Mappable ECU's - The definitive thread

[SimonN]

After a big thread hijack elsewhere and comments elsewhere, I thought it was about time that we had "the definitive thread" on ECU's because we have had one on everything else and it would be good to have it on something rather less superficial than the usual "cosmetic" subjects 

So, first the disclaimer. Although i have a fair amount of experience in the subject, I do not claim to be an expert or professional. All comments are made with the best of intentions and anybody stupid enough to make decisions based on my comments or any other on this thread deserve any problems they get . However, if this starts a conversation and clears up a few misunderstandings, great.

In general, while principals might apply to other cars, I am really discussing air cooled cars up to and including the 993. While it is a generalisation, the newer the cars, the better the engine management and the less gains there are available from changing the engine management system.

I think there are 3 main areas to discuss. First, we have the standard DME fitted to 3.2C's and cars thereafter. With these, we can look at fitting a new chip. Wong style, getting a custom chip from Wong (and others) based on some measurements taken on a rolling road or, ultimately, mapping on a rolling road to maximise the potential.

From there, we get into modifying the existing system with such things as changing the air flow meter to a better system, such as what is known as the "hot film" conversion (more later).You might have a 3.2C set up, but it can be made to work with twin spark heads and larger capacity engines.

Finally, we can look at  fully mappable ECU's, typified by Motec but including a whole host of others. Each offers something slightly different in execution, but all work on a similar set of principals. Hopefully, over time, we can discuss advantages and disadvantages of each but I think that we need to understand the general benefits first.

For me, I am using this thread as a means of forcing myself to do some revision and updating my knowledge, which is based on tuning reasonably high powered turbo rice (Evo producing 465 bhp, 0-100 in about 3.1 second ) some years ago. I did a lot of research into adding Motec (and other mods) to my 964 RS (hey, Tom, did you know I had one?) which I am pleased I never did because of the negative impact it would have made to the value. I do know a number of people in the UK who have used Motec on 964's and 993's and have driven a few of the cars.

My final thought is a warning. When researching this subject, you will find a lot of stuff on forums such as Pelican and Rennlist. They usually end up as shit fights between those who really know, those who post to protect their own commercial interests (without stating those interests) and those who have no idea. If reading those threads, be very, very careful who you listen to. It took me years to learn who the good guys are.

So, from here, I will post some introductions to the key topics. Please join in. Don't wait for me. Ask questions. Post your own experiences. I don't want this to be the world according to Simon, because that will not get us very far and would be no more than an ego trip for somebody who knows he isn't the expert he needs to be to justify that trip!!

[bumble]

Great thread topic - thanks Simon!  I have a lot of interest but little experience so look forward to slurping the accumulated wisdom here

Can I suggest looking at the MegaSquirt sites - do a search.  No one will accuse them of being leading-edge but everything they do is open - hardware, software, the lot and provides an insight into the inner workings as opposed to the purely external tuning side of things.

Mark

[Redracn]

Unless 10/10's racing i.e. you are their to win, stay with the standard unit whenever possible. Only after doing all other mods such as suspension, tyres, car setup, driver training etc or there is an absolute must have feature/ability only available on a replacement ECU should you consider one. Generally as far as lap times the bang for buck is very low.

I would generally only recomend an aftermarket ECU for a ground up outright race car build with entire replacement loom. 

I am no longer involved in the aftermarket ECU or tuning industries but have designed hardware, programmed and installed aftermarket ECUs in just about everything.

[SimonN]

Subject 1 - Chips

For many, modifying the existing chip is the first and often only step in gaining more power and improved driveability across the rev range. I would start by saying that IMO, this is a good, value for money route to a better driving experience. This route is available for all who have Bosch Motronic DME, which is the 3.2's through to 993's.

In short, Porsche mapped their ecu's very conservatively and if you look at the maps produced over time, never seemed to be able to decide what the best map should be. Some of this was because of poor fuel quality in some of the markets they sold to, and I believe it is generally accepted that today, there is a far more consistency in fuel than 20+ years ago. In addition, we saw the move from leaded to unleaded fuel, which also led to conservative initial maps. This means that there are gains to be had from remapping.

