[scheherazade]

This is moot.
We have no proof that knock was the cause of this car's cylinder 6 failure.
And if there was knock, what the root cause of the knock was.

-scheherazade

[carcars 335i]

Originally Posted by scheherazade

For an evo/sti, anything over 400 whp and you'd only be 'safe' with new pistons/rods.
Not just for the compression reduction.
I've seen engines in the ~500 whp range on stock internals, and they *eventually* threw a rod or cracked a land.
Sometimes 6 months, sometimes a year, sometimes less.
It didn't happen due to ping.
It was too much power on stock internals, and something eventually broke.


You can't blame anything on ping without looking at the piston and finding the signs.
For all we know this guy had a poorly manufactured piston and it came to bare after he ran a tune.
God knows we have people with leaky blocks, what else could be poor from time to time...

Until you do look at the insides, just chill.
You're using fear to run a propaganda campaign with this failure against BMS.
Anyone could have suffered this failure - one of YOUR customers could have had a 'relatively poor' piston, and could have broken - then what? Is Procede crap?
It's dirty to use it against your competitor.
Especially with THOUSANDS of cars running tuned, and this particular car being ran to the extreme, you're blowing this way out of the water.

Like everyone else, YOU should wait to pass judgment until you have something to judge.
Who knows, maybe you're right, maybe you're wrong.
If you don't know, you shouldn't speak.

I wouldn't have a single issue with you saying everything you're saying, if you had some data to back it up.
Pictures of this car's internals, logs of this car, etc.
But you don't. No one does.

Frankly, no one can say that the meth ever stopped. *Maybe* it worked fine the entire time...

Let's all keep in mind that <almost> 20 PSI all the way to redline is not the typical tune.
This is <almost> 2.5 times the normal pressure (+plenty of heat).
This *could possibly* be a much more common story - all around - if <almost> 20 psi were common.
We just don't know.

For that matter, there are a few (very few) cars running BMS tunes with more aggressive setups - and they're still in one piece - and have been for longer.
That alone speaks to uncertainty from one car to another.

-scheherazade

[stressdoc]

Here are my thoughts: 1) We won't get the failure analysis results on this case. That information is proprietary. 2) The pattern of (very rare) N54 engine failures indicates a weakness in cylinders 5&6. This appears to be heat-related. 3) BMW engineers their engines to withstand a wide range of conditions, including the rigors of sustained high speed loads during autobahn travel.

The OEM engine parameters, including boost levels, are decided on the basis of performance/reliability/cost algorithms. The stock 335 is a phenomenal car, providing extraordinary acceleration, handling, and braking for a relatively low cost. Aftermarket engine tunes alter the balance. In the case of the N54, dramatically so. Power increases of >60 hp for under $1k. We now have the potential for supercar performance at a bargain basement price. However there is an inevitable trade-off in reliability. In normal dd conditions in the USA/Canada, and other countries with low speed limits, the reliability consequences are likely small. In racing conditions and high speed driving on the autobahns, reliability would suffer more, in particular because of the difficulty in providing even cooling to cyl 5&6.

Look at what BMW engineered in the 335is and the engine performance kit. The focus is on cooling -- higher output fan, changes to software for H2O pump, ducts, additional radiator, ...

[enrita]

Originally Posted by stressdoc

Here are my thoughts: 1) We won't get the failure analysis results on this case. That information is proprietary. 2) The pattern of (very rare) N54 engine failures indicates a weakness in cylinders 5&6. This appears to be heat-related. 3) BMW engineers their engines to withstand a wide range of conditions, including the rigors of sustained high speed loads during autobahn travel.

The OEM engine parameters, including boost levels, are decided on the basis of performance/reliability/cost algorithms. The stock 335 is a phenomenal car, providing extraordinary acceleration, handling, and braking for a relatively low cost. Aftermarket engine tunes alter the balance. In the case of the N54, dramatically so. Power increases of >60 hp for under $1k. We now have the potential for supercar performance at a bargain basement price. However there is an inevitable trade-off in reliability. In normal dd conditions in the USA/Canada, and other countries with low speed limits, the reliability consequences are likely small. In racing conditions and high speed driving on the autobahns, reliability would suffer more, in particular because of the difficulty in providing even cooling to cyl 5&6.

