Darin's 80q…Alive...

[nsimps]

Glad to see you got things torn down so fast hope it's all back together and ready for next month's In N Out meet!

Have to commend you on your positive attitude through all this! Keep up the good work

[loxxrider]

Interesting Hank, I retract my previous comment. I still want to know the answer to my other question about Aaron's failure though.

[Hank]

Well detonation causes headgaskets to fail, so you often see these the cut out heads associated with detonation events. They are not exclusive however. I have seen just as many bone stock cars have failures like this. Usually they are helped into failure from loss of water that warps the head, or sustained throttle going up a mountain pass. If the water gets past the headgasket into the combustion chamber at low coolant pressures(in a closed system), you can bet it will find the void in the headgasket on boost when cylinder pressures get into the 2000-5500psi cylinder pressure range.

When a piston is going down in the cylinder, it is still filling up with air. If you are cramming a low relative pressure of 24psi of boost, you still are around the same pressure of the coolant system. The coolant system pressure can be more than the cylinder pressures at certain segments of the stroke, plus the quick vac situations between boost.

Vems does not have barometric compensation without adding it in, but the closed loop control will still get you to your target if the VE table is close. The Lambda Target will be met when the ECU gets data from the WB that the mixture is rich. I wish the Vems just did a rough barometric compensation table like other ECUs, but htat would require a second 1 bar map sensor on board. I understand it is coming to light, along with my biggest request of better dwell control on ignition.

[Dave]

i had asked marc about vems and if it uses the barometric pressure sensor in our cars, he said it doesn't but it does take a reading at key on that accounts for it.

[loxxrider]

Yep, VEMS does account for it somewhat at least, like Dave said.

Hank, my question about Aaron's problem is why quick transient boost increase would cause head lift or a failure like this (for whatever reason). I just don't understand why that is any different than getting to high boost under "normal" conditions.

[savagerocco]

Megasquirt has real time baro correction. My ecu has 2 4 bar pressure transducers on for map signal the other for baro. They have to be the same transducers in order for the program to scale for altitude

Sent from my SAMSUNG-SGH-I717 using Tapatalk 2

[Hank]

Like Dave said, the Vems takes the atmosphere on startup. After that, the WB02 is correcting in closed loop to compensate for atmosphere difference. They would not have to be the same transducers Brent. Just a simple reading fed into an input that would modify the VE scale in programing.

[savagerocco]

Megasquirt can do that as well but you don't always have a good reference on Start Up. It is not the same as having two sensors. There are situations where you will get erroneous readings and say you drove from your house to the top of the Mountain. Which is only 20 miles from daring house?
Sent from my SAMSUNG-SGH-I717 using Tapatalk 2

[Hank]

Yes, so say at Darin's house at 4000ft you see 87kpa worth of atmosphere, and then drive up to Park City at 8500ft without stopping(74kpa). Vems would read your base VE fuel table as 87kpa on start up. As you begin to climb, you would lose atmosphere which would cause you to go rich if the EGO was turned off. Fortunately, going from 87 to 74 kpa worth of atmosphere only represents a small percentage 15% reduction in total atmosphere. If in closed loop, the ECU will correct the rich mixture by taking out 15% of fuel and leave you with the same mixture. Sure, it would be awesome if the EGO didn't have to correct for this, but it will happily do it. I suppose the extreme elevation(say sea level up to the top of the Unitas) would be above the correctable allowance, but that is a very rare occurance where you don't stop for gas, pee breaks, ect.

You guys are in a unique geography there in Utah. Most people do not need the barometric sensor. For those that do, it is SUPER easy to add one to the Vems. Infact, Marc does it routinely with the stock secondary senor on the urS.

[turboskipper]

A speed density algo does not need baro. Only alpha-n needs to rely on a baro compensation and on a turbo car baro won't work either, you need prethrottle pressure (post compressor). I only find baro useful for target boost pressure. A turbo that can make X psi at 2000ft can't always make that at 7000ft so sometimes it is nice to make a target manifold pressure that is X kPa above atmosphere.

[Hank]

Good points Preston

[scubadave]

For those that have all the stock sensors including the stock baro sensor, and are using the Audi PnP Vems box, is it being used/ utilized?