Over the years, there have been a number of companies that have produced such chips. I have no personal experience of them, but it seems that now, Steve Wong is the "gold standard" with a range of chips to suit different engines, fuels and mild mods such as improved exhaust systems. For many, this is as far as you need to go, because it gives you a worthwhile lift without breaking the bank. I think that some expect too much from these chips, but in the UIK, I have seen regular reports that on 3.2's, with the right SSI headers, you see something like 250 bhp. This is at the top end, which you don't use so much on the road, but there is a nice improvement in the mid range. This approach is not going to make your stock 3.2 into some sort of performance hot rod, but it should lead to an improved driving experience and reduced fuel consumption. Yes, you read that correctly. In most cases, you get improved fuel consumption for the same driving style.

I believe that if you have a later, reduced power 3.2c, there are more gains to be had in relation to your lower base figure.

To take this one or two stages further, there is the opportunity to get a chip that is specific to your car. Wong provides this service whereby you provide him with data (air/fuel ratio) you gather on a rolling road and he then makes a map for your car. This can be done for cars with a wider range of modifications than his "standard" chips (more later). Finally, you can get your car "live mapped" which will get you the most improvement. Of course, unless you are in the USA, you can't get Wong to do that for you, but there are people here in Oz who do this (I only know of PR Technology, but I am sure there are others). These live remaps are particularly good if you have made a fair number of changes to your engine. It is possible to run with a DME up to a reasonable level of tune, including larger engines, upgraded cams etc. However, IMO, you aren't going to unleash the full potential for a number of reasons that will hopefully be explained later in the thread. In short, those limitations are due to batch firing of injectors, how the air is measured entering the engine, other sensor inaccuracies and how the map itself is constructed. 

[bumble]

Simon, for the more or less completely uninitiated, you might want to provide a quick overview of what a 'map' actually is and what it does...

[Redracn]

Tuning a stock NA engine will only yeild as much as the manufacturer gave up in preference for reliability. You are just tuning out the saftey margin. Due to manufacturing variability some engines in combination with ECU and sensors will have less margin than others so a custom tune is always the safest route as generic tunes may still be too safe on some engines and dangerious on another "so called" identical engine. Rule 1) No two engines are identical.

When tuning a multi cylinder engine you are in effect tuning to the cylinder with the least margin (which can change with RPM). This can be changed by either trying to get all cylinders to match more closely which is often difficult or tune each cylinder individually.  Given manifold and exhaust nuances especially on OEM parts there can be good gains by putting an injector with more flow in a better flowing cylinder along with an ignition timing adjustment for that cylinder. It's details like this that mean not all tuners are equal. 

Fuel quality is critical especially on ECU without knock control. So a critically tuned engine must always run on the fuel it was tuned on. You can not just use a higher octane as the fuel may have a different density and the injector injects volume but it is the fuel mass injected that matters.

In my experience Porsche tuned closer to the limit with less variation between cylinders than most other manufactures.

[SimonN]

Unless 10/10's racing i.e. you are their to win, stay with the standard unit whenever possible. Only after doing all other mods such as suspension, tyres, car setup, driver training etc or there is an absolute must have feature/ability only available on a replacement ECU should you consider one. Generally as far as lap times the bang for buck is very low.

I would generally only recomend an aftermarket ECU for a ground up outright race car build with entire replacement loom. 

I am no longer involved in the aftermarket ECU or tuning industries but have designed hardware, programmed and installed aftermarket ECUs in just about everything.

I understand your viewpoint but I think there are bigger issues than you raise. To start with, we are not talking about lap times. We are talking about engine management. In this discussion, things are pretty black and white. What power does an engine make with what modifications? Whether you have the skills to take advantage of that power is another matter.

So why do I have a different view? Let's take a ROW 3.2 Carrera. Most conventional wisdom suggests that with Motronic remapping (chip or live) and a decent header/exhaust system, you can get something like 255 bhp. If you change that car to Motec, I know people who have achieved 275 bhp and would argue that the throttle response and mid range power is a big improvement. Some stop the discussion there, but if, as i suspect, you know a lot more than most, you would know that only scratches the surface because we aren't comparing apples with apples, due to the change from the standard air flow meter the MAP sensor. This cleans up the airflow a lot, and allows for better mapping. To get closer to a fair comparison, we need to see a 3.2 with a hot film conversion, running on Motronics. I have never seen any figures from that configuration. In fact, to really compare apples with apples, we should run the hot film with both Motronics and Motec, seeing that you can run MAF on Motec. My belief is that you will get a better driver experience with the Motec - more power across the range and better throttle response, but it comes at a price.

