Funny that your age means you have a clue…
That couldn’t be any further from the truth normally.
It drives just like it did before. In Dynamic mode specifically there is more of a surge (someone compared it to a sprint booster, but I don’t know what that feels like) In normal mode it just feels stronger. I haven’t had any weirdness reported by others.
No one is saying it can’t provide gains. Beyond that, I think you need to say it’s your opinion. There is absolutely no logic or hard evidence that suggests anything else you are saying is accurate.
In terms of modifying the ECU software, all the major tuners have a proven track record. This idea that there is something to be afraid of is not backed up by years and years of data. Can you say the same thing about Chipwerkes? They are new and have no tie to the Audi community, make a similar product for 4000 vehicles, and the hard evidence/data is very mixed at best. What logic or hard evidence are you using that makes you think it is less risky than a flashed tune from one of the major tuners?
I understand you are saying everything is working well for you, but we also know that’s not the case for everyone. Based on the evidence and data presented on drivability issues, gas quality issues, and higher elevation issues, your argument that everything works fine again defies logic. You are ignoring all the data because it works on your car? Some of the logs I have seen have been absolutely horrible. HORRIBLE!!! I’ve never seen anything remotely similar on a flash tuned vehicle, and we know flash tunes from the major tuners do not have the same drivability issues, or problems with gas quality or elevation changes.
I guess we weigh risk very differently? Logic tells me it seems more risky to modify and manipulate stock sensor values without compensating in any other way, and then just hoping everything works out because you left the stock safegaurds in place. It’s detailed above why that can be an issue, or what issues can be caused. Evidence shows flash tuned cars don’t have those same issues.
I find it hard to believe any engineer that works in a complex hardware/software world would be concerned about such changes. I also don’t necessarily equate tuning changes to code changes so maybe my perspective is different. I’m use to seeing these types of tuning and code changes done on semiconductor equipment that cost millions of dollars. The idea of doing it on a $55k car seems pretty risk free if done properly. The thought of manipulating a hardware signal/input has never seemed like a better idea than a tuning/code change to any engineer I’ve ever worked with. That’s crazy talk coming from an engineer (and I don’t doubt you are an engineer, but you’re trying to hard to make a logical argument that doesn’t exist).
Rough crowd… I stated twice my opinion and MY logic for going with a box vs. flash. CW says they have been in business since 2001 (there is a nice picture of shiny garage with dyno on Google). Racechip since 2008 and ABT since whenever, all using boxes vs. flash and all claim TUV. ABT is +4000 Euro, so that’s a different animal. Racechip also attaches to an extra sensor (see their Youtube install guide)…maybe I’ll buy one once available for MY15 and compare.
There is +1 year log data on CW (with A6/A7 as well) and over 5 years of data on various flash issues, including several failures in early builds I’m aware of. I’m not aware (and expect it would have come out on one of the forums) of any major failures w CW or other box vs. flash. I remember the MTM debacle where it simply didn’t work (at least on US cars).
It’s not a debate for me since I won’t flash my MY15 unless Audi does indeed deliver a performance line of products.
I’ve learned for all the postings on various forums, so trying to contribute not argue.
Cheers
I found this thread interesting, sharing for learning http://audirevolution.net/forum/index.php?topic=2932.50
“The car is always in a targeted lambda closed loop mode, it never runs “open loop” as stated in this thread. The ECU targets a specific lambda based on mass air flow and RPM”
[/quote]
First off, this is very different than the bmw options. Literally all this does is modify the pressure sensor (and potentially the iat signal). If you are a programmer, then you understand what “garbage in garbage out” means. Everyone defending this seems to think factory protection is some kind of Boolean true or false. It’s not, go up and look at the Audi service manual, and see what it says can happen when the map sensor reports a false value. Secondly, fueling is done both open loop, and closed loop. If you don’t know what that means, go read up on a book on feedback controls. This sensor is fucking up the open loop strategy… is the wideband able to “compensate”…who knows, but I sure as shit don’t want to experiment to find out.
Really your argument is no different than all of the other audizine minions, and honestly, I doubt you want to actually debate this. Just label us haters who trust our cars to tuners for obscene money and just drink kool aid.