Look at what BMW engineered in the 335is and the engine performance kit. The focus is on cooling -- higher output fan, changes to software for H2O pump, ducts, additional radiator, ...

seems certainly cyl 6 goes hotter but in my case it was cylinder/piston 2 that failed. All other pistons were pristine.
I am running now the exact same injectors as before. (changed HPFP though)

[Mike@N54Tuning.com]

Originally Posted by AlterZgo

On the surface, these statements seem contradictory. As a person interested in purchasing a flash or a piggyback unit, I would appreciate a technical explanation.

According to Mike/Terry, the JB3 runs much less timing and it is readily apparent in the BT logs. Yet Shiv indicates the JB3 does not monitor of A/F ratios, timing/knock, engine temps, etc.

Shiv and/or Mike/Terry - Please answer the following to help me make a purchase decision:

1. Even if the JB3 does not monitor those critical functions, doe the JB3 still pull timing? If so, does it do this purely based on the level of boost? If it doesn't pull timing based on boost, what is used to calculate the amount of timing retard to apply?

2. Shiv - Assuming the JB3 is pulling timing dependant on boost, while not as sophisticated as what your auto tune is doing, in your opinion, is this still a reasonable approach (assuming people are not pushing their cars to the ragged edge)? If not, why? Would there be a relatively safe level of max boost where boost dependent timing controls would be a reasonable approach?

3. Mike/Terry - You've stated previously in this thread that the Procede V4 does absolutely the same thing with timing that the JB3 does and this is confirmed in the logs. Shiv indicates the logs do not accurately record exactly what the Procede is doing (there's an inference that additional timing is being pulled in a manner that cannot be logged as a way to protect his technology). Mike/Terry - Can this be factually disproved by you? (Shiv, I'm not asking you to share how this is done as it would defeat the purpose of hiding it in the first place.) Is there a way timing can be accurately measured in another method that can validate or disprove Shiv's claim that the Procede V4 is indeed adjusting timing in a different manner than the JB3?

Thanks.

You'll want to read up on CPS offsetting. The V4 does it and the JB3 goes not. Offsetting alters the maximum allowable timing advance should the knock threshold system allow it. So as a theoretical example at say 4000rpm under some conditions the maximum allowable factory curve is 11 degrees. That doesn't mean you get 11 degrees at full load. Only that the timing can float as high as 11 degrees IF the long and short term knock threshold systems allow it to rise that high. A stock car might normally run around 8-9 degrees there on 93 octane. If you add 100 octane the stock car timing will float up to
that maximum of 11 degrees. The system is designed to always ride the knock threshold system which has short term (knock detection, run to run) and long term (octane adaption, over many runs) factors.

Offsetting is used as an advance limiter preventing timing from climbing up to the maximum target should the conditions allow. If the offset induced is say 2 degrees and the OEM maximum target for those conditions is 11 degrees then the highest your advance can go is 9 degrees. Say a tuned N54 on pump wants 6 degrees at peak torque @ 12psi on 93. The highest timing could go on a JB3 would be the full 11 degrees IF it cleared the long and short term trims. The highest the 2 degree offset tune could go would be 9 degrees under those same conditions. In practice though both run the 6 degrees dictated by the knock threshold system. So the argument is that the JB3 is "riding the knock system" 5 degrees while the CPS tune is "riding the knock system" 3 degrees. Generally the offset can't be brought down enough to eliminate the knock threshold system as it completely castrates the ECU's adaption range and ultimately your performance. It's generally useful for limiting preignition in higher power applications (although is not used in this manor currently) and has some benefits in short term timing changes during map switches and transitions. The JB3 G3 handles these another way. By limiting boost instead of limiting timing.

To answer your questions:

1) The JB3 G3 box can't itself monitor knock activity or air/fuel ratio. So it is up to the high HP user pushing the boundaries to monitor this via an external tool like the CT to display them on dash or the BT for logging to a laptop. The upcoming JB3 G4 box brings full CAN logging in to the mix and is really light years ahead of the G3. It has autotuning based on timing advance and other factors, an adaptive and fully integrated feed forward boost control system, 1k fueling resistors so it can target the full range of air/fuel ratios, etc. Overall a huge leap forward over the G3 and will still retail for $349 for the pin out model.

