DME CPU?
08-28-2002, 09:16 PM
#1
Advanced
Thread Starter
Join Date: Aug 2002
Location: Ottawa, Ontario
Posts: 78
Likes: 0
Received 0 Likes
on
0 Posts
DME CPU?
Does anyone know anything about the internal of the DME? More specifically, I'm looking to find out why type of cpu is used. I'd really like to take a look at the eeprom code and see what makes the 964 tick.
08-29-2002, 05:58 AM
#2
Instructor
Join Date: Dec 2001
Location: England
Posts: 176
Likes: 0
Received 0 Likes
on
0 Posts
There is very little info on these about. However it is an easy task to lift the DME out of the car and open its case to have a look inside.
Slide the LHS seat right back and there it is. There are four bolts holding it in. It is a good idea to disconnect the battery before unpluging the DME. Once you have it on the bench level up the tabs on the underside and the case pulls apart. The EPROM is a 2764 (I think) so just about any old EPROM reader will work.
Let you know what you find.
Slide the LHS seat right back and there it is. There are four bolts holding it in. It is a good idea to disconnect the battery before unpluging the DME. Once you have it on the bench level up the tabs on the underside and the case pulls apart. The EPROM is a 2764 (I think) so just about any old EPROM reader will work.
Let you know what you find.
08-29-2002, 06:03 AM
#3
Addict
Lead Rennlist
Technical Advisor
Rennlist
Lifetime Member
Lead Rennlist
Technical Advisor
Rennlist
Lifetime Member
Good luck Steve but honestly you would be better off investing your time with something else. At least something which has a happy ending.
Ciao,
Adrian
911C4
Ciao,
Adrian
911C4
08-29-2002, 07:04 AM
#4
Racer
Join Date: May 2002
Location: West London, UK
Posts: 270
Likes: 0
Received 0 Likes
on
0 Posts
Adrian,
I'm intrigued as to why you would say this. Especially as Ruf, etc. make a lot of money from reprogramming eproms. Are you saying that this is beyond a eprom skilled person, or is that it is not possible? As a slight aside, what packages do these tuners use, as the prospect of tuning your own 964 does have an attractive ring to it?
Taj.
I'm intrigued as to why you would say this. Especially as Ruf, etc. make a lot of money from reprogramming eproms. Are you saying that this is beyond a eprom skilled person, or is that it is not possible? As a slight aside, what packages do these tuners use, as the prospect of tuning your own 964 does have an attractive ring to it?
Taj.
08-29-2002, 01:02 PM
#5
Rennlist Member
I think the DME ECU is given way too much credit as a source of horsepower and car owners are being exploited by chip re-programmers. The thought seems to be that it's like changing the CPU in a computer for a faster clock speed and therefore a faster computer - or car.
All the non-turbo 964's CPU does is control fuel mixture and timing-I know rev limit too but why change this unless you want a decontainment of the engine.
The assumption is that the factory has to make horsepower concessions in order to achieve emissions control and fuel economy. I really don't think this is the case in a non-turbo 964, unless you increase the octane of the gas the engine is tuned for. This is only effective to a point, i.e. you can't get 325 horsepower tuning for 118 octane. The Euro 964 RS has a DME that is tuned to 93 octane gas and this is worth 10 DIN horsepower over the 92 octane chip. My car has a Euro RS chip and the car pings a little on tip in to full throttle running 92 octane, so this chip is over the limit for 92 octane. I run a mix of unleaded race gas to get over 93 and the car is fine. Running the fuel mixture a little richer may make more horsepower in some engines, but with the fuel atomization achieved with Bosch fuel injection I don't think this works in the 964. The car is run at stoichiometric proportions with the loop closure of O2 sensing anyway.