There’s a reason I’m not gonna go get in line first to flash apr dsg software because I will let others prove it out first. Because I like data, and reliability…not saving a couple hundred bucks and being fine with clearing cel’s after my car runs like shit cause I didn’t have the “update” for the D1 sensor spoof.
[/quote]
Hey welcome to the site. Creepy lurker 
You’re certainly free to do whatever you’d like. Nobody can fault that
My concern is the proof of concept. One guy posted modest (at best) 1/4 mile times. (I’m not listening to anyone with a dyno sheet because
- It’s easy to make it say what you want
- It doesn’t tell me how my car accelerates
- I don’t drive on the dyno I drive on asphalt
So to that end we have seen one mediocre-at-best 1/4 mile time at the fastest track in America. And it’s from the guy selling the product over here. The guy who posted here, and when asked about his involvement and what he paid etc left out some key information that I can only call deceitful.
So that, for me, isn’t enough to buy a product. Nowhere near it. I want to so 5 people have proved acceleration gains. Then I want to see them drive their cars for 10,000 miles each. Then I want those cars to be checked for healthy compression etc. THEN I will start sniffing around.
None of that has been done here. So for me, the product is nowhere near ready to purchase. And for that reason, I’m out. /Dragon’sDen
P. S. Didn’t your warranty leave the building a year ago? Who cares about TD1
I kinda checked out when he said I’m looking for the best thing for my car and I don’t trust the people who do this for a living I’d rather not have everything working properly I Ike i paid for. I’d rather fool the car and lie to it about a very important input. The two don’t go together.
Also to say it’s working well but that you need to use specific fuel to compensate for the lack of actual linersation…is also kinda counter productive.
Also saying your 78 won’t make any one think your smarter.
I would hope most people have already read the self study on the 3.0TFSI motor. We all know what the MAP sensors do. That is why the chipwerke module actually works…because it changes the readings giving you increased boost, increased timing, and relying on the factory to provide an OEM afr. This is what my logs have proven. The fact that it allows for increased timing is also why you are seeing me making power all over the power band even down at 3000rpm on my dyno.
As for the second part about the sensor failing it is irrelevant of course unless people have a failed sensor. The sensor still works with CW hooked up obviously.
I certainly haven’t ever stated the CW is the same as a tune but for many people it provides similar or even greater in my case performance gains with zero driveability issues and safer looking logs than many people with flash tunes. Arguably if you and I were to compare logs I think if you objectively looked at them mine would look safer no?
Its just disappointing seeing the bias here. You mentioned that the logs on CW you have seen have looked so horrible you have never seen logs so bad with other tunes. But you are failing to mention that the same persons car was pulling 12.6 degrees of timing STOCK! So how is that fair to say at all. In fact what we saw is with the CW on his car the bypass even opened up showing the factory safeties were in place. That would be a good thing right?
And then you say no other tuners have had driveability issues. Really? I think you mean to say that after multiple revisions the tuners tend to get it right but I would not say no major driveability issues. I means on the first page of Audizine alone there was…
[QUOTE=sergzz;10733124]Just to update everyone. John and I had a look through the codes and decides that the best way to understand the issue is to flash the car back to stock and see if the logs show up with the same engine fault code. I have since then flashed the car back to stock; sent the logs to John - awaiting for the reply. No engine fault codes showed up and the car is shifting completely normal with the stock tune.
I appreciate the team working with me through this.
For anyone that is interested, I can also post the videos of what my car was doing with version 1, version 2 and how it behaves as stock, in case anyone is interested. I would like to see if other people are experiencing these RPM fluctuations during shifts as well.
[/quote]
[QUOTE=TexasDfwS4;10734630]John like I mention in my last post I’m having what sound to me like the same issue. The first tune that was used on my car from Unitronic was “jumpy” as in it would ack like there was no tune and bang there it would be, but not smooth. Then you said you had an update so I got this put on and since then the car feels weird. When I wot the car feels like it’s missing or stuttering. I thought this would clear up when it learned itself but no such luck. I trusted the first tune more then this one.
[/quote]
Those above posts were sitting literally two or three posts away from the APR/CW thread and I didnt see a single of you non biased folks on that thread warning people away from that tune.