Or take the 964. It's pretty well established that running a decat, cup pipe, hot film and remapping the Motronics, you will get between 300 and 310 bhp. Take the same car and fit Motec, I have seen 325bhp, not just the graphs but I have been there when the car has been dynoed and have driven it. You will have to trust me when I say that the extra power felt like a lot more, because of the shape of the curve and the throttle response, particularly lower down.

Other key factors come into play when people are building special engines, as are a number on this forum. Over here in Oz, the default seems to be to go for ITB's and something like PMO's, which is the easy option because many can easily learn how to set them up. Very few, if any, are building these new engines with Motronics. Some don't have Motronics to start with, so what should they do. For those who do have it, some choose to go for a mappable ecu (Motec, MS, etc) because they believe it is easier for them to DIY, which i think has some truth to it. When I was learning to remap the Motronics, I had almost zero resources to turn to, while for most programmable ECU's, there is a lot of support. I knew how to remap Motronics and even knew, in theory, how to remap for hot film (never did one). However, say on the 3.2, I don't know anything about moving from single to twin spark, which combined with the hot film, leaves me with either not doing it myself or going to a programmable ecu which i do understand and can get support for. And as i am doing it for my own pleasure, going elsewhere seems pointless.

What I would like to achieve is to get as many real life examples of gains from all the ways of upgrading the engine management so people can be informed

[tomo]

From Colin Ninemeister UK   :

The standard 964 injection system uses a barn door type volumetric air flow meter and two throttle switches (idle & WOT or Wide Open Throttle). The ecu runs on three basic maps: idle, WOT & part throttle where it uses the signal from the AFM. The problem with this set up is that the AFM is slow to resond and is not calibrated for the full scale of air flow demand (it opens fully at approximately 4000rpm & above 70% throttle) and therefore without accurate re-mapping of all 3 maps the fuel supplied does not perfectly match the engine demand. Hot film is another name for MAF or Mass Air Flow, in the case of these conversions the MAF unit is a hot film device (rather than the 928/968 hot wire) which cools down as the air flows over it, so giving a calibrated voltage to the ecu depending on the mass flow of air into the engine (obviously). The range of the MAF allows the ecu to run on one main map which covers the full scale of operation from idle to redline and from zero to full throttle. In theory once the MAF is calibrated (chipped) the mass air flow and fuel supply will be a constant relationship and therefore the ecu should supply optimum fuelling at all times. This is the system fitted to the 993 and later Porsches as standard. Alpha-N is the mapping strategy adopted by most race engine tuners using aftermarket ecu's. Alpha refers to throttle angle and N refers to RPM, thus ecu's running this strategy ignore air flow and instead fuel flow is calculated solely by a 3D map of the throttle position and RPM of the engine. There are usually additional sensors to apply corrections & aid the accuracy of the fuelling, like Manifold Absolute Pressure MAP, Barometric pressure (BAP) and air temperature. Generally Alpha-N mode gives an engine the fastest possible throttle response and track driveability simply because the ecu is looking at the same input that the driver controls, i.e. throttle position. Motec are just one manufacturer of aftermarket ecu's, but in the case of the 964 the term applies to a full conversion package that covers the ecu, correction sensors and larger fuel injectors (I believe that we pioneered the conversion on RS models in the UK, no doubt that many others now offer similar conversions). We found that the standard fuel injectors at the standard fuel pressure maxed out their flow at 300bhp, therefore the larger injectors were fitted in order to accurately time a more dense shot of fuel into the cylinder.  In my experience I have found little to choose between a 964 properly mapped with the standard AFM and one fitted with a MAF, whereas with a Motec based Alpha-N conversions we see appreciable gains undoubtedly because of the improvement in fuelling accuracy. Last edited by NineMeister; 09-22-2005 at 05:55 PM.

09-22-2005, 10:21 PM

[SimonN]

Simon, for the more or less completely uninitiated, you might want to provide a quick overview of what a 'map' actually is and what it does...

It's a fancy way of saying the lines of code which are processed by the ecu. For any given set of conditions, as measured by whatever sensors a system has, there is an instruction or set of instructions which determine the input parameters for the engine, such as the amount of fuel, when it is introduced and when the spark happens.