In times past, the mid 8o's, the USA was undergoing a conversion from leaded to unleaded gas and the highest octane available was 87 to 89 octane gas in some areas. Manufacturers were limited in the amount of advance that they could dial in because of this limited octane. In areas of the country where 92 octane gas was available horsepower could be made by reprogramming the computer to take advantage of the increased octane increase by advancing the timing. Thus chip tuning was born. Chip tuning's reputation was furthur enhanced by chip tuning of turbocharged cars. In a turbocharged car chip tuning can be carried to a new dimension since turbo boost is controlled by the ECU and can be increased. Increased turbo boost pressure is a real source of power (at the price of reduced engine durability).
So we have a situation where past performance of chip tuning, particularly in turbocharged engines, and the expectations of the "computer generation" meet to provide chip tuners with an exploitable market.
Car and Driver magazine did a test where only the chip, not octane and chip, was changed in several cars. The result was essentially no improvement in most and a negative result in others.
I may be wrong, I'd like to see real dyno plots showing results under controlled conditions and not just hear testimonials from people who think things "feel stronger". I think Adrian is right don't waste your time. <img src="graemlins/soapbox.gif" border="0" alt="[soapbox]" /> <img src="graemlins/icon107.gif" border="0" alt="[icon107]" />
All the non-turbo 964's CPU does is control fuel mixture and timing-I know rev limit too but why change this unless you want a decontainment of the engine.
The assumption is that the factory has to make horsepower concessions in order to achieve emissions control and fuel economy. I really don't think this is the case in a non-turbo 964, unless you increase the octane of the gas the engine is tuned for. This is only effective to a point, i.e. you can't get 325 horsepower tuning for 118 octane. The Euro 964 RS has a DME that is tuned to 93 octane gas and this is worth 10 DIN horsepower over the 92 octane chip. My car has a Euro RS chip and the car pings a little on tip in to full throttle running 92 octane, so this chip is over the limit for 92 octane. I run a mix of unleaded race gas to get over 93 and the car is fine. Running the fuel mixture a little richer may make more horsepower in some engines, but with the fuel atomization achieved with Bosch fuel injection I don't think this works in the 964. The car is run at stoichiometric proportions with the loop closure of O2 sensing anyway.
In times past, the mid 8o's, the USA was undergoing a conversion from leaded to unleaded gas and the highest octane available was 87 to 89 octane gas in some areas. Manufacturers were limited in the amount of advance that they could dial in because of this limited octane. In areas of the country where 92 octane gas was available horsepower could be made by reprogramming the computer to take advantage of the increased octane increase by advancing the timing. Thus chip tuning was born. Chip tuning's reputation was furthur enhanced by chip tuning of turbocharged cars. In a turbocharged car chip tuning can be carried to a new dimension since turbo boost is controlled by the ECU and can be increased. Increased turbo boost pressure is a real source of power (at the price of reduced engine durability).
So we have a situation where past performance of chip tuning, particularly in turbocharged engines, and the expectations of the "computer generation" meet to provide chip tuners with an exploitable market.
Car and Driver magazine did a test where only the chip, not octane and chip, was changed in several cars. The result was essentially no improvement in most and a negative result in others.
I may be wrong, I'd like to see real dyno plots showing results under controlled conditions and not just hear testimonials from people who think things "feel stronger". I think Adrian is right don't waste your time. <img src="graemlins/soapbox.gif" border="0" alt="[soapbox]" /> <img src="graemlins/icon107.gif" border="0" alt="[icon107]" />
Trending Topics
08-29-2002, 07:27 PM
#8
Addict
Rennlist Member
Rennlist Member
Nice work, Cupcar! Thanks for taking the time to share your knowledge. Yes, some may disagree, but I think we all benefit from this type of exchange. I know I do!
