Gen5 LTx series

[ringram]

Ross any ideas on what model ECU it is?
I take it, its something totally new?

Or is it an E67 or E38?

Looks like a candidate for a nice cam. Stock spec is 200/207-0.550
Heads look nice.

[GMPX]

Howard Tanner posted elsewhere it is an all new ECM, never seen before. Not surprised given it's direct injection.
Sounds like we've still got a while before it hits the streets though. I think we all know this is going to be an awesome engine, I'll be at the Holden dealer putting my name down for one when they hit our shores.

[ScarabEpic22]

E92 is what the press release says, no idea WTF is actually is (Bosch, Delphi).

Ross is gonna have a heart attack if its Bosch though... :D (please dont though)

[JOHNBOY]

This was posted on another forum. Not sure I trust source. I think he tries hard to make himself seem important.

Any ways

Gm has dramatically changed the security acess, iirc 512bit keys and a completely different algorythm, think cray super computer to reverse engineer.

and the entire method for how the do fueling and ignition timming, its just so different. IIRC allot of the calibration is a mathmatical model.

not easily adjusted. _


512bit security algorythm, not happening. Plus they seed key pair will no longer be generated, they are stored in the GM mainframe. Or so I hear. Each ECU has its own seed and key, and its tied to each vin.

Its going to be all after market ECU's for these vehicles, or nothing.

[GMPX]

Ok, so whoever is posting all this stuff, how do they know 100% for sure, do they work at GM development, at Delphi? I call BS (but, willing to be proven wrong).

So, just on he's points:
512bit key, maybe, it's conceivable.
E39 & E78 is very much mathematical model based, so probably nothing new there.
Anyway, if GM have locked these down far enough that after market tuning (and therefore parts) can't happen, watch sales plummet in 2014.

[JOHNBOY]

He does not work for GM. He has "moles" giving him information. Claims to have insider info on japanese brands also.

I dont take him to serious. I have read his posts on other subjects such as machining and metallurgy. Which is my back round. He is out in left field to put it nicley.

More awesomeness.

I know, I've seen the calibration software its called "inca" and glanced at the code snipets. If you have a few hundred hours to work out the TQ modeling for the throttle input, you might be able to clean up fueling and spark.

Pay attention to what I just wrote there very carefully.

[the_red_shark]

Howard Tanner posted elsewhere it is an all new ECM, never seen before. Not surprised given it's direct injection.
Sounds like we've still got a while before it hits the streets though. I think we all know this is going to be an awesome engine, I'll be at the Holden dealer putting my name down for one when they hit our shores.

E39 runs the sidi v6 seems odd to throw a new ecu at it, but then gm did give us the e40

[GMPX]

E39 runs the sidi v6 seems odd to throw a new ecu at it, but then gm did give us the e40
I agree, the E39 is capable of running a V8 too. It probably comes down to supplier costs per ECM, in the end the hardware doesn't matter, it's the code inside running that makes or breaks the deal for tuning.

[JOHNBOY]

I am hoping it is E39 or close to it.
Looking forward to hopefully getting the chance to work on the new Gen5 LT1.

[DrkPhx]

It sounds like the ECM will be nearly hack proof, infinitely more complex with a secondary OS and well above where the current ECM's are at. Here is a snippet: The secondary OS within the self contained ECM will overwrite any changes if compromised. And it will NOT but just in case it will. Upon over writing any tuning/tampering... the HIGH PRESSURE SOLENOID will close allow statistic pressure to build to allow for starting and normal OEM spec performance.

Check out the posts from Big Gunz in this thread on tech. Hs states he is Bigg Gunz someone within GM Advanced Combustion Engineering. Could just be a troll with no credible info, but his posts are still interesting to read.

http://ls1tech.com/forums/lt1-lt4-modifications/1595424-new-lt1-2014-6-2l-alum-block-8.html

[ringram]

Well considering Greg Bansh told me a few years ago at PRI that the newer ECU models were torque based and not airflow based some of this makes sense. Except his solenoid weirdness.
In theory pedal position will become a commanded torque input rather than throttle blade or airflow related. So the ecu will "manage" torque delivery based on pedal position and remove control over all else.
So Im sure like the VVE modelling there will be some nice new tables. I doubt messing with them will be too challenging given time. Once someone has their head around the maths there will be ways and means to adjust.
The key of course with all of this is being able to tune it and gain access to the calibration.
It will be interesting to see how things go with it. Im sure Ross is waiting with bated breath for the first file dump or hard ECU.

You are right though. Kill off the aftermarket and watch sales plummet. There are better cars out there in stock form.

[DURAtotheMAX]

512bit keys? On a mainstream automotive ECM? Haha ill believe it when I see it.

[joecar]

By declaring the ECM to be secure and hackproof, GM have implicitly issued a challenge (i.e. they are expecting it to be hacked, they want to see how soon), lol.

[GMPX]

Cummins have had 32 bit unlock keys since 2006, but there is probably no point for GM to go to 512 bit for the unlock codes, if they did something like the Euro EDC17 with RSA involved, we are probably screwed.