3) When I say the tuning on meth is similar I mean that the procede is not inducing a CPS offset under these conditions which is clearly visible in the logs. This business about a significant hidden timing pull not shown in the logs is pure fantasy and stems from short term CPS offsets induced during map switching, etc. Nothing to do with timing during a sustained high speed run. A dual position scope can be attached to a car with the V4 to monitor the CPS going in to and out of the box but hopefully it will not need to come to that. On meth both systems rely on a flow sensor system to revert to a failsafe mode before meth flows and in the event of a drop in flow during a wot run. The ECU is fast to pull timing but at very high boost levels it can't go from full advance (say 14 degrees) to where it would need to go (say 2 degrees) fast enough so you open yourself up to knock there if great care isn't taken. As a general rule race gas is much safer for high boost levels and is suggested by both tuners over certain power levels. For BMS that is at 17psi.

Mike

[Mike@N54Tuning.com]

Originally Posted by shiv@vishnu

So it's not the absolutely ridiculous tuning approach that no legitimate engine tuner would ever adopt or recommend. But rather a failed injector?

When StartupJunkie's AR brake line split on him at the end of a 125mph 1/4 mile pass, neither of us were quick to blame anyone. Because we both knew that no one intentionally design an faulty brake line. It was just a fluke. And sh%t happens. We were just happy no one got hurt. And it didn't stop either of us from ordering more lines from you.

But in the case of the JB3, you have a completely inadequate tune. One that is promoted to be a "max power" tune capable of supporting methanol and (gasp) nitrous. Yet it simply ignores 1 out of 3 ingredients that even the most basic tunes should have. That, my friend, is marketing. Not the stuff I'm saying. Everything I'm saying here is supported by basic engine theory.

I know BMS sells your parts but you can't honestly tell me that simply raising boost on a turbo engine will no regard for output (AFR, ignition timing, engine temp, etc,) is anything close to a sound approach. Does it work well enough for most people who have never tried a proper tune? Yes sure. But that doesn't make it any less incomplete than it is. Because engines will fail. Sure it may take some rough operating conditions as enrita and sevak have shown. But even their failures can be avoided by using a tune that conformed to basic tuning principles that even the most n00bish tuner learns at this first day on the job.

Shiv

Like Maxboost's motor?

Mike

[kenneth]

This is my take on this, and I'll caveat my thoughts by saying I am no expert in tuning, so to be taken with a grain of salt. No need to bash me either!

The two engine failures seem to be related to running agressive set-ups, without the required failsafe/hardware/fueling in place, but which (to me) is not representative of what most folks are using their tunes for.

The real question (again to me) then becomes: is knock present for folks running the jb3 (and the procede) for the majority of folks not running extreme set-ups? How can we measure if this is or isn't happening and under what conditions they occur? If this cannot be proven, although it makes theoretical sense, maybe (again to me) then there's nothing to worry about either tune?

[Ilma]

You know Terry/Mike.....

There is a lot of useful info in this thread that is not even related to Sevaks scenario.

In fact, I feel what you say is true.....that with high octane/meth, even the Procede is not inducing ignition correction as the octane takes care of things so it is a moot discussion.

But for those of us NOT running meth, timing control becomes even MORE relevant because we don't have the octane taking care of the knock.....only the tune.

So in that regard, octane and knock control is VERY relevant to us daily drivers.

Neither tune can eliminate knock entirely....as you say, the timing dip around the vanos shift seems to be pretty common on non-meth applications.

But I have illustrated in the above graphs, how some degree of ignition retard can reduce the number of instances of knock.

To me that is a good thing......but to you it seems irrelevant.

You seem to condone that the engine was meant to induce knock.

While that may be true as a temporary measure, I don't think I would put 87 octane in my gas tank and run it long-term.....yet according to your thinking, the DME will take care of it by pulling timing.

The question in my mind is just how much knock do you want to invoke?

A knocking engine just doesn't make sense unless that knock is harmless in some way. If it did, then why does it say Minimum octane = 91 on my gas cap?

