself have access to relatively cheap and accurate computer-controlled measuring and data-logging equipment. Which, not all that long ago, would have been way beyond my reach.

 

As an aside, I couldn't agree more about the bean can and the ability of the human ear to detect detonation far better than any known knock sensor. With my setup, I have a mike that attaches to the engine and connects to a small amplifier with volume and tone controls, driving a set of headphones. It really is amazing what you can hear.

 

The way I see it, what it all boils down to is this:

 

I can sense; measure; data-log; compute and analyse all with a known high degree of accuracy. But the one big thing I cannot do is to apply a known, steady-state load.

 

Yes, that's a BIG problem. But I say it is a problem which can be overcome. Given the time, and the right kind of data, optimum conditions can be extrapolated with a degree of accuracy that is plenty for the task in hand.

 

It's what I used to do with the race motorcycle, and I see no reason why the same techniques I learnt then, cannot be applied to my car. After all, it is basically the same combustion process.  

 

But no-one get me wrong. What Chris is saying is spot on.

 

Yours,

J