How about low fuel pressure (ie, you are going to run lean) with some of the facelift APR cars…
http://www.audizine.com/forum/showthread.php/594874-P310b-low-fuel-pressure-code
Apparently people dont have long memories but there were blown motors with Revo software, no power from GIAC revision 1 for facelift cars, and there are plenty of people pulling 8 degrees and more timing with GIAC and APR cars. I’m not saying CW is better than them but at least take the blinders off and realize that the “perfect tune” is a unicorn. Everyone has issues.
Mike
Hello there. As I posted…
It sure seems you enjoy talking about people more than talking to people. Again, if you have any questions about some kind of perceived deceit, please do ask so I can clarify for you. It is a sign of character to actually talk to a person instead of about them. If you look it will real easily show that I posted on audizine that I had obtained authorization to exhange my S unit for the Pro unit and I had planned to keep the S unit for a few days to log them against each other.
I havent posted this here either because I dont update you guys on every move I make but just to let you know I have decided to actually send the pro unit back and keep my original S unit. IMO the aluminum box heats up easier and I didnt find any performance edge of the Pro versus the S.
Mike
Mike
Mike we asked you straight up and you were far less than honest. Now you’re adding details that you quietly omitted when you were originally asked here?
At this point you running damage control all over the place isn’t going to undo that simple fact. Fortunately forums are posted in text with a time record in black and white. Or green.
And fortunately on this site new members can’t edit their posts.
Good luck with your chip-sometimes-werks enterprise. All the best
Well, I would go to quote more than just that, here is more
[quote]he factory fuel strategy in the 3.0T is modified as well with our code. The stock car will run fairly lean until crossing a modeled EGT threshold of about 900C, then will enrich heavily at that point to cool things off. If this strategy isn’t modified, the fueling will be inconsistent. The car is always in a targeted lambda closed loop mode, it never runs “open loop” as stated in this thread. The ECU targets a specific lambda based on mass air flow and RPM.
[/quote]
The “specific lambda target” is what I would call the “open-loop” component. Things can get overly semantic, the reason I said open-loop vs closed-loop is because it is well defined terminology.
Whenever the pedal input changes, and the torque request in the ECM shifts, there will be a transient period (https://ccrma.stanford.edu/~jos/fp/Transient_Response_Steady_State.html) where the measurement from the lambda sensor will not be completely reliable for controlling the fueling (I have heard on the order of 1 second). The reason this is needed is because the system must be causal. Auto companies are very heavy into the world of “calibration” where they literally drive cars for millions of miles in different conditions to refine their calibration using statistical tools (called DOE). They are not very into exotic controls algorithms (both from computational cost perspective, but also because of “inertia” in a certain way of doing things), like you might see on a F22 flight computer or satellite control system.
Here is the best analogy I can think of: it’s cold out, and your lady friend has 10 different jackets to chose from with different insulation values. You want to dress her appropriately for a walk in the park. Let’s just say you want to dress her perfectly so there is no complaining (too hot or too cold).
You have 2 opportunities to chose a jacket for her. To start, you get an initial choice, and she does a walk around the block. To make this initial choice, the only information you have is a thermometer you have sitting outside on a shed. When she comes back from the walk in the park, you then get to choose a more appropriate jacket based exactly on her feedback, which will take more information into account (her state of metabolism, maybe she’s on the rag, humidity in the air, and whether she “felt” cold).
The assumption here is that the walk around the block might create some bitching from her, but the initial choice will be good enough so that she’s not saying “fuck it we’re staying in”, or “you can go to hell cause you just made me walk around the block in 15 deg weather in a sun dress”.
In the case of CW, you have more or less made the thermometer scale completely fucked up when it gets really cold out. So every time for this walk, you are letting her deal with whatever the weather is and assuming you’ll still go for that walk once she comes back. The “open-loop” stage is the walk around the block, and the “closed-loop” stage is once she comes back (in case this wasn’t obvious lol).
Now in practice, as the revo guy said, the controller is never in only “open-loop” mode, but the issue is “The ECU targets a specific lambda based on mass air flow and RPM” implies the set point to the “closed-loop” controller is based off of calibrated values from Mass air flow (which is a calculation based on MAP and IAT I believe). This element is what I would call “open-loop” and would be what can cause the car to putter, misfire etc.