[scheherazade]

Originally Posted by shiv@vishnu

The root cause of knock is too much ignition advance and/or boost for the given conditions. This is was confirmed to be the reason Enritas engine failed as well. It's pretty easy to see a relationship here. Especially given the "crank up the boost and let the dme sort out the knock" approach used in both cases.

Shiv

There is Enrita, and there is Sevak.
Don't get confused.

There is no "as well" as of yet.
There is no "relationship" until you can present specific data showing that there is a relationship.

There are too many "what if's" in this case to peg "ignition advance" as a root culprit.
Which you are trying to do, since your product offers more control over ignition advance.

You're riding Sevak's loss as an advertising opportunity.
It's dirty.

-scheherazade

[scheherazade]

Originally Posted by Ilma

...
You seem to condone that the engine was meant to induce knock.
...
A knocking engine just doesn't make sense unless that knock is harmless in some way. If it did, then why does it say Minimum octane = 91 on my gas cap?

BMW designed the DME to induce knock, back off, learn, and adapt.
BMW condones it.


You can't make advertised power with timing relaxed too much.
If you want an engine to be equal power on lower effective octane, you're gonna have to have a larger engine with less effective compression.
(All things being equal, presuming you want to preserve equal safety headroom for both engines)

-scheherazade

[Ilma]

Originally Posted by scheherazade

BMW designed the DME to induce knock, back off, learn, and adapt.
BMW condones it.


You can't make advertised power with timing relaxed too much.
If you want an engine to be equal power on lower effective octane, you're gonna have to have a larger engine with less effective compression.

-scheherazade

Gee....that's strange.

I run quite a few datalogs including some that are at near-stock boost levels (like around 10 psi on pump gas) and the only knock I ever see is that vanos-shift.

I think Shiv is right on this one.

My datalogs confirm that the DME always targets advance of around 7.5 degrees mid-rpm up to 12 degrees near red-line. This never seems to adapt itself out.

I never saw any knock in between those values running near stock boost values.

So what I conclude is that maybe BMW designed a timing curve that in most cases will be close enough to the threshold as to not induce knock on stock boost levels and normal operating conditions.

That seems to leave the knock sensors as a backup/safety system should something get out of norm.

Like, say....you pour in 87 octane.

Seems to me that you are stating that BMW designed their engine to knock and that this is safe?

If that is so......how much knock is considered safe (ie number of instances and number of degrees of timing pulled).

[scheherazade]

I would be very surprised if the default values in the DME were not for a typically-adapted car.

What we're talking about is a system which allows you to take your car from 93 octane in greenland down to 87 octane in death valley - and back - managing to adapt to both situations. Preserving safety and performance.

-scheherazade

[Ilma]

Originally Posted by scheherazade

I would be very surprised if the default values in the DME were not for a typically-adapted car.

What we're talking about is a system which allows you to take your car from 93 octane in greenland down to 87 octane in death valley - and back - managing to adapt to both situations. Preserving safety and performance.

-scheherazade

Funny....I have run 91 octane, 94 octane and 98 octane.

On 91 octane, my logs showed a knock event that was absent on 94. This was running 11 psi with no ignition correction.....so all DME controlled.

With boost at 14 psi and 98 octane my timing curve will still go from 7.5 degrees up to 12, but sometimes it hits around 10 degrees at throttle transition and advances up to 12 from there.

So I can see how the DME is always trying the stock timing curve, listening for knock and reacting appropriately for the conditions.

I have sometimes done 3 to 4 sequential pulls and find that knock increases as the engine heatsoaks, but that damn DME keeps targeting the stock curve resulting in a lot of timing dips and knock retard.

I wish it wouldn't do that.

Also, I seem to recall that most manufacturers will advise against running lower than their specified octane for other than short periods of time. I can only presume this is because of the increased knock activity and it's long term effects.

Even if long-term adaptations do happen, at some point the DME is going to keep testing to see if the recommended octane is back in the tank.

[scheherazade]

That makes sense to me.

Each run is hotter than the previous.
I can see how the previous run's adaptation would be inadequate for the next run.

I don't know the actual adaptation duration.
I just know it takes a little while to adapt enough to not have to pull timing anymore - for any one change.
Extreme/repeated changes should only make it take longer.