[savagerocco]

For those that have all the stock sensors including the stock baro sensor, and are using the Audi PnP Vems box, is it being used/ utilized?

Mark said earlier in this thread, no the stock. Baro is not used

Sent from my SAMSUNG-SGH-I717 using Tapatalk 2

[savagerocco]

A speed density algo does not need baro. Only alpha-n needs to rely on a baro compensation and on a turbo car baro won't work either, you need prethrottle pressure (post compressor). I only find baro useful for target boost pressure. A turbo that can make X psi at 2000ft can't always make that at 7000ft so sometimes it is nice to make a target manifold pressure that is X kPa above atmosphere.

As to speed density not needing baro I disagree. I know that when my baro readings are off the main ve is affected and the car runs differently. If you were to take a look at the code and specifically the formal used to determine pulse width for a certain map reading you will see that baro will change the outcome.

Sent from my SAMSUNG-SGH-I717 using Tapatalk 2

[Marc]

jumping in late here but correct, at the moment the PnP boxes do not use the baro sensor. It is irrelevant to this discussion though, this is not 'fuel related' failure looking at the damage to the head. One of two things occured: either the coolant entered the cylinder and waterjet the path of least resistance due to a gasket failure or poor torque on the studs, or the EGTs got hot enough to soften the aluminum and blow the coolant through it. If it were the second case due to lean burn, you'd see it in the plugs. Prolonged high EGTs should also show up elsewhere in the engine, such as melted piston crowns and damaged plug electrodes, none of which are evident.

As hank pointed out, the directionality of the 'blow out' indicates the water jacket was blown out under pressure, so my best analysis is a head gasket / stud torque failure compounding into a waterjet cutter.

[savagerocco]

jumping in late here but correct, at the moment the PnP boxes do not use the baro sensor. It is irrelevant to this discussion though, this is not 'fuel related' failure looking at the damage to the head. One of two things occured: either the coolant entered the cylinder and waterjet the path of least resistance due to a gasket failure or poor torque on the studs, or the EGTs got hot enough to soften the aluminum and blow the coolant through it. If it were the second case due to lean burn, you'd see it in the plugs. Prolonged high EGTs should also show up elsewhere in the engine, such as melted piston crowns and damaged plug electrodes, none of which are evident.

As hank pointed out, the directionality of the 'blow out' indicates the water jacket was blown out under pressure, so my best analysis is a head gasket / stud torque failure compounding into a waterjet cutter.

I don't think anybody is saying this was a fuel related failure. I looked at the block personally when I pick Darin's head up to repair it. The bottom end and plugs looked great. I would say Darin needs to invest in a better torque wrench, the cost of this repair alone would have paid for a very nice snap-on unit.... but the main thrust of this turn into baro correction in general to help improve tune. I would assume that vems looks at baro correction values in the main fuel calculations and an accurate baro reading would improve the tune getting the ve closer to correct with out having to trim it with closed loop lambda correction. Some systems are set to not make corrections above a certain KPA setpoint. I guess that depends on how you have set things up.
Not everyone would benefit from a realtime baro, but here is Utah, where we can go from 4000 feet to 8000 feet and back to 4000ft in a single drive, with out stopping, this becomes a good thing to have on a SEM.

[audifreakjim]

I think Preston said it best though, you really don't need baro compensation for MAP based systems because the MAP sensor is already doing just that. It is reading pressure in the manifold, which could be at sea level, or 10,000 feet and the MAP sensor would still read the correct pressure and put you in the correct portion of the fuel table. I have played with the baro compensation in real mountains going from 5k to 14k elevation and if you have it enabled it really screws with things. I have the table in my IIc with the minimum amount off adjustment in the cells and even from sea level to Vegas at about 2000 feet if was messing with my tune.

[Hank]

Brent, why would you need baro on a speed density system? Fuel is being matched to a barometric setting, so 100kpa is 100kpa. If you are going to the top of Park City at 8000ft, you will see 75kpa and you will add proportionally 75kpa worth of fuel to get to the right AFR. Going Forced induction at Park City @ 100kpa would be requiring the same amount of fuel as a sea leveler at a static 1bar, even though you are 25kpa above ambient pressure. Pressure is Pressure, and when you are tuning for density, it doesn't matter what the relative atmosphere is, it is matching fuel. As Preston said, the only issue you have is when you are getting outside of hte turbo's efficiency, and this is mostly for boost control PID settings.