We’ve seen people’s cars run bad, and the excuse is that “they ran it too aggressively”, whereas the real reason is: this method is Frankensteined to begin with and they just experienced a certain use case/load condition/device setting that pushed the car too far outside of the proper fueling set-point and caused the engine to misfire and have issues. Changing the setting to something “more dialed back” is just changing what use case/preconditions are necessary to experience the issue again…because you fundamentally have the same fucking system manipulating things the same fucking way…in what I would consider a dumb way.
Unlike yourself, I’m a lot more hesitant to spoof my engine sensors (again very very different from BMW piggybacks) than allow some software people at APR to change the calibration tables (not the algorithms, but the tables), where their reputation is on the line, and tons of others are going to vet out the bugs before I flash to my car.
You are reading that with a hilarious bias. There is nothing there that says a failed sensor = dead sensor. If the sensor is dead then why the fuck would anything even happen (i.e. instant CEL and limp mode)? I’ll give you a hint: failed sensor = malfunctioning sensor…and guess what, chipwerke = malfunctioning sensor.
So I guess you’ll just conveniently dismiss something from Audi themselves on the subject. But we’ve already seen you really fail to engage this with any kind of intellectual curiosity “an update to the CW sensor spoof curve is the SAME as an update to the APR tune” LOLOLOLOLOL
[QUOTE=TexasDfwS4;10734630]John like I mention in my last post I’m having what sound to me like the same issue. The first tune that was used on my car from Unitronic was “jumpy” as in it would ack like there was no tune and bang there it would be, but not smooth. Then you said you had an update so I got this put on and since then the car feels weird. When I wot the car feels like it’s missing or stuttering. I thought this would clear up when it learned itself but no such luck. I trusted the first tune more then this one.
[/quote]
Those above posts were sitting literally two or three posts away from the APR/CW thread and I didnt see a single of you non biased folks on that thread warning people away from that tune.
How about low fuel pressure (ie, you are going to run lean) with some of the facelift APR cars…
http://www.audizine.com/forum/showthread.php/594874-P310b-low-fuel-pressure-code
[/quote]
Looks like I was incorrect about this fueling issue. Not really sure what to say. I’m sure APR will try to address it. But I would return my tune if I had that problem. No one is arguing that tuners are infallible…but the fact remains that they can CHANGE the software while CW can’t do anything other than change their spoof curve (how is this different than just changing the setting to D1?).
Regarding unitronic, I would never tune my car with their software. And there’s no way I’m getting the new APR tune/dsg software until the fall at the earliest.
You are continually making this out to be
ECU Flash Vs Chipwerke Piggyback
when in reality it’s
do something dumb to your car (chipwerke piggyback) Vs not do something dumb (stock or ecu flash)
Appreciate the further detail and insight. We (any of us on this forum) don’t’ know how tight the feedback loop is, but as you say I suspect that whenever a code is thrown or a surge/stumble happens it’s a case of the ECU running against a target vs. feedback loop. When the low octane/high DA issues happen, I suspect it’s because the injector duty cycle is at 100% can ECU can’t keep AFR correct. At high DA underboost similarly the ECU see’s a too low MAF reading. At my octane/DA, the D-1 curve seems to work well. Low octane/high DA works well with a less aggressive curve. I don’t know whether the ST MCU in the CW modifies the curve or it is linear…I’m going to ask them (and Racechip) to see what they say. Unichip (which does not have Audi applications) claims to modify their curves. Again, goal here is to learn…
I do know that my timing/advance, retard, injector duty cycle and AFR all look good based on logs.
I worked at Siemens (not in Automotive controls) and am going to try and get to the Simos guys through my contacts and see if I can get a response. Usually German engineers are happy to talk…
Again, I appreciate the dialogue. We are all car guys that love our Audi’s / S4’s. We are more alike than not…
Cheers
Keep in mind that you jumped into a very polarizing debate where a lot has already been discussed, so don’t be surprised that people are pretty pissy towards each other. I’ve tried to debate this with bhvdr, but he just ignored most of my posts and continued along a stubborn argument (which is pretty much completely ECU flash vs CW piggyback, which as I’ve said, is not a smart argument to justify why CW is a good solution).
