[roninsoldier83]

Originally Posted by themyst

I want to know how he reduced ignition to begin with. Flash tuning I'm assuming? Probably adjusted the secondary timing table.

That would be my theory.... and it would explain a lot.

[Black Gold]

From what I saw, the is table limits were decreased to compensate for the higher load/boost.

I am curious though what the ECU's compensation mechanism would be, if it has one, if it could recognize that boost/load had been increased instead of being "blind" as you called it from the dummy signal.

[themyst]

Originally Posted by roninsoldier83

Actually, I'm really curious to see the timing table for the 335is as I'm not so convinced that it would be much (if any) different than a normal N54 335i timing table. I say that because with a device like the JB, you're altering the MAP signal, so the DME can't see full load. If the DME sees you're hitting a load value of around 115 (throwing a random number out there), when you're really seeing load levels that should be around 145 at XXXX rpm, the DME is first going to attempt to base timing off of the value it can see (i.e. 115 in this fictional scenario). Whereas, if you weren't manipulating the MAP signal in order to get the DME to increase boost, if you were using something like a cheap MBC (for arguments sake, assuming no fuel cut, CEL's ect) or a flash program that only modified boost settings and the ECU could actually see the fictional load of 145, it would react accordingly, which would be to decrease timing by X degrees based solely on the higher load, not that it necessarily has timing values that are different, it just wouldn't be blind to the increase in load.

The point in making a full timing table is to get the ECU to react accordingly under damn near any circumstances. If BMW did this correctly, on the 335is in "normal" mode, logging values should be similar to just about any 335i, but in overboost mode, the tables could be identical, but we would never know because the 335i never hits those load calculations, whereas if it did (via MBC or flash tune that only altered boost), it could actually pull the same ~4 degrees of timing, based solely on the higher load. That's just a theory, it'll be interesting once Cobb cracks the 335is DME in order to test my theory haha.

Either way, from the factory these timing tables can be fairly aggressive in many conditions, so it makes sense that Cobb would pull a few degrees in various locations throughout the table.

Just my $.02

The good news is, every Cobb OTS stage1 datalog and/or dyno shows consistent AFRs, and near-identical ignition curves with very few instances of knock. The only AP datalog I saw with knock was boom's multiple gear pulls, and I'm sure if it is possible to write in gear-dependent ignition curves, that knock event could be have been avoided as well.

I think load targeting could have some repercussions as well, as Cobb is doing, working with the stock DME logic. If the car has a pre-existing condition, it could cause the car to boost well over intended levels, to hit the load target.

I actually know someone who was sustaining 16.5 psi on 93 octane on his Cobb OTS Stage 1 map. Situations like this are rare, but entirely possible if the flash tune can't detect this pre-existing condition.

I am not certain if it is possible to write in an actual boost limit (since the DME does indeed read pre-TB boost), I'm sure it's a possibility.

What I am excited about is the possibility of integrating methanol safeties similar to what the piggybacks offer, as we can then tune for methanol injection instead of using it as an additive. I don't see it happening, simply because, where would the flow sensor output go?

Perhaps there is an unused DME input that we could hijack using a patch cable from a piggyback PnP harness for the flow sensor that Cobb could possibly implement for progressive meth? A man can dream...

[Clap135]

Im willing to bet the 335is and 335i share the same exact timing maps. Keep in mind the stock ecu is set up for 180 load, the 335 runs 130ish falling to 105, 14psi on the 335is would put it in the 160 range, if you notice from the timing table, 160 load has obviously lower timing.

[roninsoldier83]

Originally Posted by Clap135

Im willing to bet the 335is and 335i share the same exact timing maps. Keep in mind the stock ecu is set up for 180 load, the 335 runs 130ish falling to 105, 14psi on the 335is would put it in the 160 range, if you notice from the timing table, 160 load has obviously lower timing.

+1

My theory exactly.

[harry$]

I think load targeting could have some repercussions as well, as Cobb is doing, working with the stock DME logic. If the car has a pre-existing condition, it could cause the car to boost well over intended levels, to hit the load target.

I actually know someone who was sustaining 16.5 psi on 93 octane on his Cobb OTS Stage 1 map. Situations like this are rare, but entirely possible if the flash tune can't detect this pre-existing condition.


What might be this pre-existing condition?

[themyst]

Originally Posted by harry$

I think load targeting could have some repercussions as well, as Cobb is doing, working with the stock DME logic. If the car has a pre-existing condition, it could cause the car to boost well over intended levels, to hit the load target.

