I have an 85 930 which I would like to convert to EFI. I recently otained a 3.2 intake with fuel rails fuel lines. For those of you who have converted to EFI (or are thinking of converting to EFI) what ECU are you using or planning to use? I have looked at the web information for Motec, Autronic, DTA and AEM. Clewett Engineering has a setup using the TEC-GT. Any information on experiences (good or bad) would be appreciated. Thanks.

 

I would find what the dyno guys near you are running. You can have the best ecu like motec and if your dyno guy has no idea how to program it you are s?it out of luck. I have a tec2 and when I went to my dyno guy he had no idea about it so it was a waste of time. I'm looking to buy DTA since a porsche shop down in south fl dyno them and will be able to trouble shoot any problems. Also if you install a ecu that your shop has never worked with then you may get a high bill on trying to trouble shoot what is wrong with the car. Also did you do a search on this tread?

 

I will second what Kleinbbc2 said, and go a step or two further. You need to find a true expert or two in this -- someone who has done this before, preferably a number of times, so you can obtain good guidance. You need two forms of guidance -- the mechanical/electrical build, and the tuning. The tuner may not be your best consultant on the mechanical build, and vice versa. Make sure your "dyno guy" is someone who has verifiably successful experience tuning Porsche air cooled turbo engines with the specific ECU platform you have chosen. I suggest you scour the web and do extensive searching, looking back several years, on Rennlist and Pelican to get an idea of who on these lists might be worth contacting for their advice or experience.

Most of all, don't choose an ECU unless you've chosen a tuner. The tuner should be your first selection. Almost any ECU, if handled right, will do the job. But any ECU, if tuned by someone with no experience, will cause you great heartache and expense. And don't expect that you'll have the wherewithall to tune it properly yourself the first time.

I chose a DTA ECU, because the tuner was very familiar and comfortable with it. He's tuned other brands, but that's the brand he prefers and is most efficient with. Other tuners have other strengths. Find someone you believe in, preferably someone who's geographically close, and go for it.

Rob
78 930 EFI

 

Coming from someone who looked for an ECU first, then a tuner, the above recommendations are spot on. Choose the turbo tuner first, then seriously consider their system recommendation (notice I said system - sensors, harness, plugs, injectors, etc). Selecting the ECU might only be 1/2 the total price once you add the peripherals, and don't look here to cut costs.

FYI - I chose the Link system from Performance Developments.

Good luck

 

Motec M48 would be our first choice.

 

I agree with the statements above, I chose the Motec M600 and the price of the ECU alone does not tell the whole story, the parts i.e. sensors harness etc can more than double the price.

 

If you go for serious power with big injectors it's better to choose an ECU that has a good injector control circuitry and flyback because otherwise it's going to be hard to get the car to idle and drive at low loads smoothly. AEM has sometimes very serious problems with injectors over 800cc.

But otherwise I'll second the guys above: if you're not familiar with the tuning of a standalone, choose the one that your nearest tuner knows the best but I definitely would stay away from TEC2 on a turbocharged car.

 

The reason why I recommended the M48 over the M600 is that it is one of the most stable hardware platforms that you can buy and the dos based software is rock solid and extremely straightforward to use, so the upside is that once your installer has the car is running nicely, fine tuning could be done by the owner. In contrast, although the M600 is a newer, faster and more capable ecu and with the new version 3 software is probably about as good as it gets in terms of functionality, the real truth is that there are very few Motec Dealers who could get the best from this ecu/software combination, so the average self-tuner is unlikely to be able to benefit greatly from the additional cost and complexity. I suppose the conundrum still boils down to the fact that a cheaper system well mapped is better than an expensive system badly mapped.

 

I second what NineMeister wrote. I have the M600 and while I know my tuner has tuned many engines with the M800, I can tell he's fighting the windows software vs the old M48 DOS stuff.

 

I have spent months with the MoTeC version 3 software it it is very flexible and detailed and contains functions FAR beyond what is required for a street/track environment. I think the sofware is awesome and even the Beta version was extremely stable, however, I agree, the average do-it-yourself tuner may have trouble navigating through. I haven't found any issues with the Version 3 software and have used it on several DBW cars and some with variable camshafts. The increased control makes for a much better mapping than the Version 3 software WHEN USING advanced features. The Version 2 software is simply a Windows version of the DOS M48 software with added functionality to match the new capability of the M48 such as drive-by-wire and variable camshaft but will effectively control the engine nonetheless.

 

If you are going to map /install the system yourself , then pay very close attn, to the ease of installation and use of the operating system along with reliability and functionability and most of all SUPPORT! If you are going to have a specialist install and map then of course this will be of lesser importance to you and what is of more importance is the competence of the person performing the installation and there ability to deliver.

You should first have an idea of where do you want to go for power , drivability , reliabilty and use , street /track or both , this will determine the necessary features you will need to reach your goals as to boost control, fuel type , adjustable maps , hi/low boost , boost protection etc.

 

I have been reading this for several days now and today I was present at a dyno session where EFI was tuned. I thought I would add something here.

I agree about the need to have someone who knows the software and it's use to tune. I also think the system itself is important. If you can get both then your well on your way. I think some systems are well known by name. Some offer good software and control, while others still rely upon their good name. I think some systems need and deserve looking at as they have come on leaps and bounds. These systems are often overlooked in the mix, as they may not have the name some others have. These systems can offer as much for a lot less money. There is nothing wrong with going with the best, but today the best has company.

Today I was present when some engine testing was to done. The test was conducted on a Turbocharged 4 valve V8 engine. They were testing different Camshafts and Intake systems. They were trying to get the best combination of Camshaft, Camshaft timing and Intake runner length/butterfly position. What was interesting was how the EFI was tuned and how it applied to this thread.

