Separate names with a comma.
Discussion in 'GS Boxers' started by Poolside, Nov 18, 2010.
Any particular reasons why the 4th option of tuning the stock ECU is not a option?
So, How soon?
Will you sell it to the other side of the Atlantic?
I was researching the "accelarator" hardware that works with the air temperature, and found this thread, so I'll wait.... but don't take too long.
We discussed that option. But the logistics were a killah.
<table border="0" cellpadding="5" width="650"><tr><td>No SG, that isn't me. And I'm not these either.
ReCyclizer http://www.novitech.nl/abouteng.html €35/$45.91
Booster Plug http://www.boosterplug.com $149.95
Accelerator http://sol2.be/Accelerator/English/index.html €40/$52.44 €48/$62.93
Power FRK http://www.powerfrkusa.net $293.00
MemJet http://www.memjetmodule.com $198.00
I've been noticing that EFI product descriptions in general have changed. I'm not singling anyone out. I just mean in general. EFI aftermarket product descriptions have changed to reflect the reality of fuel injection, as opposed to taking advantage of common misconceptions.
JJ calls that stuff 'super hyper marketing glitz'.
<table border="0" cellpadding="5" width="650"><tr><td>It is an option at some level, it just didn't occur to us as practical. I mean, the ECUs are programmable by some method or another. Whether a JTAG or BDM port is available externally I do not know. If not, you'd have to open the case and solder a connector to the PCB. The legacy development software may yet be available from Bosch.
I was under the assumption the 1100 and 1150 gs used a bosch motronic with a removable eprom, like the pre OBDII automotive systems?
Why would anything need to be added inside the ecu vs just changing the stock eprom with a tuned eprom? ECU limitations? Concerns/ difficulty for the end user installing a chip and encryption board? Software tuning issues?
<table border="0" cellpadding="5" width="650"><tr><td>I see what you're asking now, and those are good questions. You're right, the Motronic uses a removable EPROM chip. As you probably know there are several vendors of EPROM chips for the Motronic.
Will aftermarket EPROM chips produce the same result as these IICE temperature offset products? I can't say.
What I can say is, the temperature offsets produce as much power and drivability improvements as is possible with fueling changes.
Technical information, propellerhead alert.
The Motronic processor.
The Motronic uses a licensed Intel processor, more than likely an 8051 derivative. The chip itself is made by various manufacturers, Infineon, Siemens, etc.
The 8051 core can support both internal and external program memory. Program memory is often referred to as 'firmware', and represents the 'instructions' the processor follows. Now, the external EPROM may contain only simple fuel map data, but the chances are good the external EPROM also contains program instructions.
Changing the program on the EPROM chip.
The only sensible way to change the Motronic program is to have possession of the original Source Code, and use the ECU Software Development Kit (SDK) from Bosch. That is, if it's still available for the legacy Motronic ECUs.
What if you couldn't do the sensible thing, and didn't have the Source Code, or the Development Kit? Luckily, programmers and guys wanting to modify their bikes are not always sensible. Count me in that group.
Without the original Source Code, to make changes to the program memory requires that the binary data on the chip be 'decompiled'. That means to revert the binary data on the chip back into an 'approximation' of the original cryptic text instructions written by the people who wrote the program. Rendering the decompiled code into an understandable form is an intensive process, and one that rarely succeeds.
But, let's say they get lucky. The code is decompiled, and the team labels enough variables to allow at least partial understanding of the subroutines and the program flow. They now must make changes to the code and test the results.
To have a realistic chance of making changes to the decompiled ECU code requires on-chip JTAG or BDM debugging access. This allows the software engineer to watch the code while it's being executed.
What about these aftermarket Motronic EPROMs? Did the people who made them do all of the above?
In all likelihood, all the aftermarket EPROM chips are identical. It seems likely that everyone using a copy of the same EPROM.
Where did they get the 'original' copy of the EPROM data?
Direct from BMW! The EPROM chip was available as a factory part. I'd say the aftermarket chips are either copies of the 'retuned' EPROM available as a factory part in Europe, or they are copies of the factory-prepared EPROMs used in the 1100S Boxer Cup race series bikes. The series ran between 1999 and 2004.
That's a bold conclusion, Jim. How do you justify it?
If the EPROM contains program memory, and if someone had decompiled the code, then it would be on the interwebs.
Also, if anyone had done that, it would be serious bragging rights. The pressure to claim those bragging rights would be uncontainable. In addition, if the vendors had used these processes, they would have proudly spoken about it on their websites. (It could be that after reading this their website content may change.)
The most compelling thing is, since the modified EPROM was available from BMW, it is highly unlikely the vendors are doing anything more complicated than making copies of the BMW Boxer Cup EPROM.
I think you are right. In particular: Knowing that someone has done it doesn't make it less hard for you to do it.