If you want to see some analysis done here, go look at Jim’s post here http://audirevolution.net/forum/index.php?topic=3310.msg96401#msg96401
The figure is reproduced here
http://i616.photobucket.com/albums/tt241/jgreat_bucket/S5/map_sensor_alter_zps3btanwjz.png
which is based on CW’s post here http://www.audizine.com/forum/showthread.php/645409-B8-5-S4-stock-versus-APR-Stage-1-tune-versus-Chipwerke-piggy-back?p=10659805&viewfull=1#post10659805 with data here
http://i205.photobucket.com/albums/bb96/YILMAZM/Ekran%20Alnts.png
Regarding your points
I know for a fact that “certain OEM” fueling strategies are on the order of 1 second until steady-state lambda closed-loop mode becomes the dominant strategy. Is this how seimens (conti) does things…dunno. But the physics require some kind of calibrated open-loop stage. In the future, I suppose this will be done with MPC (model predictive control) rather than calibrated look-up tables (which are less robust to certain emissions DTC’s, which are one of the major metrics that controllers have to satisfy), which is computationally expensive for current generation ECU’s. This will probably change in the next couple years when ECU’s transition to more cores.
I’m not sure exactly how the octane is factoring into the equation, since the only way this effects the combustion is if knock becomes present, which in my opinion is an output of the combustion system. Bad octane should result in timing pull, but I don’t understand how this changes the fueling? The lower AKI (anti-knock index) is leading to larger EGT’s and thus requires more fueling?
The issue with altitude is you are having a larger delta between the ambient pressure and the boost capable post blower. The difference between the requested boost at low pressure (where CW doesn’t modify the MAP signal) vs the difference in requested boost in charge mode (audi’s terminology for bypass = closed) is probably what’s throwing CEL’s.
But in my opinion, the misfires and stumbling are from the engine input parameters being out of spec. Which doesn’t mean necessarily that the gas is bad or the altitude is too high…in my opinion it just means this is a bad way to tune things, and the people at altitude are experiencing the issue more frequently. I’d wager it’s a matter of time before it happens to someone at a different elevation or climate. So in my opinion, this is just rolling the dice.
Would be interested in what your past colleagues say. I know people in engine controls and was told this would screw up fueling and a bunch of other calculated values (EGT, torque etc). MAP or calculated MAF is a critical signal in the control of a DI engine.
I would have a lot less shit to talk if this box somehow changing ignition or fueling or something on the actuator side. But changing a measurement like this seems like a horrible idea based on my experience in controls (this is what I do for a living).
Well thats clear as mudd. So the CW is dumb and apparently all flashes are not dumb. I am the one making this out to be piggy back versus flash??? I find it interesting you state that “stock or ecu flash” is “not dumb” but then admit that you would not be the first one to get a flash and admit we had issues with flashes blowing motors and flashes that showed low fuel pressure (which as you know is a lot more serious than a simple underboost code).
All I am saying is at least lets drop these biased comments and concentrate on logs and see what is going on. Obviously for many people this is not a “dumb” mod as there have been A6 people with 30K miles on their cars with no problems and my car has 3K with it on and logging out terrifically and plenty of people have reported no problems. But this may not be right for everyone, definitely. So lets look at it without bias and see where we can find problems and areas for improvement.
It looks like every single log (all can be seen in the original post) looks pretty darn good on all of the 93 octane cars.
It looks like one issue is gas (again, only one issue). When you have a person pulling 12.6 degrees timing on a completely stock car the last thing I would do is recommend they put ANY TUNE on their car that is going to request more boost and timing. I dont care if it is CW, APR, GIAC, whatever. So we know that is an issue for SOME PEOPLE running the crap winter blend 91 octane.
The other issue for some seems to be driveability and it seems to happen mostly in the regular “drive” mode or in the “comfort” type economy mode and obviously only with DSG cars. I can help but wonder if this is some people experiencing a feeling of the “transition point” should we call it between the non modified signal and when the modified CW signal ramps up. It would be interesting to see if the change in the map creates a more smooth or earlier transition.