<img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />
<img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />
08-30-2002, 04:35 AM
#9
Addict
Lead Rennlist
Technical Advisor
Rennlist
Lifetime Member
Lead Rennlist
Technical Advisor
Rennlist
Lifetime Member
Dear Taj,
You need the address codes. RUF and others paid BIG money to get them. From PAG and Bosch. Mere mortals such as ourselves cannot get access to them. I have been doing this for a long time and many many people tried to crack the codes, make laptop interfaces and change the DME programming, remapping it is called and every one has failed. Some even got so disgusted with the whole process, they sold their Porsches and changed marques. Very few people know really how to do this and they aren´t telling anyone. I happen to know that a RUF optimisation of your engine can give you 288HP. No internal engine parts changed. Remapping of the DME is part of the whole package. For more details please call RUF. This is available in Germany but I do no tknow about elsewhere.
I agree totally with CupCar as well I am on the record for many years saying the same thing.
Ciao,
Adrian
911C4
PS: Check the internet for details on 964 DMEs. You will hardly find anything at all on any Porsche DME system.
You need the address codes. RUF and others paid BIG money to get them. From PAG and Bosch. Mere mortals such as ourselves cannot get access to them. I have been doing this for a long time and many many people tried to crack the codes, make laptop interfaces and change the DME programming, remapping it is called and every one has failed. Some even got so disgusted with the whole process, they sold their Porsches and changed marques. Very few people know really how to do this and they aren´t telling anyone. I happen to know that a RUF optimisation of your engine can give you 288HP. No internal engine parts changed. Remapping of the DME is part of the whole package. For more details please call RUF. This is available in Germany but I do no tknow about elsewhere.
I agree totally with CupCar as well I am on the record for many years saying the same thing.
Ciao,
Adrian
911C4
PS: Check the internet for details on 964 DMEs. You will hardly find anything at all on any Porsche DME system.
08-30-2002, 05:53 AM
#10
Instructor
Join Date: Dec 2001
Location: England
Posts: 176
Likes: 0
Received 0 Likes
on
0 Posts
I agree - just tuning up the chip which is just changing ignition timing and injector sprat time will not get you any more power. The whole thing needs to be approached as a package. On my C2 with K&N element, cat by-pass and cup pipe then remapping I got 225bhp at the wheels with 95RON.
It would be useful to be able to do your own remapping if you were going to make a number of changes over a period of time but you really need a dyno to be able to do this properly. You should have a look at the UNICHIP <a href="http://www.dastek.co.za" target="_blank">www.dastek.co.za</a> which is a piggy back DME so you change do a remap with out having to crack the Bosch/Porsche DME codes.
BTW: The original map is pretty good. On part throttle the mixture is set closed loop with the O2 sensor so you can only change the ignition and all we found was the odd degree here and there. Full throttle you can play with all the settings and we found that as standard it was a bit rich in the mid rpm and leaned it out a bit and got more power, again not too much to play with in the ignition. NOTE: do not be tempted to raise the rev limit to get some more power as you will get valve float and have the pistons closing the valves
It would be useful to be able to do your own remapping if you were going to make a number of changes over a period of time but you really need a dyno to be able to do this properly. You should have a look at the UNICHIP <a href="http://www.dastek.co.za" target="_blank">www.dastek.co.za</a> which is a piggy back DME so you change do a remap with out having to crack the Bosch/Porsche DME codes.
BTW: The original map is pretty good. On part throttle the mixture is set closed loop with the O2 sensor so you can only change the ignition and all we found was the odd degree here and there. Full throttle you can play with all the settings and we found that as standard it was a bit rich in the mid rpm and leaned it out a bit and got more power, again not too much to play with in the ignition. NOTE: do not be tempted to raise the rev limit to get some more power as you will get valve float and have the pistons closing the valves
08-30-2002, 11:04 AM
#11
Advanced
Join Date: Jun 2001
Location: Florida
Posts: 99
Likes: 0
Received 0 Likes
on
0 Posts
steve_eo,
I have written software that allows you to edit Motronic images. Send me your e-mail and I'll send you a copy of it. Of course you'll need an EPROM reader to get the image on your computer.
You can also dissassemble the EPROM image using any 8051 dissassembler if you really want to dig deep into what makes these things tick.