So bragging about your achievement does not help you competitors.
If you would like to know more about EPROM tuning here is the guy you want to talk to. Marc Salvisberg At FACTORY PRO (415) 491-5920. I had him fly into town and and burn a custom chip for my Aprilia a few years back. really nice guy. I ran a chip in my R1150GS and it was a night and day diff. The chip still had the diff maps which could be activated via the cat code plug. Heres a little more on that. http://www.2ridetheworld.com/new_mod_pages/bbpower_chip.htm
Yes, and you're about to slander them at the end of your post...
And spark angle changes? Which sensor do you tweak for more spark advance at idle? At WOT?
I am the guy that does the stuff you're talking about. I make my living reverse engineering the software inside Motronic boxes and all sorts of other stuff. So, yes... it can be done.
The only sensible way to tune the Motronic is to figure out how it works, figure out how to change it, and then tune your changes on a real bike on a real dyno with real data equipment.
Oh man, this is just sad. And obviously bullshit. (Can I say that on here?)
I suppose they are running LTs on R1100S Boxer Cup software? Or maybe they got the software from the K1200LT Bricker Cup model?
Don't be so naive. There are plenty of people who are very capable of figuring out how the Motronic ECU stores its data and modifying said data for performance gain, even without "decompiling" the code. (Lets call it disassembly, because that's what we actually do out here in the industry.)
Disclaimer: I am not associated with Rhinewest, I don't even think they know who I am. They were just a convenient example.
Nope. People have done it. Done it regularly, for things much more complicated than bike Motronics, and done it since the 1980s. I do it every day.
The vendors that have used these processes have the good sense to keep quiet about it because it means less competition. The contractors that they hire to do these processes () keep quiet because it means job security.
So, if you think that you've cracked some sort of ridiculous BMW bike conspiracy and you're going to set everything right with a couple of pots connected to some motronic sensors, you're nuts.
The fact is, if you're smart enough to know how your changes will effect EVERYTHING that the ECU does, you're smart enough to go in and make the changes in the chip where its easy.
Edit: I probably come across as a dick in this post, which sucks but probably reflects reality. Consider this an invitation to PM me and talk about ECU hacking. Merry Christmas, if that's your thing. :slurp
Interesting thread. Keep it coming.
I don't think you come across as dick.
Perhaps a bit arrogant, but not a dick. A certain amount of arrogance goes with the territory.
I'm one of those guys that did this kind of thing back in the 80's, got the court injunctions to prove it. I think my biggest was a network box with 64K of code. Took me the better part of a month; the end result was the afore-mentioned injunction, plus a $big contract from the OEM to fix some bugs in the code. They reasoned that if I could figure it out the hard way, I could certainly also fix the bugs in it, which I did. Disassembly of large amounts of code is non-trivial to do, but certainly far from impossible.
Sorry for the hi-jack, but it's perhaps a bit relevant. I'm still very interested in the "pots connected to the motronic sensors".
Are you not aware that the temperature inputs change the spark advance?
A modified chip was available for the K motor as well.
just wanted to express my thanks to Poolside and JohnJen for explaining a lot of whats going on with our fuel injection systems.I suspect that a lot of the time you chaps are having to over simplify things so that those of us with no grasp of the concepts involved can get a basic idea of whats going on,I feel that because of the audiences general ignorance of the subject its necessary to take a lowest common denominator approach.For the last few days I have been trawling through the entire archive reading anything FI related(about 50% done now!),time and time again the same 2 names occur.Well done gentlemen,over a period of years you have helped a lot of people in a logical and concise manner.Cheers Chris
I am aware that pretty much everything changes pretty much everything. I also know how to get into the code and get the specifics.
The various changes you can make with modified inputs are not predictable unless you know the maps and how they work. Thus, if you have that sort of understanding you may as well do your tuning with a chip.
Hell you may as well use a PC, atleast they hijack outputs so that you know what the actual result will be.
<table border="0" cellpadding="5" width="650"><tr><td>I think you pretty much need changing.
The maps are based on combusting gasoline and operating the catalyst afterburner with the combustion products. Few processes are more well-known than that.
The maps are the least relevant to leading throttle and transient throttle response.
Leading throttle and transient throttle are based on air density and motor temperature respectively.
You want to reprogram the chip to do that, I'll see you at the finish line.
Are you saying there is no map I can modify to increase or decrease throttle pump enrichment?
Sure, but they're based on tables. That we can get in and modify.
Keep in mind, the ECU did not spring forth from some engineer's brain ready to run every bike ever, no matter how well understood combustion is. It takes tuning for every specific application. So, there is a framework within every ECU (including production ones) to support modifications to the tune. This includes transient fueling, spark and everything else.
Did you say that your products would work on the parallel twins? Like the 800GS? If so I look forward to trying your product.