[ringram]

So interesting then :)

Someone could put up an anonymous prize and place binary dumps on say a Ukrainian server offering cash for anyone who broke and decoded the binary...

ProjectX

[ls1mike]

Didn't they say the skyline ecu would never be broken into? Took people like 3 months.

Drop in ECUs from a new manufacturer would be awesome.

Have we hit our heyday, our high-water-mark? The LSX engine is why most of us are here so will we forever just work on the 99-12 cars/trucks/engines?

Will GM get its act together enough so that STOCK cars won't NEED to be tuned because they are so awful? Has anyone here driven a new car that fits that description? My 2010 Audi does.

[ScarabEpic22]

I agree, my brothers Jetta 2.5L and moms Audi Q5 3.2L dont need a tune. The only engines that get tunes are the boosted ones or if someone decides to boost a factory n/a engine.

GM needs to use WBO2s like VAG does, problem is I dont know if any current GM ECM can actually use that so they might have to either develop something new or use what VAG does...Bosch. *gasp*

[Redline Motorsports]

Well the PCM is definitely a DELPHI unit...saw the connectors. I think it will be a cross between an E67 and an E39 to run the DI system. GM know we are getting in and every time a new model comes out we freak. Granted they are getting tougher but they know better then to add stress to a billion dollar a year industry.

We may be incorporating a piggy back for a bit while this is sorted out but we plan on jumping all over this. For sure its SERIES 11 Global Architecture which is all TORQUE based logic....which I think with some good math skills can be solved.

One thing for sure is the hack tuners will be separated by the "calibrators" as this TORQUE based logic gets more wide spread. No more slush room for improperly dialing in the MAF or the air models as they will both reflect actual torque which is jammed back into everything....

Should be interesting for sure!

Howard

[DURAtotheMAX]

Well the PCM is definitely a DELPHI unit...saw the connectors. I think it will be a cross between an E67 and an E39 to run the DI system.

Do the connectors tell for sure though who makes it? Some delphi ECM's use Molex (MX123) connectors...and then the LML duramax ECM also uses MX123 connectors, and of course its completely Bosch even though it looks 100% delphi. The Molex MX123 connectors look almost identical to the Delphi micro-64 lever-lock connectors.

Ben

[Redline Motorsports]

It distinct-fully looks like a DELPHI

[GMPX]

We all thought the LML ECM was a Delphi too, just look at it!
I would be pretty confident it's a Delphi running the same or similar code to the E39 SIDI ECM as well, I can't see GM (these days) completely redesigning something to do the same job as something they already have.
It probably came down to cost or maybe Delphi didn't have an ECM ready to go with DI capability until now which is why GM has in the past used Bosch and the E39 (Continental / Siemens).

http://download.efilive.com/Staff/GMPX/E86_External.jpg

[redhardsupra]

512bit key for RSA is actually quite weak. Now if you're talking AES with 512bit key, then yea, we're proper screwed.
Straight out of Wikipedia: "RSA claims that 1024-bit keys are likely to become crackable some time between 2006 and 2010 and that 2048-bit keys are sufficient until 2030."
So just out of sheer curiosity, how much would it be worth it to you for someone to crack it?


Cummins have had 32 bit unlock keys since 2006, but there is probably no point for GM to go to 512 bit for the unlock codes, if they did something like the Euro EDC17 with RSA involved, we are probably screwed.

[ringram]

Marcin you binary whore :)

FWIW comments on ls1tech suggest as per the current ECU's there are multiple controllers and logic. Just as now you cant access the ETC tables and controller unless your name is Chuck. GM dont want people messing with the DI injector tables. They say they flow enough for well over 1000bhp in any case.

Also there is suggestion it runs a closed loop with EGT sensors to remove the need to access the calibration as much for most mods.

Of course things like gear changes, Torque management, max VSS, open loop, commanded afr etc are all things we will want to change. Plus this ECU is going to work across the entire range of vehicles. Trucks etc. So Its going to be target 1.

The dude, who claims he is a combustion engineer said that the DI logic was out of bounds for legal reasons hence the encryption. The implication was that the rest of the cal may be fairly open as per now. He also said fuel was injected at weird times which may seem to defy logic and that 87 and 93 octane gave about the same peak power levels.

Also that cam technology was where the power was at because the DI and combustion process was very touchy and they had spent a lot of time on this to optimise this aspect and he thought it was a close to perfect as they could make it,

I assume DI is done only fractionally before ignition and timing is probably retarded from what we are used to. Though there was suggestion fuel would be injected at other times to address EGT control.

Anyway, she looks like a nice motor. I think Ill shelve my Gen5 Viper plans and go high tech. Especially with the C7 planned to hit the scales around 3000lb with a 300 reduction. Though perhaps thats in Z06 form..

I cant help thinking that the AFM lifters are the weak point in an otherwise good setup. A nice VVT cam and DOD disable would be good. 550rwhp cam only!?

[GMPX]

So just out of sheer curiosity, how much would it be worth it to you for someone to crack it?
Not worth the multi million dollar law suit that would follow, it didn't work out too well for the last company who cracked the RSA keys on the EDC17's.