I actually know someone who was sustaining 16.5 psi on 93 octane on his Cobb OTS Stage 1 map. Situations like this are rare, but entirely possible if the flash tune can't detect this pre-existing condition.


What might be this pre-existing condition?

boost leak? bad turbos? who knows. I just know what I heard.

[themyst]

Another question I have regarding load targeting vs PID-based boost control -

Will there ever be a fix, per se, for these cars with undesirable throttle trimming and/or boost overshoot at the WOT event onset? I believe what Shiv stated regarding wastegate stiffness being different in each and every N54 holding water. I am not sure if this is something that can be addressed via ATR, but I don't think anyone here can disagree both throttle trimming and/or boost overshoot is a bad thing.

Granted, many of us would like to run a custom tune created through ATR (or even the Procede, from what I understand, is capable of doing) but I guess a few questions remain.

FYI - I have zero tuning experience. I just read and soak up as much information as possible.

[Morpheus]

Originally Posted by roninsoldier83

In truth, this thread was in no way, shape or form intended to slander their respective businesses. After reading much of the other thread, when creating this one, my sole intention was to attempt to share knowledge I've found over the years playing with/tuning cars, in order for more of this community to gain a greater understanding of some of these hot topic debates that pop up frequently around here.

I've emailed back and forth with Terry, and I don't have a problem with him, his business or his ability to make a profit. We both agree that avoiding detonation is the ultimate goal, and he shared his experiences with me over the years. As such, I also shared my points of view/experiences, and belief that by building a custom map could increase consistency and improve the look/feel of the torque curve.

Terry informed me that about an endeavor he had a year ago, attempting to develop a flash based program for these cars, and that the timing tables he played with seemed to have little bearing on the actual timing limits the car would impose, but admitted that there was some difficulty in using and identifying tables, or at least getting the desired results on this car, also mentioning that Cobb might have access to more tables than he was able to uncover (he even sent me a screen shot of the interface he was using). I have however also spoken with Rob at Cobb, who seemed pretty clear on as long as the tune is happy, the base timing figures should be what you're seeing, confirming everything I've learned about timing tables in the past.

I will say this about Terry, he has always been cordial and fairly explanatory (without attempting to lecture) about his reasoning in our conversations, more so than many would expect when having a conversation with someone who has creating a large thread that could have a negative impact on one of your products (albeit not intentionally). I have no problems with him as a person, and like I've mentioned before, when it comes to customer service, I would give him top marks on responsiveness (regardless of the time of day) as well as effort in attempting to resolve issues.

With that said, I'm a firm believer in the term "put your money where your mouth is". As such, that's exactly what I plan on doing. I've dyno'd a JB4 on my car (map 1 on 1/1/2011 firmware/sent the dyno to Terry as well), logged it as well for a couple days. I was less than impressed and it has not been on the car for a while (dyno'd on 1/29/11, removed 1/30/11). When the AP comes out for '07's, I fully intend on purchasing one (have the cash now if anyone at Cobb wants to allow me to "test" a beta haha! I have to stop hassling those guys ). At that time, I'm also going to throw the JB4 back in the car, update firmware, set it to autotune/autoPID make another pull or two on the dyno, gather a few logs for comparison and remove it. I'm going to perform the same process for the AP, comparing logs & dyno curves... and ultimately, when released, I'm going to tune the car via ATR software and compare it (dyno curves & logs) to the OTS maps from both the JB4, as well as the AP's OTS stage 1 map.

Ultimately, my goal is to let the numbers and shape of the curves speak for themselves. However, along with this data, I plan on posting my own personal impressions along with them, and I will attempt to do this in an unbiased manner. Hey, if I can't create a smoother curve, more consistent power and cleaner logs, I'm not selling anything, I have no problem saying so/admitting I was wrong... but I have a feeling that won't be a problem

Again, I'm a "put your money where your mouth is" kind of guy, and the combination of dyno time and purchasing another tuning device (Cobb AP) isn't exactly free haha!

I've informed Terry of my intentions to test my basic tuning theories (based on tuning cars in the past), document them and share them openly, to which he had no problem.

Based on playing with these motors for the past few years, I'm sure there's a few things Terry can teach me about them (he's more specialized on this platform/whereas my experience is with several other platforms), and I'm an open minded enough individual to listen, even if I don't fully agree with something. As such, I hope he's the same way. If he implements a new feature set in his future products that could improve overall performance and consistency, the way I see it, it's win/win for everyone. If we can all grow as a community and become more knowledgeable, with a greater number of tools at our disposal, I'm all for it.