-doug
I have written software that allows you to edit Motronic images. Send me your e-mail and I'll send you a copy of it. Of course you'll need an EPROM reader to get the image on your computer.
You can also dissassemble the EPROM image using any 8051 dissassembler if you really want to dig deep into what makes these things tick.
-doug
08-30-2002, 02:51 PM
#12
Advanced
Thread Starter
Join Date: Aug 2002
Location: Ottawa, Ontario
Posts: 78
Likes: 0
Received 0 Likes
on
0 Posts
DougB, thanks for the software offer...I'll email you with my address.
I'm actually not interested in trying to tune the engine better than what Porsche already had...after all what the heck do I know. What I do want to know is what are all the inputs and what does the computer do with them. Example I'm read that there is an air temp sensor that backs off the timing when the air intake is more than 25C. Also what are the criteria for the AWD light to kick in....things like that. I really doubt I'll try and change anything...just being nosey..
Steve
I'm actually not interested in trying to tune the engine better than what Porsche already had...after all what the heck do I know. What I do want to know is what are all the inputs and what does the computer do with them. Example I'm read that there is an air temp sensor that backs off the timing when the air intake is more than 25C. Also what are the criteria for the AWD light to kick in....things like that. I really doubt I'll try and change anything...just being nosey..
Steve
08-30-2002, 04:14 PM
#13
Rennlist Member
Steve- You may want to go to <a href="http://www.Amazoncom" target="_blank">www.Amazoncom</a> and look up Bosch fuel injection. There are several good books, some published by the Bosch company, on the subjects you are interested in. <img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />
08-30-2002, 10:50 PM
#14
Advanced
Thread Starter
Join Date: Aug 2002
Location: Ottawa, Ontario
Posts: 78
Likes: 0
Received 0 Likes
on
0 Posts
Cupcar - Thank you.
I've ordered the Bosch book (author Probst). I think one of the biggest hurdles that I will encounter is knowing what the sensor is going to what input on the processor-then mapping those to the code. Whatever I do learn....if anything at all, I'll create a webpage to share this with everyone.
I've ordered the Bosch book (author Probst). I think one of the biggest hurdles that I will encounter is knowing what the sensor is going to what input on the processor-then mapping those to the code. Whatever I do learn....if anything at all, I'll create a webpage to share this with everyone.
08-31-2002, 02:19 AM
#15
DougB:
I'm not interested in doing performance mods to the program, but I might be interested in doing some stuff to gain better access to the diagnostics of the car instead of being limited to what the check engine light provides on my US model.
Where I work we use Power PC chips for the embedded systems we put in trucks. On ours the chip sits in an interface to a regular PC and a bus connection to the PC loads the programs we write (in C) to ram (not eeprom) on the Power PC host. When we have the program right, we burn the eeproms and put the eeproms and PowerPC chip into their mother board. Being able to edit and debug on a PC is obviously much better than the good 'ol days of using a programmer with a hex keypad.
I assume that the 8051 has similar tools, but how much money am I look at? A few hundred or a few thousand dollars (I think ours were about $2K for the boards and software).
Thanks in advance,
<img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />
I'm not interested in doing performance mods to the program, but I might be interested in doing some stuff to gain better access to the diagnostics of the car instead of being limited to what the check engine light provides on my US model.
Where I work we use Power PC chips for the embedded systems we put in trucks. On ours the chip sits in an interface to a regular PC and a bus connection to the PC loads the programs we write (in C) to ram (not eeprom) on the Power PC host. When we have the program right, we burn the eeproms and put the eeproms and PowerPC chip into their mother board. Being able to edit and debug on a PC is obviously much better than the good 'ol days of using a programmer with a hex keypad.
I assume that the 8051 has similar tools, but how much money am I look at? A few hundred or a few thousand dollars (I think ours were about $2K for the boards and software).
Thanks in advance,
<img src="graemlins/beerchug.gif" border="0" alt="[cheers]" />