[Clap135]

Let me put my surprised face on........

[Jonny550]

Originally Posted by shiv@vishnu

The nonCAN channels log at 32/sec. The CAN channels are currently limited to 10/sec. And the rest of what Terry told you to write is rubbish.

lol kinda hostile.

[Jeff@TopGearSolutions]

Originally Posted by Clap135

Let me put my surprised face on........

I'll be over here with that as well...

[Clap135]

I find it funny how a Meth kit fails and people blame the tune. How bout we figure out why the kit failed and what kit it was. Enrita had a cm kit as well?

[adrian@vishnu]

The non-CAN channels are actually logged alot faster than that internally Shiv (like kHz), but the software only logs them a 32Hz. The Procede itself can react very quickly to events.

The key here is that cracked ringlands with reasonable design pistons (as these appear to be) are really only caused by detonation. Detonation can happen at any boost level, but will become more likely with higher boost levels... particularly as the DME still thinks it is running stock boost and is therefore targetting ignition timing to suit stock boost. Terry has long said that the stock knock control system has the capability to pull enough timing to control knock reactively (meaning it sees some knock and then retards timing). It appears that in some cases this does not hold, as the enigne clearly suffered severe knock.

The difference with the Procede is that it retards timing proactively. We can argue over the semantics of ignition timing limitting that is a term Terry introduced to confuse the populous, but at worst, the Procede is providing the DME with more authority range to retard timing reactively. At best the Procede is proactively reducing timing to avoid the knock occurring in the first place. Either way, the liklihood of knock is significantly reduced.

My guess on this situation is that the stock DME does have good knock control that has managed to provide reliability to many tunes that have no timing control... but in certain situations (sustained high loads to very high speeds), the conditions add up to increase the knock and therefore increase the ignition retard requirements for the knock control system. In stock cars, the ignition target is close to the level required, so 5 degrees of knock retard is enough to control knock in any situation. However in JB3 cars, most of that 5 degrees is used up to compensate for the extra 10psi of boost that requires less timing, so when you need another few degrees to control knock in "bad" conditions, you run out of authority range, so the knock continues.... and melted piston.

Adrian

[kennejp]

Was your meth working properly? And what map were you one? What were you doing at the time of the incident?

[Mike@N54Tuning.com]

Originally Posted by adrian@vishnu

The non-CAN channels are actually logged alot faster than that internally Shiv (like kHz), but the software only logs them a 32Hz. The Procede itself can react very quickly to events.

The key here is that cracked ringlands with reasonable design pistons (as these appear to be) are really only caused by detonation. Detonation can happen at any boost level, but will become more likely with higher boost levels... particularly as the DME still thinks it is running stock boost and is therefore targetting ignition timing to suit stock boost. Terry has long said that the stock knock control system has the capability to pull enough timing to control knock reactively (meaning it sees some knock and then retards timing). It appears that in some cases this does not hold, as the enigne clearly suffered severe knock.

The difference with the Procede is that it retards timing proactively. We can argue over the semantics of ignition timing limitting that is a term Terry introduced to confuse the populous, but at worst, the Procede is providing the DME with more authority range to retard timing reactively. At best the Procede is proactively reducing timing to avoid the knock occurring in the first place. Either way, the liklihood of knock is significantly reduced.

My guess on this situation is that the stock DME does have good knock control that has managed to provide reliability to many tunes that have no timing control... but in certain situations (sustained high loads to very high speeds), the conditions add up to increase the knock and therefore increase the ignition retard requirements for the knock control system. In stock cars, the ignition target is close to the level required, so 5 degrees of knock retard is enough to control knock in any situation. However in JB3 cars, most of that 5 degrees is used up to compensate for the extra 10psi of boost that requires less timing, so when you need another few degrees to control knock in "bad" conditions, you run out of authority range, so the knock continues.... and melted piston.

Adrian

There is no lower limit on dropping timing. It will happily go negative during WOT and I've posted logs before to show this. That is not relevent in this case. On meth both systems run effectively the same setup. The full factory advance when meth is flowing properly.
The question here is was the failsafe connected and working properly. If not then this is no surprise. Boost can not be increased until meth flow is within the window or you risk knock at these extreme boost levels. If so then there may be some other limitation/risk floating around at 19+psi.

Mike

[itsbrokeagain]

given your reputation in the Subaru and Mitsubishi camp Shiv, I would choose my words before attacking other tuners on here...

there are quite a few variables that can point to failure, but you seem really quick to jump right to a conclusion....without knowing all the facts or the full story.