With that said, Terry was working on the PRO board which utilizes CPS offsetting (timing control) LONG before I ever thought about making this thread... so the reality of the whole thing is that this thread isn't really teaching him much (if anything), that he didn't already know.

At least that's my $.02

-Brandon

looking forward to your post as well. I already have a Cobb, but I have yet to install it. I'm due a service, so after everything comes off and my car gets updated I'll give the Cobb a go. My comparison will be much more personal and less scientific and I don't feel the need to post the results publicly as I have nothing to prove. However, I am looking forward to your results.

[Morpheus]

Originally Posted by shifterboy45

one reason i am concerned is that the more advanced tuning solutions are getting close to release.

now im sure the tuners have zero malicious intent and will properly document the features available on these tuning tools. but as in each case of good intentions, bad actions follow.

without these types of discussions, the willingness of the newbie to buy something (that they have little or no knowledge of) and test it out with possibly devestating results. pushing this type of information to the forefront especially if this is a informational forum -- people seeking others approval for their personal decisions should be set aside for best practices in a tuning environment.

why foster the idea that anything goes, when ultimately it will cost someone in the long run. i always hear the word FAIL -- its my honest hope that the "baby doesnt get thrown out with the bath water"

I agree. I keep hearing ATR this and ATR that. In the wrong hands I think ATR will blow many more motors than the JB ever did. But maybe my "open loop bias" is showing as much as the rest of you.

I remember when the FC-datalogit came out and PFC map swapping was rampant. BOOM. That's when I bit the bullet and had the FD professionally tuned.

[Morpheus]

Originally Posted by roninsoldier83

Notice the overall shape and smoothness of the torque curve (STI) in relation to the JB4.... and in the STI's defense, I was time limited, and cleaned up the AFR a bit more later on the road. Ask me which one was smoother/more consistent.... then ask me which one was prone to more random drops in timing.... although, I think the graphs are pretty evident without me saying anything. As noted above, pay no attention to the overly inflated fully corrected dyno numbers (we're at 6000 ft elevation, turbo cars don't lose as much power up here/machine spits out overly corrected numbers). Needless to say, I believe that a custom tune can clean up the curve a good bit and make the car more consistent. I suppose we'll see soon

Is the STI closed loop all of the time? I don't know anything about them. Also, isn't the JB4 still being developed? Just making sure we are apples to apples.

[Morpheus]

Originally Posted by themyst

The whole point here is, the stock ignition curve is too aggressive for increased boost. Ask BMW why they dropped the targeted curve 4 degrees across the board on the 335is in "overboost" mode.

It is nice that the stock ignition curve is set so aggressively from the factory though in that there are gains to be realized with meth and/or race gas on increased boost without altering the ignition tables directly (JB, for example).

It was my understanding that the tables were for max ignition advance and the ECU would attempt to hit those values, but might not depending on octane, temp, etc. And like others have stated it will not exceed those, even with race gas. So instead of thinking of those tables as absolutes like an open loop map, you have to think of them as caps. I know you guys will burn me at the stake, but I'm thinking BMW left some timing (power) on the table on the stock max ignition advance. Otherwise you could fill a stock car with race gas and beat up on more expensive product lines. Same with the iS. Max advance (cap) is 3 degrees less. Is this for safety? Or so the 335iS doesn't beat up on its big brother M3? Just playing devil's advocate here.

[Jake@MOTIV]

Originally Posted by Morpheus

It was my understanding that the tables were for max ignition advance and the ECU would attempt to hit those values, but might not depending on octane, temp, etc. And like others have stated it will not exceed those, even with race gas. So instead of thinking of those tables as absolutes like an open loop map, you have to think of them as caps. I know you guys will burn me at the stake, but I'm thinking BMW left some timing (power) on the table on the stock max ignition advance. Otherwise you could fill a stock car with race gas and beat up on more expensive product lines. Same with the iS. Max advance (cap) is 3 degrees less. Is this for safety? Or so the 335iS doesn't beat up on its big brother M3? Just playing devil's advocate here.

Without a way to actually advance timing while sitting on a dyno there is no way to check what is left on the table. Filling up with the highest octane out there and running meth on top - there may be a few more degrees left on the table.

It is too bad some of the guys who think we are hating through creating these discussions to enlighten people how the rest of the world follows tuning theory. I brought up EFI University which does a ton of training for professional tuners and enthusiasts. A friend of mine from Haltech USA used to instruct a lot of these types of courses. I am happy to see that some people are catching on and able to see through some of the BS that is spoken. This platform is itching for a tuning solution that is complete. By the looks of it COBB will be first with ATR unless Shiv actually opens up the Halwin software for people to tune.