Besides being code that issues instructions, the map actually consists of a series of tables which gives a "value" for the parameters needed to control the engine, such as air/fuel ratio. On a computer, it loooks like this

What you are seeing here is part of the map for a 996TT. On the left is the menu for all the tables and other variables. You can see 2 tables which look like spreadsheets full of numbers, which is what they are, and below is a graphical representation of those numbers. Each cell is target data point - the larger set of numbers is the timing based on load vs RPM. As you can see, there are a lot of numbers, or values and the bad news is that you need a value in each box. The good news is that the software can extrapolate between points. The mapper then looks at the results and adjusts accordingly. The graphical representation can give you a quick way of pinpointing something out of place. It looks very complex, but once you understand the basics of what you are targeting and you know the limits either end of the scale, the software can do much of the rest.

From Colin Ninemeister UK   :

The standard 964 injection system uses a barn door type volumetric air flow meter and two throttle switches (idle & WOT or Wide Open Throttle). The ecu runs on three basic maps: idle, WOT & part throttle where it uses the signal from the AFM. The problem with this set up is that the AFM is slow to resond and is not calibrated for the full scale of air flow demand (it opens fully at approximately 4000rpm & above 70% throttle) and therefore without accurate re-mapping of all 3 maps the fuel supplied does not perfectly match the engine demand. Hot film is another name for MAF or Mass Air Flow, in the case of these conversions the MAF unit is a hot film device (rather than the 928/968 hot wire) which cools down as the air flows over it, so giving a calibrated voltage to the ecu depending on the mass flow of air into the engine (obviously). The range of the MAF allows the ecu to run on one main map which covers the full scale of operation from idle to redline and from zero to full throttle. In theory once the MAF is calibrated (chipped) the mass air flow and fuel supply will be a constant relationship and therefore the ecu should supply optimum fuelling at all times. This is the system fitted to the 993 and later Porsches as standard. Alpha-N is the mapping strategy adopted by most race engine tuners using aftermarket ecu's. Alpha refers to throttle angle and N refers to RPM, thus ecu's running this strategy ignore air flow and instead fuel flow is calculated solely by a 3D map of the throttle position and RPM of the engine. There are usually additional sensors to apply corrections & aid the accuracy of the fuelling, like Manifold Absolute Pressure MAP, Barometric pressure (BAP) and air temperature. Generally Alpha-N mode gives an engine the fastest possible throttle response and track driveability simply because the ecu is looking at the same input that the driver controls, i.e. throttle position. Motec are just one manufacturer of aftermarket ecu's, but in the case of the 964 the term applies to a full conversion package that covers the ecu, correction sensors and larger fuel injectors (I believe that we pioneered the conversion on RS models in the UK, no doubt that many others now offer similar conversions). We found that the standard fuel injectors at the standard fuel pressure maxed out their flow at 300bhp, therefore the larger injectors were fitted in order to accurately time a more dense shot of fuel into the cylinder.  In my experience I have found little to choose between a 964 properly mapped with the standard AFM and one fitted with a MAF, whereas with a Motec based Alpha-N conversions we see appreciable gains undoubtedly because of the improvement in fuelling accuracy. Last edited by NineMeister; 09-22-2005 at 05:55 PM.

09-22-2005, 10:21 PM

How old is that? Colin and I had a long running disagreement on the topic of AFM vs MAF to the extent that I went to his place and we tested my car vs the best AFM car.  My car, with MAF, made 309 bhp and the best AFM car made 301 on the day. The guy who did my car now maps at Colin's place (and still does MAF conversions. The one thing I totally agree with Colin on, and why I will constantly recommend him, is that with his motec conversions, we are talking significant gains and a very different driving experience.

Edit to say I just saw the date, being 2005. I went to his place in April 2006, on the way to Alton Park for one of the best track days ever, where Colin beat everybody up in his beast of a 993 track car which was producing something like 400 bhp (or was it 425?)

[Redracn]

Hi Simon

While I understand you want to keep this about ECUs I do believe that the decision to go down the path of a replacement ECU should be based on what you want to achieve. My point being that generaly for a street based car it should be well down the list of must have.

To some degree you answer your own question in that you are not comparing apples to apples. A motec will not make more power or torque than a tuned standard ECU given the motec is plugged into the same OEM loom and no other changes whatsoever. Removing the flowmeter does change the volumetric effeciency of the engine as a whole. Some improvement in mid range torque could be expected if you use sequential injection but this requires addition of a cam shaft sensor in your example along with rewiring. You could then individual cylinder tune for further improvements. 