[Ilma]

Originally Posted by Mike@N54Tuning.com

That is not relevent in this case. On meth both systems run effectively the same setup. The full factory advance when meth is flowing properly.

Mike

I was wondering about this very same thing.

There are enough logs out here from guys running meth that show timing between 10-14 degrees of advance.

Many of the Procede guys are running little to no ignition correction on these logs.

So then how is that any different from the JB3 setup which has no timing correction

[adrian@vishnu]

Originally Posted by Mike@N54Tuning.com

There is no lower limit on dropping timing. It will happily go negative during WOT and I've posted logs before to show this. That is not relevent in this case. On meth both systems run effectively the same setup. The full factory advance when meth is flowing properly.
The question here is was the failsafe connected and working properly. If not then this is no surprise. Boost can not be increased until meth flow is within the window or you risk knock at these extreme boost levels. If so then there may be some other limitation/risk floating around at 19+psi.

Mike

I very much doubt you are correct that this is no limit. The fact it goes negatie proves nothing, but please provide data, as I would be interested to view it.

Also to say the setup is the same on meth is not true. In ideal conditions they may run the same, but if conditions deteriate, the Autotune on the Procede will see this and reduce timing and boost. The JB3 would not even know that the conditions are deteriating, and will continue to run the same timing (because it cannot change it, and does not know to change it anyway).

It is fine for you to look at all the Procede logs for cars in good conditions and say the JB3 runs the same, but clearly the conditions were not great when the engine went. I suspect the difference in Swedish cars is that they run them at high speeds which requires sustained high load/RPM that is not common in the USA. THis is the worst possible conditions for build up combustion chamber heat which then results in detonation which the Procede would react to but the JB3 does not. These bad conditions would build up gradually over many seconds giving a tune plenty of time to react if it has its eyes open.

Adrian

[OpenFlash]

Originally Posted by Mike@N54Tuning.com

There is no lower limit on dropping timing. It will happily go negative during WOT and I've posted logs before to show this. That is not relevent in this case. On meth both systems run effectively the same setup. The full factory advance when meth is flowing properly.
The question here is was the failsafe connected and working properly. If not then this is no surprise. Boost can not be increased until meth flow is within the window or you risk knock at these extreme boost levels. If so then there may be some other limitation/risk floating around at 19+psi.

That's where you are mistaken. Yes, timing will go negative if long term knock adaption PUSHES IT DOWN over a period of sustained cumulative load. The problem here is that meth failure keeps the engine knock free when it is working as it should work. As such, knock adaption is PULLED UP (more advance). But if it fails or stops flowing for a second or two, timing will only be pulled out quickly in 3 degree increments. If 3 degrees isn't enough (and it isn't if you were running stock timing (10+ degrees) at 17psi, you will knock again until another 3 deg is taken out. And then again until another 3 degrees is taken out. On a datalog, this looks like it happens very quickly. But in a fraction of a second is an eternity for an engine spinning at 6000rpm (100 events in 1 second!). If just some of those events results in detonation, you will (at best) induce engine wear or (at worst) crack a piston land. Again, basic tuning fundamentals on how relying on the knock sensor for ALL TIMING CONTROL is negligent.

Shiv

[OpenFlash]

Originally Posted by Ilma

I was wondering about this very same thing.

There are enough logs out here from guys running meth that show timing between 10-14 degrees of advance.

Many of the Procede guys are running little to no ignition correction on these logs.

So then how is that any different from the JB3 setup which has no timing correction

The confusion comes from the fact the Procedes "ignition correction" datalog channel doesn't display the final actual output. When secondary timing corrections are induced, they act "downstream" of the ignition correction log value. Just a bit of IP protection built into v4.

[Ilma]

Originally Posted by adrian@vishnu

In ideal conditions they may run the same, but if conditions deteriate, the Autotune on the Procede will see this and reduce timing and boost.

Adrian

Thus the answer to my previous question.

You can run zero ignition correction on the Procede, but it will still add-in timing correction via autotune as it senses knock.

So in some sense, it is reactively being proactive when needed.

[Ilma]

Originally Posted by shiv@vishnu

The confusion comes from the fact the Procedes "ignition correction" datalog channel doesn't display the final actual output. When secondary timing corrections are induced, they act "downstream" of the ignition correction log value. Just a bit of IP protection built into v4.

So when we view our Can actual ignition advance logs, timing could sometimes actually be lower than indicated or does it just affect the amount of correction?

But I presume the knock dips are still relative?

[adrian@vishnu]

Originally Posted by shiv@vishnu

The confusion comes from the fact the Procedes "ignition correction" datalog channel doesn't display the final actual output. When secondary timing corrections are induced, they act "downstream" of the ignition correction log value. Just a bit of IP protection built into v4.