I personally don't think the end of motors will happen when ATR is released. Just as BMS is doing - the users will be able to rely on the stock DME to protect them. I have seen first hand as in developing software for this platform we tested the effects of lean burns, and timing errors, etc. The DME is very fast and very capable. That backup/safety logic is great to have - just not to use as a primary means.

[Black Gold]

Originally Posted by Morpheus

I agree. I keep hearing ATR this and ATR that. In the wrong hands I think ATR will blow many more motors than the JB ever did. But maybe my "open loop bias" is showing as much as the rest of you.

I remember when the FC-datalogit came out and PFC map swapping was rampant. BOOM. That's when I bit the bullet and had the FD professionally tuned.

Well atr also means that tuner shops get the software ad can then protune your car (this includes Cobb themselves) who typically can do an excellent job tuning your car and provide safer and more powerful results than the ots Cobb maps.

This is why I am excited for atr. I will not be tuning my car myself, I will leave that to the experts.

[roninsoldier83]

Originally Posted by Morpheus

I agree. I keep hearing ATR this and ATR that. In the wrong hands I think ATR will blow many more motors than the JB ever did. But maybe my "open loop bias" is showing as much as the rest of you.

I remember when the FC-datalogit came out and PFC map swapping was rampant. BOOM. That's when I bit the bullet and had the FD professionally tuned.

I'm actually going to agree with that statement. Giving users the ability to run 20+ psi on an otherwise stock car, combined with some users not having a great affinity for adjusting timing and other parameters necessary for a safe tune will surely result in more popped motors in the wrong hands. Although, I would like to think that most people that aren't experienced/familiar with tuning interfaces will have their car protuned by an experienced tuner.... but just like any other community, there will always be those that push too far.

Originally Posted by Morpheus

Is the STI closed loop all of the time? I don't know anything about them. Also, isn't the JB4 still being developed? Just making sure we are apples to apples.

That's a complicated question to answer.... the STI's fuel system is closed loop under idle/cruise/light acceleration, but under WOT the fuel system is open loop (even though LTFT's will alter OL fuel maps slightly, providing a smaller level of control/compensation based on historic fuel correction in closed loop). In comparison the N54's fuel system is fully closed loop (no map/table, just a targeted lambda).

Although each employ timing tables, as well as a plethora of compensation values for things like knock correction, IAT's, ect. My bet is that by playing with targeted lambda (fuel), boost and the timing tables I can make a much smoother curve than would be achieved with an OTS map. I used my old STI as an example because I just happened to have a couple of my old dyno graphs laying around that were on the same dyno with smoothing set to 0 with the same 2000-7000rpm scaling, but in reality, most of the experiences I've had have shown that a car's torque curve can typically be made a bit smoother with a good custom tune.

They're releasing new firmware versions for the JB4, but it is fully functional, so I would suppose we would have to define "being developed". Quite a few tuning devices come out with constant firmware revisions (Cobb AP's included on numerous platforms), so I suppose that would be based on how you want to look at it. When I dyno'd on the JB4, IIRC it was on the firmware it shipped with, which was either their 2nd or 3rd firmware version for the JB4 if I'm not mistaken. However, they've come out with a couple new firmware releases in the last couple weeks. I would say "no", the hardware is in final production and for sale to the public as a whole, however it appears that Terry is still making minor adjustments as time goes on to firmware, which is to be expected, and will likely continue on and off for while.

Although, I do plan on doing a comparison of that JB4 firmware version map 1 to their autoPID map on their latest firmware version release.

The goal is to improve the curve as well as the consistency of the vehicle through custom tuning and to display the results. On the majority of the cars I've worked with in the past, I've typically been able to improve peak power/torque as well... but we'll see how much is left on the table compared to these OTS maps when I start playing with it.

I'm looking forward to the challenge to say the least

[Joshboody]

This is continuing to be a very interesting thread and I’m just now catching up on the reading. Some comments:

First, the table produced by Terry compared to the Cobb posting is very different. I have NOT seen any evidence of the DME raising timing above the 10 to 13deg at >120 units of load. Now his table doesn’t have these load values, but in comparing the 2 can would be concluded that timing would be considerably higher. Like mentioned, maybe a secondary table for limp mode.

In my experiences in logging, I could make the claim that there is no timing compensation with IAT… I know this goes against all other platforms.

