John

I guess the proof is in the pudding. The Motec is proven but costs 10 times more - let's see what the Unichip does or doesn't do.

Christer

 

Dear John,
The guys with superchargers and the required supercharger management systems run into these problems as you write about all the time. It is almost impossible to help them once these systems are installed. I agree totally. For road use just work on remapping the existing ECU/DME,
Ciao,
Adrian
911C4

 

[quote]Originally posted by Adrian:
<strong>Dear Steve,
Would you expand a little more on your statement about the 964 ECU "other functions".
Your statement about replacing them is correct but I feel the explanation why is a little vague. I would like to know what you mean exactly before answering in more detail.
Ciao,
Adrian
911C4</strong><hr></blockquote>

Adrian,

You know a lot more than me about what goes on inside the 964. All I know is that the DME does more than just the ignition and fuel sprat times. These other functions still need to be performed even if you are having another ECU doing the engine stuff. The UNICHIP has about 8 connections which go between the engine/senors/injectors etc and the original ECU. The UNICHIP allows you to modify the signals between the existing ECU and the engine to alter their timings. If there is no value in the UNICHIP for a particular engine load and rpm then the default ECU timing are passed through.

This is all a bit advanced for me, I miss the days of 45 DCOE Webers and points <img src="confused.gif" border="0">

 

John,

I disagree with you about the difficulty of mapping an aftermarket ECU. A rolling road like a DynoPack dyno will hold the car at a specific RPM regardless of the throttle position (engine load). This allows for mapping a specific load site whether you are mapping TP vs RPM or KPA vs RPM. Motec and Haltech are very reliable and flexible. They contain parameters to map overrun, accel enrichment, air temp compensation, MAP compensation, etc. For a point of comparision, an 84-89 carrera has a 12x12 mapping grid, the Haltech has a 16x32 grid and Motec has 21x40 grid to map fuel and ignition.

I have a MoTec M48Pro installed in my 930 and there are NO drivability issues and the car successfully starts at-20 C. I've also installed an Electromotive Tec II in a friend's Ferrari and had it running perfectly after about 5 days.

Don't kid yourself, a Porsche DME from an EFI Carrera or 964 is much less powerful than either Haltech or MoTec.

 

[quote]As I understand it the more processing power the more mapping points you can have and thus the better the map is. Better map == more power + better fuel comsumption.

If the ECU did not make any difference MOTEC etc would be out of business.
<hr></blockquote>

Good point. <img src="graemlins/c.gif" border="0" alt="[ouch]" />
Does the UniChip add mapping points though? I thought it just modified the signals going into the stock ECU to change the fuel and ignition curves.

 

Hi Geoffrey,

My knowledge of rolling roads is limited, from your description I guess the dyno you mention is a brake type rather than inertia so that would certainly make the mapping a lot easier, but you still have problens getting correct mapping at small throttle openings because the dyno's inertia masks the response of the engine (this is a problem even on engine dynos which generally have less inertia than rolling roads).

I know the Motec and similar systems will offer the usual complement of transient fuelling corrections, the difficult part is getting the parameters set up correctly, which cannot be done properly on the dyno (inertia issue again) and is difficult and time consuming even when you have a race track to test on and all the data acquisition and individual cylinder wideband lambda probes you could want (in my experience).

I have no illusions about the power (or lack thereof) of stock Porsche DME - most people would be shocked if they knew how little money the car manufacturer is prepared to give the DME manufacturer for these units, they are designed to very tight budgets - in contrast, the kinds of control units with which I am directly familiar cost more than my 993 , so I have a good understanding of what can be achieved, but getting a more powerful unit set up for a road car from scratch is a big job for anyone.

Your 930 sounds like it could be quite a beast - what sort of power does it produce? Courtesy of 930 Motorsport?

 

[quote]Originally posted by Geoffrey:
<strong>John,

For a point of comparision, an 84-89 carrera has a 12x12 mapping grid, the Haltech has a 16x32 grid and Motec has 21x40 grid to map fuel and ignition.

</strong><hr></blockquote>

I have just checked and the UNICHIP has 17x12 sites for fuel and ignition giving a total of 408 mapping points. Not quite up to MOTEC standards but better than the original bit of kit.

 

Steve,

I think the Unichip is a unique response to the way ECUs are built today. As mentioned earlier, the latest version of Motronic controls so much about how the car runs, PSM, ABS, emissions, etc. I'm not sure how viable (I'm sure it's possible) to remove and replace a later car's ecu with something like MoTec. In general, a street oriented customer wants some bolt on goodies to improve the performance of their car, but retain emissions compliance. An engine is usually tuned for a primary objective - fuel economy, emissions, or performance and cannot acheive all three. An engine being just an air pump, will need additional fuel and tuning to achieve increased performance. The Unichip allows for this fine tuning while retaining the stock ECU - Cool!