I thought we were keeping that to ourselves!!

Unfortunately our datalogging has served to make it easier to copy our algorithms, so we have changed our tact a bit with datalogging in recent times.

[OpenFlash]

Originally Posted by Ilma

So when we view our Can actual ignition advance logs, timing could sometimes actually be lower than indicated.

But I presume the knock dips are still relatively scaled?

The short term timing corrections that aren't represented in the logs are just that, short term (ie, less than 1 second) and only trigger under certain transient conditions. So without revealing too much, they wont trigger during the conditions you are probably thinking about (ie, looking for knock dips, or timing level during a WOT single gear pull)

Shiv

[Ilma]

Originally Posted by shiv@vishnu

The short term timing corrections that aren't represented in the logs are just that, short term (ie, less than 1 second) and only trigger under certain transient conditions. So without revealing too much, they wont trigger during the conditions you are probably thinking about (ie, looking for knock dips, or timing level during a WOT single gear pull)

Shiv

Good enough for me

Now back to our regular programming

[themyst]

Feel bad for the owner. This goes to show that when you push the limits, you gotta pay to play.

[techlogik]

Originally Posted by Matt@Camber-Toe

given your reputation in the Subaru and Mitsubishi camp Shiv, I would choose my words before attacking other tuners on here...

there are quite a few variables that can point to failure, but you seem really quick to jump right to a conclusion....without knowing all the facts or the full story.

This statement makes no sense. To anybody who understands what Shiv is talking about, there is a reason the Procede is the most sophisticated tune available out there. Everybody who has been around these forums for the last 3+ years knows the progression of Terry and what he has tried to do since day 1, steal/ripoff what Shiv has been doing and develop a product to make money. JB works good enough for most applications. But as you can read, when on the ragged edge and running meth/high boost, being good enough just doesn't cut it.

Shiv has so much technology that people don't even know what they are really doing, and the only way we find out is when they decide to leak something out like in this thread to prove a point.

And this is coming from someone who removed the Procede v4 and returned it for a refund and went with a GIAC flash. It wasn't because the Procede isn't the best, I just didn't want to deal with having to remove the piggyback in case of warranty/maintenance work. But when my warranty is near end, the Procede is going back in with more mods. Not JB/Terry and their poor implementations and lack of expeirence compared to competitors.

This whole Sub/Mits stuff with no backing and information except to incite and inflame things more is worthless.

Lastly, we still don't have all the facts about what the setup actually was at the time of the problem. But I think the conversations between Mike/Terry and Shiv posting in the threads regarding these issues is very important so people can judge and make their own decisions as to who knows WTF they are really talking about.

[themyst]

Weren't those guys that blew their motors in the Mitsu and Subie world pushing the limits past the tune? So in that case it wasn't Shiv's fault.

[OpenFlash]

Originally Posted by themyst

Weren't those guys that blew their motors in the Mitsu and Subie world pushing the limits past the tune? So in that case it wasn't Shiv's fault.

I know of 2 mitsu and 1 subaru engine failure. Both evo engines were stock with big turbos and running upwards to 25-30psi of boost on c16 race gas. Making close to 600hp out of 2.0-2.5 liters. So imagine a stock 3.0L n54 making 800-800hp for an idea of how hard they work. IIRC, the Subaru threw a rod at just 1psi over stock during deccel after a dyno pull so who knows what happened there.

And if a BMS reseller like Matt at Camber-Toe wants to know, I've personally tuned over 1000 Suburus and Evos over the last 10 years. And the race cars of Tanner Foust, Pat Richard, Stephan Verdier and a few other national rally champions. All over the world. For all types of competition. So if Matt thinks that citing isolated engine failure incidences gives himself some credibility and puts my tuning experience in question, then he can go ahead and knock himself out. But I'll say this, at just 150hp/liter we shouldn't even be working these n54 engines hard.

Shiv

[joyride]

You may have tuned over 1000 Suburus and Evos in the past, but you're still out of favor with that crowd for the most part. Which is surprising, since that would seem to be a bigger market, guess some things just can't be undone with basic engine tuning fundamentals and other marketing key words.

Regardless of the past, people do learn from things from time to time and move on.

Timing control seems to be the latest thing that everyone's keyed up on. Which it is a big part of the fundamentals, but when someone takes the stance of "we can control timing", i'd honestly expect more power then someone who "lets the stock DME" control it. But that doesn't seem to be the case, only that it's safer?

Eh... don't wanna take this too much off topic, so just make this a statement for another thread.