To add to Marc's comments, if it was indeed EGT, the garrett GT turbine would be gone. These turbos spool nicely because of the lack of mass on the turbine, but the flip side ist hat they do not handle high EGT well.

[savagerocco]

I have noticed a difference with baro connected and not connected. I would think that you would hit different cells in your map based on atmospheric pressure that you wouldn't normally hit. If you have higher atmospheric pressure, you might hit boost earlier and thus a different cell than you would normally be in and vice versa, if at higher altitude, you would be in a different cell. Thus, unless you have been in that cell before, a barometric correction would be beneficial. Most SEM's come with provisions for baro correction and manufactures use it as well, so that tells me there is some benefit to its use. I wont tell you that it wont work without it, but obviously there is a reason that it is used. I like having it on my car and believe that it allows my VE maps to be closer to correct without need for an after that fact adjustment.

[turboskipper]

I have noticed a difference with baro connected and not connected. I would think that you would hit different cells in your map based on atmospheric pressure that you wouldn't normally hit. If you have higher atmospheric pressure, you might hit boost earlier and thus a different cell than you would normally be in and vice versa, if at higher altitude, you would be in a different cell. Thus, unless you have been in that cell before, a barometric correction would be beneficial. Most SEM's come with provisions for baro correction and manufactures use it as well, so that tells me there is some benefit to its use. I wont tell you that it wont work without it, but obviously there is a reason that it is used. I like having it on my car and believe that it allows my VE maps to be closer to correct without need for an after that fact adjustment.

Some ECU's have a baro correction but the physical model that speed density represents does not have a baro correction. It is simply the ideal gas law. At different baro's you are right, you might hit other spots in the VE map but they are still just mapped to a manifold pressure. You might also push the turbo harder at high altitude which adds heat. This factors into the ideal gas law and thus a speed density algo needs to be corrected by the charge air temp. I write the software for bosch motorsport controllers. I have a pretty good understanding of the bosch controller and others out there. So trust me that if you map things correct baro correction is not needed.

[3BI5]

In non detonation events( stock ECU's, stock turbos, stock boost levels and stock components), heads get cut out like this because water gets into the combustion chamber via a weak headgasket, a warped head due to excessive heat, or debree on the surface of the deck. That water is sucked in on off stroke and expelled out at high cylinder pressures on the compression stroke. This is vaporized water with mass at this point carrying a tremendous amount of energy, along with the 1600 degree "burn" you reference. 1600 degree water plus a modest 2000psi at 24psi that Darin is running makes for a very powerful jet stream. If given given the option of least resistance into the coolant system, the pressure will cut out the soft alloy in the head, and the .010" layers of steel to get there. 1600 degree flame mixed with vapor water will make it look like a burn, but it is just a waterjet part! It is erosion. No aluminum was melted, contrary to your theroy, but eroded away. Look on the back of both failures. The backboard of the waterjet eroded into back of the failure.

EGT's may be in the 1600 degrees but actual pump gas burning temps are somewhere in the range of maybe 3600-4000F. So it's way above aluminium melting temps, even without problems.

[Hank]

For no duration long enough to be relevant to this discussion. Like you said, it is not where near sustained at those temperatures or you would not have aluminum(or blocks/valves/rings/turbines/ect) components survive. Aluminum simply can't take in the heat fast enough to be affected by that temperature. The temperature is partial to the gas burning, but is mostly the effect of compressing gas 3 bar worth of gas 9.5 times.

[3BI5]

Yes, that's how correctly working combustion engines survive, the heat cycle is short enough to not to melt the engine. But if you have too much ignition advance or preignition and then you're compressing already burning mixture and the heat transfer skyrockets and that will melt any headgasget or what ever the weakest part of the engine is.
And there is no need for audible knock to this to happen. One compression cycle is enough if the preignition happens early enough, because of spark hop or some other reason.

[death 4kqt]

this is a good discussion. I have nothing to add, but it is a good read.

[nsimps]

:stupid: