View Single Post
      03-01-2011, 05:12 PM   #187
roninsoldier83
Forced Induction FTW!
roninsoldier83's Avatar
United_States
5
Rep
214
Posts

 
Drives: E90 BMW 335xi N54 6MT
Join Date: Oct 2010
Location: Denver, CO

iTrader: (1)

Quote:
Originally Posted by vasillalov View Post
Ok, cool! In this case, if he has agreed that proper timing control (not just relying on stock DME) is of utmost importance, why on earth would he not implement it in all of his JB products from the get go? Why would he "unleash" Mike into endless blasphemy over what's proper tuning? ...

Something is not right here....
He never said having direct timing control was a necessity (obviously with these cars it's not or else we would have a lot more blown motor threads)... but the man is no fool, I'm sure he's debated with tuners like Shiv on and off for years on end, and by him incorporating CPS offsetting, well, I'm reading between the lines here

I'll just post the conversation (asked for his permission/visible with the text):

Quote:
On Fri, 2/25/11, Terry Burger <terenceburger@gmail.com> wrote:

From: Terry Burger <terenceburger@gmail.com>
Subject: Re: JB4 ignition timing?
To: "Brandon Meadows" <roninsoldier83@yahoo.com>
Cc: "Mike @ N54Tuning" <mike@n54tuning.com>
Date: Friday, February 25, 2011, 5:37 PM


Hi Brandon,

Hope all is well. One of our dealers, Mike @ N54Tuning, forwarded a forum link created by you asking for my input. So I figured I'd just email you directly.

I think your post covers the basics and there is certainly something we can strongly agree with. Avoiding detonation is top priority. I used to tune 500-600whp supercharged LS1 motors, with 10:1 static compression and flimsy pistons, so I'm very familiar with how damaging detonation can be. It's hard to tell from your writing whether you're lecturing the public on detonation, or me, but I certainly hope that was not directed to me. I did not just wake up one morning and release these products. I've been tuning for almost 15 years. With regard to N54 motors I've tuned more than any other tuner.

Anyway while much of what you write is correct, there are areas you've taken a lot of liberty, maybe because you are new to this platform, maybe becuase you don't like my tuning, etc.. One big area you and most don't understand is how the timing system on this platform works. It's much different than a high/low table system you might have tuned in the past. It uses an adaptive knock threshold system and it's very good at running timing right on the edge of knock threshold. That was how the DME is designed to work and it's difficult to get it to operate any other way. Although it probably can be done.

A year ago or so I developed an N54 flash tune with hopes of having users load at home using technology from a partner that ultimately didn't pan out. Here's an attachment of the interface and a stock map. I personally found it is extremely difficult to produce the behaviors I wanted with flash tuning on this platform. Part of that was is a limitation in the software I was using and identified tables, but most of it has to do with inherent DME logic. For example I like to add boost/advance during very cold IATs and lower it during very hot IATs. Instead I could only raise or lower an advance factor based on IAT and had to accept boost would only work with the opposite effect. Less when cold more when hot. On advance I found the tables are established as limits, analogous to a high timing table on say an LS1, but that timing could and did float above those limits. Ultimately it became clear the adaptive system has a huge authority range to set timing based on knock feedback. On pump gas in 4th gear sometimes I'd see a smooth 4 degrees, but with race gas it would float up to a smooth 10 degrees, with a limiter supposedly set to 5-6 degrees. In this platform that is a huge variance. Anyway my point with all this is I hope you have more luck with flash tuning the platform than I did. I found it not to be very compelling compared to what we can deliver with the piggyback systems. Especially when you consider the feature set this market desires (in dash controls, gauges, meth integration, etc).

I see in that thread you guys picking on the JB4s approach to timing management, which is to largely let the DME run it. The DME has proven itself capable of catching knock very early in the process and capable of sinking 2-4 degrees of advance globally in to a long term octane offset. We have hardware here that allows CPS offsetting and it does offer some advantages in areas where the DME's logic is counter the what you want to do from a tuning perspective. Like with nitrous. But largely there have not been big gains seen in terms of output or consistency in the 13-14psi range most run. I also noticed you referenced torque drops in dyno charts. If you look at corresponding logs you'll see they are primarily throttle / boost control related. Remember logged timing is only one cylinder and has only a minor effect on output. Knock on the other hand has a very distinct dyno footprint. AFR goes rich, power levels out for 1000 RPM, etc. It takes a trained eye to spot throttle closure vs. knock on a dyno chart. As an aside you'll find controlling boost and throttle to be a major pitfall with flash tuning. But I'll let you find that out for yourself.

Ultimately it's a free country and I don't have nor do I want any control over your opinion or what you state. But keep in mind I've been at N54 tuning for 4 years now, almost as long as anyone else, and I've become very familiar with it. We have almost 10,000 N54 tunes out there in use and if a customer has a problem their first call is always to me for answers. The problems you and others predict with detonation damage have not materialized not becuase not enough time has passed but because the DME management of advance is that good. We both know if knock was a factor these motors would not hold together more than a few passes and the plugs would show clear signs of distress. I've personally found the real challenge with tuning this platform is preventing users from pushing boost/power beyond their octane limits. And that is exactly what we've accomplished, in an early stage, with our new map 5.

Best,
Terry
Quote:
On Fri, Feb 25, 2011 at 6:11 PM, Brandon Meadows <roninsoldier83@yahoo.com> wrote:
Hey Terry,

Thanks for the email, it's appreciated. I also appreciate your insight and hearing a bit of background on yourself. I've actually played with a few LS-motors myself (my dad currently drives a C5 Corvette/LS1, and my brother rocks an LS2 powered GTO), good to hear.

To be very clear, I have no problems with you, nor was my thread directed towards yourself in any way, shape or form. As a matter of fact, I said as much more than once, and mentioned more than once that while I do not believe letting the DME handle timing blindly (manipulating the MAP sensor/not allowing it to see full load). If you check the "two threads" combined, you'll even find compliments in regards to your overall level of customer service (even had a negative comment about me "kissing ass" over it haha).

When Clap135 started the uber-popular Mike/Terry thread, I saw what I believed to be a lack of clarity in how timing works on a modern EFI car. Comparatively, the BMW community at large does not seem to be as savvy at grasping basic tuning concepts as some of the other communities I've been fortunate enough to be a part of. As such, I created a thread that had nothing at all to do with you or Mike. It was originally intended to be a "clean" thread, as I simply wanted to keep trash talking to a minimum and educate N54 owners on the basics of timing. Wasn't intended to start a tuner war, or bash your product.

However, we do have an area of fundamental disagreement, but as stated previously, this has nothing to do with you. I respect that you have been doing this for several years now, and have sold numerous N54 based piggybacks. With that said, while I do agree that the stock DME appears to be capable enough to catch knock at its lowest levels in order to save the motor for catastrophic failure, I do believe adjusting timing settings can increase overall consistency. To be honest with you, not to start a tuner war, but a couple of your competitors appear to feel the same way. Granted, they are your competition, but everything in my tuning background leaves me inclined to agree with them.

In truth, I ran the JB3 (map 3) for a very short while... and honestly, because it works on a simple multiplication algorithm, I think it had a very tough time adjusting to my elevation (5200 ft-Denver). In the cold, inconsistent would be the nicest word I could use to describe it. We're talking, pulls like a freight train for a second, and drops down to seemingly stock power, almost like hitting a brick wall. I pulled quite a few logs (ala BT scanner) and timing was all over the place on the "clean runs", whereas during the "hitting a wall" runs, boost dropped frequently as well.

If I could actually keep track of knock count accurately on this car, I have a feeling I would have not liked what I saw (I honestly don't trust the "knock flag" feature). Either way, after the way the JB3 behaved, I removed it from the car. I actually put it up for sale over on E90post, but then I realized that I qualified for a JB4 upgrade (didn't have the JB3 long at all), and was curious to see the differences between the two, so I held onto it. I received the JB4, swapped the pins on the wiring harness/added the power wire, and slapped it in the car.

The next day, I went to a local dyno, co-owned by a friend of mine (Nick at MAC Auto), and slapped the thing on the dyno. It made decent power on that dyno (more than my brother's LS2 GTO when it was stock, but at the same time, the dyno is at 6000 ft elevation, turbo cars have a HUGE advantage up here, so it's not exactly fair haha!!). However, I didn't care for the look of the torque curve (very choppy, not nearly as smooth as any car I've ever tuned on that dyno). I drove the car with the JB4 (map 1) for a few days, and I will say it never drastically cut boost like the JB3, but consistency was not on par with many other turbocharged vehicles I've driven/tuned, and logging again yielded random drops in timing on a fairly regular basis (pretty much any 3rd or 4th gear pull). As such, I pulled it off the car as well.

I live at a high elevation, where only 91 octane piss water is available (well, I used to run E85 in my old STI, but obviously without extra fuel overhead that's not an option for my N54 haha). From my experience, OTS maps on many platforms typically don't play too nice with our conditions up here.

I have tuned with Cobb's ATR software in the past (on Subaru's, EVO's & Mazdaspeed's). I'm very familiar with it, and I know what I have been able to do with it in the past. Granted, this is a different platform, with it's own unique quirks that I'm sure I'll run across along the way, but at this point, the next logical steps would be one of the following:

1. Contact you, and email you logs back and forth attempting to correct consistency issues with the JB4, without addressing timing. Honestly, this has never been my style. I prefer to tune my cars for myself, I'm not a big fan of OTS maps, and I highly value user adjustability.

2. Purchase a Procede and assume autotuning will resolve the issues I'm having. Again, I prefer to tune myself. I don't care for the idea of paying extra money in order to toggle settings myself, I also don't care for the VIN lock feature, making it harder for me to sell it, and in truth, while I've tuned several piggybacks in the past (mostly UTEC's, Xede's, ect), I've always preferred flash tuning for a number of reasons.

3. Purchase Cobb AP and tune the car myself with ATR software. I'm familiar with the interface (to include their hot keys ect), I've worked with numerous AP's in the past. I prefer flash tuning, there's no VIN lock, no need for me to rely on anyone else to tune my car remotely, and I honestly believe I can smooth out the torque curve and make the car more consistent by adjusting timing settings (along with boost and fuel of course).

Maybe this is foolhardy of me, maybe you're correct, in that the DME will exceed base timing values (Rob at Cobb disagrees), but it is largely uncharted territory on this platform and I would like to see for myself. You have your experiences, and I have my own experiences, which is to say that I believe I can make a solid timing table on this vehicle that will increase the overall consistency of the vehicle. Truly, there's only one way to find out, which is my intention.

Now, with all of that said, please realize I have not brought my negative experiences (JB3/4) to light over on E90post... this truly was not meant to be an attack on you or your company, and honestly, I would have tried to give as much of an unbiased opinion as possible if your old buddy Mike hadn't repeatedly interjected himself into the conversation. I understand he means well, and is trying to defend his product, but doing so as often as he does gives off a certain type of impression.

It's one thing when someone makes a thread called (Mike/Terry can you please join the discussion), and it's quite another when someone is trying to simply educate the population on how timing works, and help people to at least understand why some members have such strong feelings about this topic.

With all of that said, I haven't sold my JB3 or JB4 as of yet... there's a reason for that. I have planned on doing a true unbiased comparison eventually. Once Cobb releases '07 software, I intend on purchasing an AP, and making a couple pulls on the same dyno in comparison. I also intend on trying to log both methods of tuning on the same day, in the same conditions, on the same tank of gas. I am a man of my word, and as such, I don't intend to show the "best" log from one, and the "worst" log from the other, but instead to make a few pulls with each on post each log for comparison. I figure, let the information speak for itself, and test many theories.

At that point, when ATR software is released, I intend on dyno tuning my car with ATR software, and comparing the tuned dyno chart and datalogs to both OTS maps (JB4 & Cobb AP). My goal is to hopefully be able show the value in custom tuning a vehicle. If you're correct and the DME will not allow me to fully adjust the timing curve (I doubt it, but we'll see), I will still gladly share logs and dyno charts with the community, and I will be out a good deal of money haha!! However, I'm a big fan of the statement "put your money where your mouth is", so that is exactly what I'm doing. Based on past experience, along with the concurrence of other N54 tuners, I believe I can tune a better map for my own car than the OTS maps currently offered, based mostly on being able to set my own timing curve. I suppose we'll see shortly haha.