MAP sensors are not prefered on highly tuned NA engines as they pulse to much and removal of the pulsing slows down response. Also they do not measure the effective pressure being delivered to the cylinder due to resonances in the induction system. 

Since both the fuel and ignition are calculated from sensor inputs any jittering of the sensor values causes a jittering/scatter in the applied fuel and ignition advance meaning that you need more saftey in the tune so the most advanced spark is still safe. This is one area aftermarket ECU will differ. 

To make the most power the ECU only has to do the following. (Excluding cam timing etc)

Supply the optimum amount of fuel at the optimum time in the induction cycle and ignite it at the optimium crank angle for each cylinder.

 

[tomo]

It's a fancy way of saying the lines of code which are processed by the ecu. For any given set of conditions, as measured by whatever sensors a system has, there is an instruction or set of instructions which determine the input parameters for the engine, such as the amount of fuel, when it is introduced and when the spark happens.

Besides being code that issues instructions, the map actually consists of a series of tables which gives a "value" for the parameters needed to control the engine, such as air/fuel ratio. On a computer, it loooks like this

What you are seeing here is part of the map for a 996TT. On the left is the menu for all the tables and other variables. You can see 2 tables which look like spreadsheets full of numbers, which is what they are, and below is a graphical representation of those numbers. Each cell is target data point - the larger set of numbers is the timing based on load vs RPM. As you can see, there are a lot of numbers, or values and the bad news is that you need a value in each box. The good news is that the software can extrapolate between points. The mapper then looks at the results and adjusts accordingly. The graphical representation can give you a quick way of pinpointing something out of place. It looks very complex, but once you understand the basics of what you are targeting and you know the limits either end of the scale, the software can do much of the rest.

How old is that? Colin and I had a long running disagreement on the topic of AFM vs MAF to the extent that I went to his place and we tested my car vs the best AFM car.  My car, with MAF, made 309 bhp and the best AFM car made 301 on the day. The guy who did my car now maps at Colin's place (and still does MAF conversions. The one thing I totally agree with Colin on, and why I will constantly recommend him, is that with his motec conversions, we are talking significant gains and a very different driving experience.

Edit to say I just saw the date, being 2005. I went to his place in April 2006, on the way to Alton Park for one of the best track days ever, where Colin beat everybody up in his beast of a 993 track car which was producing something like 400 bhp (or was it 425?)

Yes Mate from the 964 /993 years , About the same time he built the black 993 RSR.

[SimonN]

Although slightly off subject, but I think the black RSR was more like 2008/9. In 2005, he was running his 993RS, which was a total weapon

It had a 3.8 engine that he built with some very special parts and it had Motec engine management. Early iterations saw something like 415bhp and that was with standard heat exchangers and a cat! I think the final iteration ended up at 450 bhp. A key point worth noting is that from 415 to 450 bhp, which included changes to cams, lifters, custom headers and custom intake, he kept exactly the same engine management system. The car made my car look slow, despite most thinking mine was fairly quick!  Seeing him peddle it around Oulton Park (note I spelt it right this time!) or on hill climbs was something to behold. And the noise!

 

[MFX]

So this is where I am at. I am building an interesting engine. Using a 2.7 case, the current thinking is we are trying to source larger custom P's & C's to in crease the bore (not sure at this stage what the final capacity should be). Looking at using a 3.2 crank with titanium rods from a GT3, and twin plugging. That is the plan anyway. I still have to work out ignition although the guys doing the machining swear by M&W for the ignition. I need to work out what I do for the management and inlet manifold for that matter. I was thinking of making up something custom myself, either butchering the CIS or welding up a custom plenum.

Obviously in my case I need to go down the aftermarket management route (and I am not sure on what one I will use as yet), but I am wondering, as I am starting from scratch what sensor system I use? MAP? MAF? Hotfilm? Basically I am still on a conservative budget (I am building the base engine with A LOT of help), and this is going to be a street car with occasional track use.

[SimonN]

MAP sensors are not prefered on highly tuned NA engines as they pulse to much and removal of the pulsing slows down response. Also they do not measure the effective pressure being delivered to the cylinder due to resonances in the induction system. 

