David@Vishnu

iTrader: (2)

The XEDE is very dissimilar to the UTEC. From the start, it was designed to work with the factory fuel and ignition control, not in place of it. I see no real similarities other than they are both plug-n-play.

just my 2c,
shiv

Imyurturboluva

iTrader: (4)

Let's not forget about the knock detection too with the XEDE, Shiv! It allows the OE ECU to detect and compensate for knock. IMHO, I trust two things...1st, my ears with det cans and 2nd, the OE ECU. That's just my $.02.

-chris

SoR

iTrader: (19)

Utec also works with factory fuel and ignition controls and you can only offset them. Only with new (not-out-yet) firmware will you be able to choose to take full control. Am I missing something?

SoR

iTrader: (19)

And how does Utec detect knock?
Didn't you say you're going with AEM?

rebelzx

iTrader: (4)

Why not get a cheap flash to raise the rev limit until the update comes out?

BigBoogieman

That's true for fuel, but the UTEC doesn't allow you to offset ignition timing, you have to specify absolute values. I really wish it had an option for specifying offsets for timing, but I think I'm in the minority here The problem I have with absolute values is that the ECU pulls timing as boost starts to build, and with a simple 2D map like the UTEC has, there's no way to do something like that. You can come close to getting things smooth in all occasions, but I don't think it's really possible to get timing to work as well as the ECU does in all situations.

But other than absolute vs. offset timing, the UTEC and XEDE are VERY similar...

BigBoogieman

The UTEC monitors the stock knock sensor, and pulls timing when it detects knock. You can even configure how much timing it pulls for how long, and the sensitivity in different RPM ranges. It works very well, and you can see in the logs when the ECU wants to pull timing, and whether the UTEC pulled timing in that spot. In my car, the default UTEC settings are almost spot-on with what the ECU is seeing.

Imyurturboluva

iTrader: (4)

Nope, never said that. Standalones are great for racing, but as far as a street driven vehicle is concerned, I'd rather use offsets of the factory ECU. It's much safer that way.

As far as the UTEC detecting knock and utilizing the factory sensor, that's great. However, I use det cans (with headphones) to listen to knock and the UTEC did not pick it up, somtimes, when I heard it. Also, it was detecting when there was no knock. It created a lot of suspension with me from the start. Say whatever you want...I know what I hear and don't. Maybe some of you are having better luck than I did. Flame away!

shiv@vishnu

iTrader: (20)

How have you confirmed that you are picking up every instance of knock and that all (or even most) of the knock signal you are picking up is indeed knock? And how can you be certain that the actions you take (with retard to rate of retard per engine event and post-knock advance is appropriate)?

Shiv

MalibuJack

iTrader: (5)

Well, I really don't want to say too much since this is a bit of a trick question and nobody can really answer this appropriately.

However, my experience so far with the UTEC has been that when monitoring the stock ECU, and seeing (And hearing) how it responds to knock events (and non-knock events, since I found one or two of those) was almost identical to how I saw the UTEC respond. The rate at which the timing comes back (Post knock advance) is different though (it doesnt just suddenly come back though).

On my car, the Stock Engine doesnt seem to produce the same "Racket" that alot of other people seem to have trouble with. Its also not as detonation prone as some other cars that I've heard of. So what is good for me, may not be good for others. In my case what I see the UTEC do, and its sensitivity and retard levels in its knock algorithm, appeared to closely mirror the factory ECU.

There's just no right answer for this.. Because if you allow the factory ECU to do its job, there will be occasions when your fighting with it to get your tune, or tuning on top of non-optimal conditions. If you take complete control over this (which the UTEC does) you have to be comfortable that the configuration of the knock control, And even more importantly, your tuning is going to be appropriate. I'm comfortable with that since I have seen that it works. The other side of the fence is if your offsetting timing from the stock ECU's desired timing.. Well, then your playing with a moving target since you don't have control over the circumstances where it may adjust timing and why. Plus there is a limit to how much adjustment you can make (realistically it makes sense, and is not an important point). But finally, adjusting knock in that environment may require attenuating or desensitising the stock knock sensor so it doesnt pick up every event anymore...

You really can't argue which is better.. Their just... Different..

I do agree on one thing, the stock ECU does have a very complex compensation system for handling less than optimal situations, however if your car is experiencing less than optimal situations, I would think you would be trying to find the problem before it can do damage then just tuning around it or tuning the activity out...

shiv@vishnu

iTrader: (20)

FWIW, running an engine at 20+psi of boost on pump gas is, without a doubt, a "less than optimal situation." A point proven by being able to gain 30-40bhp and greater run-to-run consistency and lower head temps with just the addition of race gas and appropriate remapping.

Unlike aftermarket stand-alone fuel/ignition systems, the factory knock control system is both reactive and proactive. There are two ignition maps (and 2 fuel maps) that it actively "floats" between depending on knock conditions. It is constantly predicting where the knock threshold is and keeping its active timing advance within check. It doesn't simply run a fixed ignition map and pull back timing when knock is detected. This old-school approach is rarely, if ever, done in modern engine control systems when turbo cars are designed to safely run at an always-moving knock threshold for max power with hopes of adaptability.