Themyst, you mention a flash may NOT be good with certain "pre-existing conditions", but this would be the same with a piggy targeting boost instead of load. I believe the DME has some reaction to DC... too high or too low... this should be the case with a flash also. I have read that piggies have some method to "piggy" off these DC limits also.

[shifterboy45]

Originally Posted by FlyLow335i

Without a way to actually advance timing while sitting on a dyno there is no way to check what is left on the table. Filling up with the highest octane out there and running meth on top - there may be a few more degrees left on the table.

It is too bad some of the guys who think we are hating through creating these discussions to enlighten people how the rest of the world follows tuning theory. I brought up EFI University which does a ton of training for professional tuners and enthusiasts. A friend of mine from Haltech USA used to instruct a lot of these types of courses. I am happy to see that some people are catching on and able to see through some of the BS that is spoken. This platform is itching for a tuning solution that is complete. By the looks of it COBB will be first with ATR unless Shiv actually opens up the Halwin software for people to tune.

I personally don't think the end of motors will happen when ATR is released. Just as BMS is doing - the users will be able to rely on the stock DME to protect them. I have seen first hand as in developing software for this platform we tested the effects of lean burns, and timing errors, etc. The DME is very fast and very capable. That backup/safety logic is great to have - just not to use as a primary means.

very true statement...

the average tuner will show some discretion when reaching the limits of octane and timing, and you can "feel" the car pull back on you, u can also feel a mis-fire if you are well aquainted with the engine in question.

that level of expertise does not come by reading or examining a product review -- it comes solely with experience.

there quite a few incidents that the car is pulling timing due to combustion problems, but are side-stepped as adaptation or otherwise -- high speed surging -- and most guys that are not familiar with tuning will think its just an anomoly that will go away -- and guess what? it does, becuause the DME took care of the problem by degrading the timing curve. so this extremely smart DME now allows errors (misjudgements) in tuning, saving the engine for another battle against detonation -- if you dont know there is something wrong, you cant fix it.

now those who do have the expertise can do the same thing, and say "ah ha!" -- i need to back the timing down, look at my fueling, assess the load range and make adjustment accordingly. the will look at the TPS signal, and possibly have a reliable EGT to make sense of the whole equation. if not integrated into the actual toolset, who is going to invest in a $500 Fluke meter to monitor voltages and resistance?

im not against novice tuning, but i hope that when this task is taken on, there is preparation for the task. its fun, but costly to make a mistake -- and relying on the DME should be the last resort for any tuner.

[rKi]

Definitely a great read, thanks dude

[Syndicategt]

Great thread, I didn't realize this was stickied. All newbies to tuning need to read this before they purchase "x" tune.

[UFGatorGuy20]

Great read. Definitely a must read before you buy "X" piggyback or "X" flash tuner. I come from the 03-04 mustang cobra world (cough cough). All about live dyno tuning and proactively adjusting timing for your specific application.

[georgelb]

What a delightfully educational thread.

[Jeffman]

Originally Posted by SennaZ

Great post Brandon!

Would like to repeat myself from some of the other threads and emphasize your point quoted above. People who subject these engines to excessively agressive tunes owe it to the future buyer to do an engine oil analysis at least once a year. Search for some of the Mr. 5 threads on the subject to get baseline numbers by which you can judge the effectiveness of your tune and put all of the hypothetical arguments for and against one type of tune or the other to bed. Some dude doing WOT runs as a part of his daily drive on a mild flash tune is probably doing the same engine wear damage as a guy who is rarely gets into it with a JB3 on off-brand cheap gas. But it's easy to find out.

I find all of the hypotheticals rather tiring because the reality is that none of us know the Siemens MSD80/81 control logic well enough to have any real idea as to what it's capable of doing or not.

This is a great thread! As Siemens seems to have partnered with BMW AG on turbo technology, I was wondering whether Siemens has any relevant patents filed in the early 2000's on controlling turbocharged, direct-injected engines that could actually tell us something about how the controller works. I.e., such patents could very well describe some of the MSD80/81 control logic. I found US patent 7155899 which I believe may be helpful. It can be read at

http://www.google.com/patents?id=zox...ed=0CCwQ6AEwAA

The original patent was filed in Germany in 2002. Based on typical R&D cycles (four years to commercialize a new mechanical engineering invention), I have a sneaking suspicion that some of the N54 control logic may be described in this patent.

Let me know if anyone finds this useful and I'll send links to BMW's patents on turbocharging (if any).