-scheherazade

[Ilma]

Originally Posted by scheherazade

That makes sense to me.

Each run is hotter than the previous.
I can see how the previous run's adaptation would be inadequate for the next run.

I don't know the actual adaptation duration.
I just know it takes a little while to adapt enough to not have to pull timing anymore. Especially if there is an extreme change.

-scheherazade

Oh, I don't mind the timing being pulled at all....in fact I welcome that.

It's just that I don't like it constantly happening because of a knock event.

It's the knock that bothers me.....not reduced timing.

In any case, I just tune for as close to the stock timing curve as possible, given my boost and octane.

I think this is the safest approach until someone proves otherwise. Especially considering that the stock curve is most likely designed to give the best mechanical leverage of the piston onto the crankshaft.

As far as adaptations go......I think most of us who drive passively in traffic and then go WOT on occassion will never get the benefit of any long-term adaptation. Conditions are just too dynamic in daily driving for any kind of adaptation to stick when you go from one extreme to the other.

So really, all we are left with is short term adaptation......which is knock reaction, no?

[Ilma]

This is some good information on the differences between pre-ignition and detonation:

http://www.contactmagazine.com/Issue...ineBasics.html

Here is a snippet: "Engine test data determines how much compression an engine can have and run at the optimum spark advance.

For emphasis, the design compression ratio is adjusted to maximize efficiency/power on the available fuel. Many times in the aftermarket the opposite occurs. A compression ratio is "picked" and the end user tries to find good enough fuel and/or retards the spark to live with the situation...or suffers engine damage due to detonation"

[scheherazade]

Originally Posted by Ilma

So really, all we are left with is short term adaptation......which is knock reaction, no?

Knock reaction is an immediate response (short term timing pull).
Timing targets are slowly adjusted by every knock. The adaptation takes time.
This will happen until the targets are such that there is no more knock.

Timing targets are stored in tables. Not a single value.
Adaptations stick to the portion that they belong to. WOT in WOT, cruising in cruising.





The issue is not whether timing/knock will be taken care of (short or long term).
It's whether you want BMW's system taking care of it, or if you want Vishnu's pre-emptive approach in combination with BMW's system.

-scheherazade

[Sniz]

good discussion guys, keep it up.

[Ilma]

Originally Posted by scheherazade

Knock reaction is an immediate response (short term timing pull).
Timing targets are slowly adjusted by every knock. The adaptation takes time.
This will happen until the targets are such that there is no more knock.

Timing targets are stored in tables. Not a single value.
Adaptations stick to the portion that they belong to. WOT in WOT, cruising in cruising.

-scheherazade

Your statement sat in the back of my mind and didn't seem consistent with my experience.......so I went out and did another couple of datalogs just to double check.

I am running around 14ish psi on 98 octane....no meth at all.....1 degree of ignition correction at mid rpms.

This first pull of the evening shows a knock event just as I shifted into 4th gear and timing immediately fell from 9.5 to 6.5 degrees.

So according to your thinking....the DME should remember and adjust to this knock event by adapting timing.

[Ilma]

But much to my surprise.....after driving around for 15 minutes to let engine temps stabilize, I did another pull - being careful not to do any WOT in between runs.

Notice how timing in 4th gear actually started at 9.5 degrees and stayed there (just barely as it did not rise at all).

I wasn't expecting this.

According to this, the previous WOT knock event was not adjusted for in the timing curve.

Can anyone explain this?

[dzenno]

Very interesting...The way I'd explain it is that the first time around the DME reduced timing to prevent knock and see if it'd knock...it didn't..so at the end of that run the DME figured last timing curve went up to 8-9 no knock and on the next run it started on 9 and left it there ... Had it knocked on this second run I think on the third run, if you did one, it'd drop from 9 to 6-6.5 again to try knock free..

[Ilma]

Originally Posted by dzenno

Very interesting...The way I'd explain it is that the first time around the DME reduced timing to prevent knock and see if it'd knock...

But if it was going to "prevent" knock why would it start at 9.5 degrees?

Wouldn't be more conservative to start at 6.5 and rise upwards....testing for knock on the way up?

This also has the classic 3 degree drop that most knock events seem to register....so I don't know.