Just my 2c,
shiv

MalibuJack

iTrader: (5)

I agree with your points for the most part.. The one thing is that the prediction is based on a long term and short term knock value, and may take several run cycles for it to drift back into the primary map.

With 91 octane gas, I can see why you would say 20+ psi is a less than optimum situation. However I'd much rather have the control over how it handles knock and how quick its recovery is.

I do agree that in some circumstances, allowing the ECU to continue to control how it handles timing and knock can be an advantage.

There's no correct answer, just preference..

If I wanted the stock ECU to continue to do all the work, and just intercepted/latered the fuel/timing, I'd probably use the ECU+ or the Emanage. For that matter, I would have just stuck with the S-AFC since it would be pointless to even try to control the timing if it were drifting on a run to run basis. The other big problem with remapping/intercept devices is that they will never be able to bypass the MAF fuel cut without clamping the MAF signal (and potentially running dangerously lean) or using larger fuel injectors (which also works with the S-AFC).

The advantage of the UTEC is that its new firmware takes control of the coil drivers and the injector drivers, so although its altering and clamping the values to the ECU, its not in any way limited by it because its in direct control. To some they may see that as a disadvantage, to others its an advantage.. For my needs its what I think is the better choice. My original choices were always between a full standalone, or the UTEC, I chose the UTEC because I have the option to allow the stock ECU to continue to run in a stock mode if for some reason I don't want or need to adjust the car's tuning.. Or even portions of the tuning.. In my opinion its the best of both worlds.

shiv@vishnu

iTrader: (20)

Not exactly true. It is possible to drift back to the primary map in just 1 run. The adaptive features are very fast-acting. For example, after an ECU reset, the car defaults to the high det maps (very rich with less ign. advance). Within 4-5 seconds of being on boost during the first dyno run, it will (if knock is nonexistant) slide on over to the low det maps. In fact, you can even see the transition during that one dyno pull when AFRs suddenly bump up a point leaner and torque jumps 15-20ft-lbs.

This adaptive fuel/spark strategies were even obvious during our One Lap adventures during the 6000 miles of transits where we were towing 800lbs of supplies, while running 20-22psi of boost on a GT30R turbo, on 91-94 octane in 100+ degree temps through traffic, mountain ranges, texan plains, etc,. After periods of long sustained boost (4000rpm, 15psi, up a hill for 5 minutes), you could watch ignition advance (through a scan tool) float around in a 3 degree range from, say 3 deg BTDC to 6 deg BTDC. I can only assume fueling was floating around with it. The neat thing about this behavior is that it is predictive as well as reactive to knock. In other words, knock retard activity isn't obvious as it would be with an aftermarket computer with reactive knock control where you would suddenly feel the timing being yanked out, torque falling on its face, recovering slowly, going back to normal, then again timing being yanked out, etc,...

It's also interesting to note that even on high octane race gas running 25+psi of boost, this adaptive behavior can be found even without the presence of audible detonation (although audible through det cans if you listened carefully). I still stand by my belief that these features kept our engine in one piece during that week of hell and uncertainty Trying to get a stand-alone tuned to that level of adaptability and safety margin while not giving up big power requires a level of tuning that most people (including myself) aren't capable of without a *lot* of testing time and full access to load bearing dynos and lots of data acquisition, IMHO.

But I don't think you know how many knock control related parameter exists in a factory computer. They are not just dictating the rate of retard, rate of advance, thesholds and max retard. They employ algorithms which monitor historic knock activity. They dictate load and rpm points where knock control is inhibitied or activated. They take into account the sonic signature of the block and how noise travels in it. They calculate the likelihood of knock by monitoring knock noise in conjunction with RPM and engine load. And so on... In the WRX, for instance, there are over a dozen knock control-related parameters not counting the 3D knock correction table.

I agree.


It's always important to appropriately control timing even with adaptive systems. For one, the amount of timing offset that is needed is often larger than the window of drift. Especially when you are inducing MAF signal offsets which can induce timing shifts by itself. So to keep the stock ECU's adaptive features effective, the "raw" underlying ignition advance output needs to be reaonable for the given conditions. This allows the adaptive features to simply ride on top of a well mapped "foundation" instead of constantly intervening to keep the motor from failing (although I have seen some applications which unfortunately employ the latter strategy and get away with it ).

Also, there is no reason to ever use a MAF clamp with the XEDE since all boost limit functions (as well as rev limit changes and/or injector scaling) is handled by the XEDEFlash.

Just my 2c,
shiv

Imyurturboluva

iTrader: (4)

Take the challenge, Jack. Try the XEDE for four weeks and see which you prefer. I can, with almost 100% certainty, predict which system you will want to keep. Walk toward the light, Jack. Trust me, I've had both the UTEC and now the XEDE.
-chris

BigBoogieman

Because you can see when the ECU wants to pull timing in the data logs, and therefore when it has detected knock. Even when you're running your own timing, you still have a column in the logger that shows what timing the ECU would run there. Of course I don't REALLY know if knock is being detected in all cases and that there are no false detections, but what I'm saying is that the UTEC's knock detection seems to work as well as the ECU's, and the sensitivity is adjustable if you need it. And I would say the rate of retard is appropriate because the knock event count is always low. The UTEC gives you quite a bit of control over all this, but I've found the defaults work very well.