I honestly understand completely why you would want to keep most users from pushing the limit and blowing a motor, so I understand why you would want to limit users from turning up the boost beyond a certain point haha. Someone blows a motor with your piggyback, even if it was completely their fault, and all of the sudden you'll be hearing "JB's blow motors all the time, they're not safe, ect ect", it's bad for business to say the least. I respect you communicating with me like this, but again, this had nothing to do with you, please don't take it personally (it doesn't sound like you are haha, but just want to reiterate that).

I know you can't post on E90post anymore, but if you would like, I can certainly make your correspondence known over there (word for word, no cut/edit nonsense)?!? If not, I completely respect your privacy and the privacy of this conversation, and am not the type of person who would post this type of communication up maliciously. Either way, I wish you a good night, and a good weekend sir.

Cheers,

Brandon
Quote:
On Fri, 2/25/11, Terry Burger <terenceburger@gmail.com> wrote:

From: Terry Burger <terenceburger@gmail.com>
Subject: Re: JB4 ignition timing?
To: "Brandon Meadows" <roninsoldier83@yahoo.com>
Date: Friday, February 25, 2011, 7:35 PM


Hi Brandon,

Thanks for the reply and I can understand your position. On the JB3 boost control at elevation, there were definitely some issues there that often required car to car custom tuning. That was one of the main catylists for designing the JB4. I'm glad you were able to take advantage of our free upgrade program. I would certainly be interested in seeing your JB4 dyno and logs. There are areas of the curve that the knock sensors are overly sensitive, around 4200rpm, and other areas from 4800-6200rpm that clearly indicate overly aggressive tuning (advance and/or boost). Other variables such as throttle position and load also play in to the timing curve. Map 5 development is going well but we haven't spend much time optimizing that at elevation yet. I do expect it to work well though. When the boost curve is correct for the octane, afr, mods, etc timing optimizes itself perfectly without knock related drop outs. The key is designing the proper autotuning algorithms to get every car perfectly tuned continuously.

On the flash tuning, Cobb has more extensive flash resources available to them so it is possible they are using tables I didn't have access too. It will be interesting to see how their product evolves and whether the self tuning takes off. Much of that in my opinion will depend on whether they can deliver the high boost / high performance maps users have become accustomed to with their piggybacks. Also shade tree tuners are normally used to tuning in terms of boost, afr, and advance. N54 flash tuning doens't really lend itself to that. You wind up tuning duty cycle, load target, an offset form of lambda, and then hit advance over many additive tables. No doubt many will catch on but there is still that big inherent problem with how load scales to baro, air density, etc. A great tune @ 70 degrees ambient may cause major problems @ 100 degrees ambient. We'll see.

On my email above, you're free to post it if you'd like, but most who are interested in what I have to say visit our n54tech forum to hear it from the horses mouth.

Best,
T
Quote:
On Fri, Feb 25, 2011 at 7:34 PM, Brandon Meadows <roninsoldier83@yahoo.com> wrote:
Hi Terry,

I've attached a copy of the smoothest pull I was able to get in on MAC's dyno with the 335xi. For reference, I also attached a copy of a dyno of my old STI with bolt-ons on the same dyno. However, in the STI's defense, I was time limited that day, and cleaned up the AFR's a bit more on the road afterward (I used to rock an LM-1 wideband/road tune). As you can see, both graphs have the same scaling (2000-7500), both also have smoothing set to 0, so it is a direct apples to apples comparison. Needless to say, I believe I can work on smoothing out the tune a bit on the N54, haha. No car I've ever tuned has ever looked that choppy when I was done.

Granted, the dyno is at 6000 ft elevation (Parker, CO), so both are displaying a grossly overrated correction factor, as both are turbocharged, and turbo cars never lose as much power at altitude as NA cars, and the dyno only spits out either fully corrected or uncorrected numbers, not that the peak values matter much.

Honestly, I would love to send you logs, but unfortunately I didn't save any (likely story right? haha). I know, I know, "very convenient" haha, but honestly, I didn't plan on posting them, but I give you my word, if/when I put the JB4 back in, in order to do a comparison, I will send you the logs, as well as make them public. Although, I will tell you, on pretty much every 3rd or 4th gear pull I made, timing seemed to go negative at least a couple times. Wasn't very pretty at all.