This comment confused me and was totally unexpected, so before replying, i thought i should do my homework. I checked what  2 different shops, one here in Oz and one in the UK, do in this regard Both I would consider among the leaders at producing high powered, NA Porsche engines and have been doing so for close to 20 years each. Both use Motec as their preferred engine management system, but both also live map Motronic DME's where appropriate, so they don't have to do the replacement ECU route. In both cases, they use MAP sensors and both say that is an essential part of their conversion. Because of this, i have always believed that MAP was the way to go. Please can you provide me with some links to evidence on this subject so i can understand it better. Maybe I am drawing wrong conclusions or am missing some important part of what you or they are saying.

MFX

I need to write a lot more base stuff which should feed into your situation. It takes a while because I want to make sure I research everything properly, to make sure I don't give totally wrong info because that will not be a good basis for discussion. In the course of this, I am also asking questions elsewhere and reading a load of stuff I haven't looked at for years. However, it is ime consuming and I also need to do some normal work! I am sure the thread will keep moving over the next few days. I am sure with a mix of my input, comments and questions, plus informed comments from people like Redracn, a picture will develop. In some ways, i am in a similar position to you, needing to make a decision on engine management. The main difference is I actually have the option of doing nothing, which in your case, would lead to a very expensive garage ornament! 

[clutch-monkey]

This comment confused me and was totally unexpected, so before replying, i thought i should do my homework. I checked what  2 different shops, one here in Oz and one in the UK, do in this regard Both I would consider among the leaders at producing high powered, NA Porsche engines and have been doing so for close to 20 years each. Both use Motec as their preferred engine management system, but both also live map Motronic DME's where appropriate, so they don't have to do the replacement ECU route. In both cases, they use MAP sensors and both say that is an essential part of their conversion. Because of this, i have always believed that MAP was the way to go. Please can you provide me with some links to evidence on this subject so i can understand it better. Maybe I am drawing wrong conclusions or am missing some important part of what you or they are saying.

would also like to know this, as in the commodore world maf-less tunes are the first upgrade for most when tuning Ls engines

[Redracn]

What I said relates to NA engines (turbos are another story) and more so to tuned engines that have resonant intake systems.  Simply put there is a lot of pulsation going on in the inlet manifold which the MAP sensor will respond to. Just locating a spot to attach the MAP sensor is a big challenge.  Given the fueling and ignition are directly dependant on the reading from the MAP sensor they will also pulsate which means you can not acheive the best possible tune. You can of course compromise and filter the pulsation to be acceptable while trading off response. In doing this you will place more reliance on transient fueling to correct the error while the MAP signal catches up. Any tube to the sensor acts as a signal filter and in many cases a small restrictor hole is used at the manifold. In many applications where the throttle position sensor is used for mapping a MAP sensor is still used but as a Barrametric Pressure sensor BAP.  Another issue is the ECU sampling point in the engine cycle. Most ECU use a fixed rate rather than actual crank position which is way to slow to update at low RPM. 

Aftermarket MAP based load sensing was used a lot on older cars simply because it is easy to fit. Adapting a Throttle Position sensor to a TB not designed for it is not easy so adds to the cost as well as the TP sensor itself. 

If fitting a remote MAP sensor always make sure it is mounted above the plenum take off point with its nipple facing down and the hose has a continious down slop from the sensor to the plenum. This also applies if the aftermarket ECU has an internal MAP sensor which a lot did/do. The reason? Water condensation forms in the tube and needs to drain out along with oil from the PCV system. The problem is that water/oil has mass and so damps the MAP reading with the amount depending on the quantity in the tube. It will also cause a higher pressure reading  which changes the tuning and can affect some MAP sensors internals reducing reliability. 

Keep searching I am sure there is more out there on the downside/problems of MAP based load sensing. 

[TrevMcRev]

I have a Motec in my Lancer rally car, but you guys might as well be speaking chinese to me!

I'm more usually interested in hanging skids than how it works, so as long as it works I just drive it!

Incredible knowledge though guys....carry on, Im sure to learn something!

快速车计算机

 

 

[bumble]

So this is where I am at. I am building an interesting engine. Using a 2.7 case, the current thinking is we are trying to source larger custom P's & C's to in crease the bore (not sure at this stage what the final capacity should be). Looking at using a 3.2 crank with titanium rods from a GT3, and twin plugging. That is the plan anyway. I still have to work out ignition although the guys doing the machining swear by M&W for the ignition. I need to work out what I do for the management and inlet manifold for that matter. I was thinking of making up something custom myself, either butchering the CIS or welding up a custom plenum.