Your observations about race cars vs street cars is right on. However, the aftermarket ECUs are flexible enough to work with a street driven application.

The older cars are simple and if you really want to be a DYI, you can download the ECU code of a Motronic with an eeprom reader, look at the hex code, change it, and burn your own eeprom for your specifc application. This is basically exactly what you are doing with an aftermarket engine management system. I believe there is even a program for sale for $900 to read and change the download file from the chip.

I find it funny that chip manufactures say a 35hp gain by adding their chip. ALL they can do is increase fuel and/or increase timing which in itself can't make that kind of gain. However, if the engine breaths better (exhaust, intake, air flow sensor) then it will require more fuel and possibly be able to run additional timing and then the chip might make that much difference.

17x12 gives you 204 load sites which is in my opinion on the low side. The Electromotive I installed in my friend's ferrari had an 8x8 grid which I think is way too low, but the car runs great so go figure. I would think that you would want to be able to map at least every 500 RPM and 250 is better. That means that you need at least 16 speed sites for a car that runs to 7000 redline giving yourself some room for revlimit and overrev to 8000. I would think that you want at least 11 load sites 0-100% TP or 0-100KPA for naturally aspirated cars.

 

[quote]Originally posted by Geoffrey:
<strong>Steve,

I would think that you would want to be able to map at least every 500 RPM and 250 is better. That means that you need at least 16 speed sites for a car that runs to 7000 redline giving yourself some room for revlimit and overrev to 8000. I would think that you want at least 11 load sites 0-100% TP or 0-100KPA for naturally aspirated cars.</strong><hr></blockquote>

The UNICHIP is about right then 17 sites 500-8000 RPM and 12 load sites. I am still not decided though. The UNICHIP dealers I have found do not have much if any Porsche knowledge. Here in the UK is a company called AMD who take the conventional chip approach but tweek each chip to car on the dyno. More importantly they know about Porsche and they owner runs a 964 himself.

I am not after ultimate power so I think I will go the safe route and stick with a conventional chip from AMD plus save some money. Well not actually save any just spend it on something for the Porsche

 

Steve,

I'll be at AmD on Friday getting the map tweaked on my 993, I'll let you know how it goes.

 

John,

You'd be suprised by the new fluid dynos, they are very accurate and will hold a car within 1% of the desired value. Mapping really isn't all that difficult, especially with a new wide band lambda sensor connected to the dyno to help you with A/F ratio. There are some standards to start with, for instance for every 10 degrees C change in air temp, you need approximately 3% less fuel. For turbo applications, every 10Kpa in pressure beyond 1bar absolute, you need approximately 10% additional fuel. You want to increase timing and fuel during cold start and ramp down until the engine is at 40 or 50 degrees C. On a Porsche 911 engine, 28 degrees advanced is about the most you will want to run. You really need a dyno to do it properly and the lates ones are the best. Additionally, there will be some things to get straight that only driving it will solve, things like the cold start where you get only 1 shot per day.

My car put out 521hp on an engine dyno when initially tuned and after 15,000 miles puts out 443hp on a Dynopack chassis dyno - tested last month with no air moving across the intercooler.

 

Hi Geoffrey,

My mapping experience has been solely on racing engines, but we always had to do the final tweaking at the track, especially for transient fuelling, even after having complemented the engine dyno work with further sessions on combined engine & transmission dynos where we can control the dyno to accurately simulate the details of any particular racetrack.

Your car must be great fun to drive.

 

John,

My car is a joy to drive. I like messing with the data logging of MoTec. Right now I'm working on connecting up traction control and working on a 3 axis acclerometer to draw a track map. In my last DE there was a guy with a new 996TT and it was fun passing him like he was standing still.

 

Geoof & John

Absolutely brilliant to have two people on the forum who have experience of DME & MoTec use on the P cars as well as dyno & race tuning experience. How satisfying to have confirmation that engines are air pumps & chips alone won't add 35 HP without other improvements in engine breathing.


I looked into after market engine management with a company int the UK for my 964 - they politely said not to bother with a street car, which I thought wa reassuring, as most chip-change merchants would have gladly lightened my wallet. These guys said if I didn't intend to replace throttel bodies, increase inlet valve size, change exhaust etc etc a new management system wouldn't reap great benefits - hurrah for science!!

 

Dear Steve,
Sorry I am a bit late back on your reply to my question of you. I see what you are getting at now and you are quite correct. One thing that cannot be duplicated unless the entire sotware contents are downloaded somehow and then loaded up into an aftermarket unit is the adaptive system. All the factory maps from the lab and the test track. I also miss the old days of carbs and points. I still have some old dual Weber 40s lying around somewhere.
I go back to my research because this thread is repeating what we already know. Regardless of the system offered the truth is that for every genuine 1HP increase on the rear wheels of a 964 3.6L engine, it is going to cost you circa $US208. The Porsche engine kits circa 400 Euros/HP.
Ciao,
Adrian
911C4