The engine was started and a very quick adjustment was made to get it running where the engine could be warmed up. Once it was warm the base map allowed it to be throttled. The dyno was programmed to allow the engine to rev through the RPM range. The EFI software was logging the A/F through the rev range. The dyno logged it as well. The data showed that the map was rich in some places and lean in others. The desired A/F was to be set at 12.5. The map showed what the recorded A/F numbers were. It is this that caught my eye. The number recorded at ceratin zones were 13.7, 11.6 and in some places 14.4, and so on. Instead of turning on the closed loop control, or running the engine and adjusting the fuel, simple math was used to establish what the adjustment to the fuel value should be. The % change was calculated using the 2 A/F numbers. The changes were made very quick and within minutes the map was dead on across the board at 12.5. I asked why he did not use the close loop control. I was told this is way quicker and less time running of the engine. I was told that there are 4 ways to use the Lambda control with this system. Manually where the % is calculated by using the A/F numbers to get it really close very quickly, let the ECU record the A/F numbers and then let the ECU calculate the change required based upon the Lambda look up table, or run the engine in full closed loop. What I learnt today is that the real "closed loop" control is a ECU based correction where the ECU continually makes small changes based upon sensor inputs, and the so called " self tuning" is a front end control or software based correction. Its this correction that makes the big changes. It really showed the importance of a good system and a person who knows how to use it. In this case, the person had a huge part in the development of the software and the system as a whole. I was shown how good the software is as the rough tune was then subjected to the ECU closed loop control. The ECU made minor changes as the Air Temp was changing in the room, but really it made very small changes that resulted in very small Torque number changes. The dyno air was heated and the ECU made several changes but the A/F number stayed at 12.5. The control was amazing. Now for the interesting part. Once the engine was tuned for the test, this took less than 10 mins to get it to this point, the map was set. The engine made good power and the engine was running really well and safe. The changes to the engine were made (Cams etc) and the recorded A/F numbers were logged. The engine was fitted with a water cooled sensor in the Head to record pressure as well as a rotary device on the Crank and some other air monitoring systems. These were not even looked at until the tests was almost complete. The differences in Camshafts etc were calculated using the A/F numbers. If the engine went lean, the amount of extra air was calculated from the A/F number. If it went rich same thing was used. This way they saw where the Cams were better and how much more air the engine was using. The Flow charts from the Flow bench, and the Compressor maps were used as well. Again I thought how important it is to know the system you are using. Once the best combination was found the engine was mapped very precisely, the fuel adjusted to make the best Torque, the Timing adjusted so that it did not knock and the best air volume measured along with the cylinder pressures.

After the tests were finished I was told that a system that allowed for very fast changes to be made that were accurate allowed this test to be conducted quickly, the engine would not deteriorate thus give good accurate results. This made me think about what has been said to date is so true. Good systems with good control and a user who knows how to use it. I saw today just how easy it looks when these conditions are present.

I am sure other systems offer this as well. It is because of this that I have not said what the system is. This is not inportant here. What is, is the system itself and the knowledge of how to use it.

 

Thanks for sharing. I had never thought about using the AFR to show where cams were working good or bad, very clever.
Did the intake use a map sensor ? No air metering ?

 

Hi Simon,

I think you mention some good points. The better ECUs have capability to use a integrated wide band sensor and a Lambda "look up" table. Here you input your desired AFR for the engine and as you are running, the ECU will display the desired AFR and the actual AFR from an integrated wideband sensor. From there you can as you say use a manual method to type in the correct value in the base fuel table, or you can use a key and allow the ECU to automatically do the change for you. Once the engine has been tuned, you can run it in closed loop, but I never use closed loop while tuning, unless I am working on tuning variable camshafts.

You don't mention about how the ignition timing was set which is where the power comes from. Unless you are doing steady state tuning or swept runs with an in cylinder pressure logging and analysis equipement, ignition timing cannot reliably set, and that is where the poer comes from.

 

The Ignition was set using knock sensors. The dyno has a system as does the ECU use this input. The engine had a Ratary device fitted to the Crank which I understand measures Crank deflection or rotation caused by knock. The head was fitted with a special watercooled pressure sensor, but I think this was never used for knock in this case. It was used to log the Cylinder pressure when the engine testing was performed. The ECU mapping was done very quickly. Once the base map was set, it was never changed. This map was used as the base line to check the Camshaft testing. The dyno computer was measuring fuel and air mass to get the A/F number. From this information they could see where the engine used more or less air. The ECU system was so stable that this test could be conducted. That was what impressed me. The ECU was used as a tool to measure Camshaft and air flow.

To add to your comment about the closed loop tuning, I never knew that the closed loop in most cases is a PC controlled function. I learned yesterday that this sytem has both. The ECU system uses the inputs from all of the sensors to make the changes, not just the Lambda sensor. These changes are very small and continue all the time to obtain a steady A/F number within 1 decimal point. It was impressive.

The manual tuning was so quick. I was told it is quicker than using the ECU to do the tuning and less time running the engine. The map displayed the recorded A/F numbers and the math was used to bring the map in very quickly. It was then the table was used and the trim was done. I was told that if the desired A/F number is far from the recorded number, the ECU often takes several changes to get it right. The way I saw it done was just by simple math and the % change done. What was impressive was that thuis quick and dirty way brought the map in dead on and the ECY trim was almost nothing. The only change I saw was because of the air temp in the dyno room air supply. They have a heater system in the air supply and when this was programmed to heat the in coming air by +10 degrees, I saw the fuel numbers change to keep the A/F numbers right at 12.5. Good system and knowledgeable users.