Obviously in my case I need to go down the aftermarket management route (and I am not sure on what one I will use as yet), but I am wondering, as I am starting from scratch what sensor system I use? MAP? MAF? Hotfilm? Basically I am still on a conservative budget (I am building the base engine with A LOT of help), and this is going to be a street car with occasional track use.

A 74.4mm 9 bolt crank in a mag case???  Even if it can be done, (which would surprise me) I wouldn't.

Max 'sensible' bore for a mag case is 92mm.  93 is getting awfully thin...

[LeeM]

 Fit carbs. Sorted

[MFX]

A 74.4mm 9 bolt crank in a mag case???  Even if it can be done, (which would surprise me) I wouldn't.

Max 'sensible' bore for a mag case is 92mm.  93 is getting awfully thin...

I have no idea to be honest. The guys at Performance Developments build a lot of crazy race engines for Porsches (and others), so I am trusting their judgement a lot. I know they do a lot of work that others say can't be done, like repairing damaged Mag cases, etc, where numbers matching makes it worth while. I will mention it to them to see what they think?  

At this stage we are still gathering parts and building it on paper before any final decisions, but it will be in the next few weeks.

[SimonN]

 Fit carbs. Sorted

If you want to leave a load of power on the table and have high fuel consumption!

[Redracn]

If you want to leave a load of power on the table and have high fuel consumption!

Add crap drivability as well.

 

[clutch-monkey]

If you want to leave a load of power on the table and have high fuel consumption!

to be fair peak power there would be nothing in it, on the old V8's the upstream mixture and slight charge cooling effect leaves not much on the table compared to port injection throwing fuel at the back of the valve.  

 

...just everywhere else the efi is more efficient flexible lol. 

Knight engines in South Australia work a lot with both systems, of note would be a monster Datsun l20b displacing 2.4 litres putting out close to 250hp with efi, and Bmw m20 displacing 2.8 litres putting out a mental 400+HP on carbs (on e85). The latter is a front runner on many historic races. Well worth looking up for headwork or engine building. Hell. Built a 27 litre merlin engine, and put two turbos on it for a tractor puller. nobody knows what hp it makes, because nobody has a dyno rated for it apparently.

[SimonN]

Redracn

I have done my "homework" and now understand what at least one of these tuners is doing. While they do use MAP, it is secondary. In effect, they are using a hybrid Alpha-N set up, using the MAP for correction to achieve things like stable fuel mixture under idle control and corrections for altitude and natural air pressure variation. In short, and this is my take on it, this makes the car more driveable in street conditions.

I can tell you that this certainly works and works very well because the particular tuner i am talking to is known for just how good his stuff behaves at idle and off throttle. This is particularly noticeable with lightweight flywheels. 

The question that some will have to consider is whether it is worthwhile for their needs here in Oz. For instance, I am not sure that altitude correction is a huge priority and many put up with less than perfect idle, so long as you aren't stalling all the time. But as this tuner is producing very top end, mixed use performance engines and prides himself on the overall driveability, it is worth the trouble for him.

I guess that I should just say something about Alpha-N before the questions start! In simple terms, it is mapping TPS vs RPM where TPS is throttle position sensor. 

I also found out some more about the standard 3.2 engine with Motec. I was mistaken, because that engine only used Alpha-N, so in that respect, it was comparing apples with apples. The car had B&B headers, stock injectors and the only other upgrade was running the fuel rail pressure at 3.8 bar. That car made 270 bhp, vs the best seen on a Motronic remap is 255bhp. In addition, the curve was far nicer and throttle response was better.

 

[SteamWolf]

like any purchase decision, it should be based on requirement.  An aftermarket ECU isn't needed on a road car with minor mods unless changing from carby.

Stock engine management is going to be unlikely to be able to be tuned for anything other thas minor fuel tweaks, and earlier stuff (90's and earlier) probably uses very VERY outdated and limiting technology such as air flow meters (particularly those horrible flappy gate units).  Aftermarket is good here.

If going aftermarket, list your requirements.  Most people don't need anti lag and all the other fancy hard core stuff. Plenty of cooking level units are affordable and do everything that is needed.

Megasquirt used to be good when it was a kit as you could build it for about $300. Now it's a full deal and costs about the same as an entry level Haltech.