Just thoughts.
Mike
OK, I found something interesting here. This guy was on my committee and is a world expert in nonlinear controls. Here’s a good quote
Keep in mind that this was published back in 1994, but that doesn’t mean that the physics have changed. The purpose of this paper was to develop an air charge estimator, which could function upon first starting up the vehicle (since he mentions EGO sensors are not accurate when the sensor is not at steady state operating temps) with likely lower bandwidth/accuracy sensors and/or microcontrollers.
http://web.eecs.umich.edu/~grizzle/papers/airchg.pdf
This thread, the same as the one on AZ is just going to round in circles.
Lack of knowledge and a serious amount of blind faith is leading to nigh on ignorance/stubbornness from the CW users. You can’t argue with ignorance!!
And whoever said 6 degrees of knock retard is acceptable is talkin absolute shite, 2 is the very maximum you should ever see.
Found a picture of the board - you can see it by going into the photobucket folder from the picture the AZ vendor posted: http://s205.photobucket.com/user/YILMAZM/media/Ekran%20Alnts.png.html?sort=3&o=0
http://i205.photobucket.com/albums/bb96/YILMAZM/Chiptuning-2009-RaceChip-PRO.jpg
Looks like he has some other brand called wheelwerke?
I’m guessing by blind faith what you mean is multiple user data logs, 1/4 mile time comparisons, past history of users with 30K miles on their cars with no problems.
I do tend to agree with you and have mentioned that I dont like to see more than 0-2 degrees of retard. I dont think more than that is eminently dangerous at all but I just dont consider it tight tuning. That is the problem with these canned tunes though from the big tuners. It’s why you see alot of giac and apr users pulling 8 degrees of timing or more. I know some people dont consider it a big deal and even will track their cars like that…
[QUOTE=drob23;10679885] My car isn’t running top of the board times (12.45@110) because I think something is wrong with my plugs and/or fuel system. Lots of timing pull when I log. But I’m going to a road course this weekend and going to run the car hard. Why? Because I’ve done it before and have no reason to believe I will see anything of concern.
[/quote]
I know Jran mentioned he is pulling 8 degrees as well. I dont think his car is going to blow up over that but it would be nice to see that tightened up a bit.
One of the problems with canned tunes is that you cant plan for individual differences in cars tolerances, regional fuel quality, seasonal fuel quality, etc. I mean we have people pulling 8-13 degrees STOCK. If a programmer were to create their 91 octane program for those stock cars that are pulling 12 degrees of timing stock then nobody would be making any power at all from the program. The tuners have to pick a decent quality 91 octane and then rely on the ecu retarding or advancing as needed. Same for the 100 octane programs. If they tune for 104 octane then you will retard on 100 octane. If they only tune for 100 octane though, then you arent going to get much benefit running 109. Nothings perfect. Neither is the CW. If you have crappy gas, you are going get retard.
But, yes, i’m with you. I dont like to see much pull. Thats why I have mine down at B1 for exactly zero degrees of pull everywhere. Its why people should always log too.
Mike
I know Jran mentioned he is pulling 8 degrees as well. I dont think his car is going to blow up over that but it would be nice to see that tightened up a bit.
One of the problems with canned tunes is that you cant plan for individual differences in cars tolerances, regional fuel quality, seasonal fuel quality, etc. I mean we have people pulling 8-13 degrees STOCK. If a programmer were to create their 91 octane program for those stock cars that are pulling 12 degrees of timing stock then nobody would be making any power at all from the program. The tuners have to pick a decent quality 91 octane and then rely on the ecu retarding or advancing as needed. Same for the 100 octane programs. If they tune for 104 octane then you will retard on 100 octane. If they only tune for 100 octane though, then you arent going to get much beit running 109. Nothings perfect. Neither is the CW. If you have crappy gas, you are going get retard.
But, yes, i’m with you. I dont like to see much pull. Thats why I have mine down at B1 for exactly zero degrees of pull everywhere. Its why people should always log too.
Mike
[/quote]
Tracked my car and it ran great. I’m going for another DE in June. But then again, I have a working MAP sensor, so at least I got that going for me.