I've talked to Rob at Cobb a few times, and he's pretty confident that as long as the car is "happy", he tends to see base timing values across the board when logging. Cobb has been doing this for quite a while now, I have a feeling that they may have this one figured out haha. However, I won't lie, while Cobb has always been good to me in the past, I've held no loyalty to them. I rocked an AP on my '04 WRX and my '06 STI, but I eventually sold my STI AP, and converted over to OpenECU. Nothing against Cobb or their software, but OpenECU is free (well, you have to buy a $70 tactrix cable), allowing me to recoup some cash (selling the AP), and by that point, I had played with OpenECU on several cars, and become very comfortable with it.

In my honest opinion, I see most communities eventually going this route. I've been part of multiple communities, and they pretty much all start out being dominated by piggybacks... then flash based tuning becomes available, and piggybacks slowly start to fade away (DSM's/DSMLink, Mazda's, Subaru's, EVO's, GM's/HPTuners, all the Diablo stuff ect). There seems to be a bit of a pattern. Ultimately, when a community grows to a certain level, and members become more active, people start hacking ECU's themselves (I'm a big fan of open source haha). Honestly, from one enthusiast to another, depending on what you see from Cobb, it might not be a bad idea to look into flash tuning again. Just my $.02

To be honest, I'm actually pretty excited about learning something new/different techniques with these motors! I have a BS/IT, and while tuning is only a hobby of mine, I've been programming machines for a living for the better part of a decade (IT & telecom), so I'm always a big fan of learning something new and figuring out how a new machine works/getting it to do what I want it to do. A couple good friends of mine tune for a living, and I must say, you're a brave man haha! It seems like for them it's either feast or famine on a constant basis, which is by far the largest factor why I never wanted to work in that field for a living. Again, you're a brave man haha!

Either way, have a good one, and thank you for your time and insight.

-Brandon
Quote:
On Fri, 2/25/11, Terry Burger <terenceburger@gmail.com> wrote:

From: Terry Burger <terenceburger@gmail.com>
Subject: Re: JB4 ignition timing?
To: "Brandon Meadows" <roninsoldier83@yahoo.com>
Date: Friday, February 25, 2011, 8:53 PM


Hi Brandon,

My guess on the dyno is throttle trimming around VANOS. Given its smoothing 0 I don't see any signs of knock up top. Was this before our autoPID system? Did you log it by chance? We made some adjustments there to allow it to adapt to various conditions. Negative timing is normal but frequent sharp drops above 4700rpm do indicate increased knock activity and need for attention. Although I will say, we've had customers run 1-2 years with 2-3 dips per gear, against our suggestions, and I've personally seen the plugs after. No metal flakes, melting, or even uneven wear. Crazy stuff.

Cobb is a big company and well respected. I wish nothing but the best for them with this platform. I would be happy to offer a BMS tuned Cobb flash at some point in the future should I see something in the flash technology that I can't achieve with piggyback technology.

When you do get the flash, we offer a tool that gives you an in dash boost gauge, timing gauge, shift light, etc. Fun to have around when tuning: http://burgertuning.com/CAN_reader_tool_for_BMW.html

Tuning is a cut throat business but I've always enjoyed helping people make their cars faster. I come from an IT background as well, BSCS, several years as a software engineer and later an IT director, and then COO, before deciding desk jobs were not for me. Was at the dragstrip last night tuning and ran in to a local raving about his JB3 and suggesting I should get one. Good times. Makes it all worth while. Truth be told most of our profit does not come from the JB3/JB4 product line up but 90% of our forum drama does. I've gotten used to it over the years.

Best,
T


^^^Those were all the pertinent correspondences from our conversations that related to this thread in their entirety (no cutting/pasting/editing).


EDIT: In reference to the emails above, just so everyone knows what I'm referring to when I'm talking about curves, here is the smoothest run I was able to achieve running a JB4 map 1:



By comparison, here is a pull I made in my old OpenECU tuned STI, on the same dyno, using the same scaling (2000-7000rpm):



^^^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
__________________
E90 BMW 335xi 6MT N54 - SOLD
Kawasaki Ninja 650R - Play Toy
Former cars: 2009 Infiniti G37x, 2006 Subaru STI, 2006 Mazdaspeed 6, 2003 Nissan 350Z, 2004 Subaru WRX, 1991 Toyota MR2 Turbo, 2nd Gen DSM & 1990 Mazda Miata

Last edited by roninsoldier83; 03-01-2011 at 05:56 PM.