Cylinder wall mark, what is it?

[Primer_Drift]

Long story sorry:
I'm in the middle of pulling my motor, for some teardown and rebuild, among other things. I decided to pull of my head and take a look to see how things were doing.
The motor has about 3000kms on it, and was running really well. I had a few problems at first tuning AEM at first due to faulty sensors, 1000cc injectors etc.. needless to say I ran RICH for what I consider a crucial period of 100kms after first startup. Break in was done under very low engine loads up to 1000kms (2.5psi) , and low rpms (max 4800rpm) until about 2000kms. Compression at this point was 155 accross the board, which I was told is normal for low comp pistons.
So I decided I would go up to Budpark for the last race of the 2004 season to log some pulls and do a little tuning. Car ran great for the first 2 runs, after the third it started to burn oil. Not good, fearing the worst I limped back to Calgary, stopping several times to check oil levels, which were fine. I won't know for sure the cause of the blue, until I check everything out.

Now I've never seen an engine with 3000kms on it so I don't know what this should look like, but these marks look kinda funny to me. They are only on the intake side of the block only, all 4 cylinders, with #1 having the "worst" looking mark in the middle of cylinder. Anyone know what it could be from?

http://members.shaw.ca/nibbs333/delsol/Pict0233small.jpg

[rage2]

Ouch... scored cylinders. At 3000km, wear like that could only mean a problem with the buildup. You'll need to overbore those sleeves and put in bigger pistons now unfortunately. Before you do, obviously try to find out what happened, what caused it, so it doesn't happen again.

Even if you ran really rich for a while, I can't see washing oil off the rings to cause that much wear. Over time, yes, but 3000km, no way. It's gotta be a problem with the buildup.

[Primer_Drift]

:( Yeah I was afraid I might have washed out the rings with that initial rich condition. As for overbore.. I dunno bout that, since they already are 50 over.(87.5mm). I'll let my builders look at it, they'll let me know what the real damage ($$$) is. Thanks for the reply rage.

[redline]

Originally posted by Primer_Drift
:( Yeah I was afraid I might have washed out the rings with that initial rich condition. As for overbore.. I dunno bout that, since they already are 50 over.(87.5mm). I'll let my builders look at it, they'll let me know what the real damage ($$$) is. Thanks for the reply rage.

is this in lude engine? who did the machine work?

[legendboy]

I know your saying it was rich, but if it did that all of a sudden as you were making a pass at the drag strip it really looks like the car ran lean! Time to find out why.

[Primer_Drift]

Originally posted by redline


is this in lude engine? who did the machine work?

Yeah, h22a4. Precise engine rebuilders did the final build, (bottom end assembly).


Originally posted by legendboy
I know your saying it was rich, but if it did that all of a sudden as you were making a pass at the drag strip it really looks like the car ran lean! Time to find out why.

Well the car was running rich under boost conditions, but that was by design, around 12.8:1 and stoich or slightly higher at cruise. I have the logs of the 2nd and 3rd run and I don't show any lean spots, knock values are a little higher in the final run, possibly due to the change in IAT (20* hotter) or oil consumption. I'm thinking the oil rings were already clogged with buildup, hence the scorching, but were not detrimental enough until I started to push it. At any rate I am going to redo the ignition system and have my injectors flowbenched to ensure they are all flowing properly, that I am not leaning out one cylinder and running rich in the other 3 to compensate, whatever the case may be.

[legendboy]

what type of engine management are you using?

[Primer_Drift]

AEM EMS, AEM UEGO wideband gauge, Autometer EGT. I'm using the boost comp. volumetric efficiency (with throttle inj. modifer) method. If I had used the boost comp method from day 1, I would probably not be pulling the motor, instead at first I used a generic turbo map, which I scaled for my injectors.
AEM made me do this at first -> :confused:
Even with reading the forum and the manual twice it was difficult to get going.

[rage2]

Originally posted by Primer_Drift
:( Yeah I was afraid I might have washed out the rings with that initial rich condition.
Washing out the rings does NOT give u that much cylinder wall wear. At 3000km, if u were running as rich as the motor can handle without stalling, like worse case scenario rich, you'll get light scuff marks on the cylinder walls. It's definately something else.

[Primer_Drift]

Rings were file fit, although I don't doubt Peter's (from precise) ability to build these motors, is it possible ring clearance was to tight for the application? If not what else would cause it?

[legendboy]

email me your CAL and the log if you have it and i'll have a look at it for you. [email protected]

[legendboy]

the only other thing that would cause that becides running lean is if the cylinder was out of round but it would have worn from the very first time you started it up.

[heavyD]

Looks like you have detonated on that piston pretty hard and your burning oil because you smoked the ring or the ring land dropped on that piston which is why your scraping the wall hard like that.

[Alpine Autowerks]

pop those pistons out to get the rest of the story, bet you have broken rings or lands.

[Dr. Lightspeed]

Hard to tell by the cosshatch you can still see looks shallow in pattern which could have led to lack of oil retention during break in. Also excessively rich also could have led to washing the rings out. Without looking at the piston I think the ring gap was too tight and you butted them won't know till the piston comes out.

[Primer_Drift]

I'll keep posting updates as they come, thanks for all your input guys.

[rage2]

Originally posted by Dr. Lightspeed
Also excessively rich also could have led to washing the rings out.
Not with that much damage in 3000km. The damage from running overly rich is negligible.

[legendboy]

Ok i had a look at your datalog and your cal. Your calibration looks pretty fucked. Your trying to boostcomp your car but your boost correct fuel table is setup wrong. It looks like you tried to make the slope of the graph go thru the vertex, but thats not the case at our elevation. So problem #1.

Second, this car was never setup with boost comp properly in the first place. Your "VE" graph doesn't resemble anything close to the ve of any motor. Its basically set flat across all RPM and load points.

This car needs to be on a dyno, have 1 load column tuned to a specific afr for all rpm breakpoints. THEN you copy that to the rest of the map to obtain your "VE" Problem #2.


It looks like your were able to tune the car to around avg of mid 12's under boost but your accel fueling looks like it needs alot of work, your hitting mid to high 13 afr's every time you are building boost. So your running super lean for 0.5 to 0.75 seconds, boost between 4 to 6psi, when you stomp on it. Problem #3 (this one is veddy bad! http://www.telusplanet.net/public/legend88/Smiles/nono.gif)

You did no fine tuning of your ignition values at all! They are directly off the aem basemap! Even the advanced ignition trims haven't been touched.


Finally, i'm not sure which wb02 you have but your o2#1 gain is set at .95 which tells me there is probably some difference from what YOUR o2 sensor (and logs) say to what your actuall afr's were. If you could tell me how exactly you calibrated your wb and what brand you have.

Also, you have internal logging turned on, it be nice if you pulled that off as soon at the motor let go after that run at bud park.

[Primer_Drift]

Originally posted by legendboy
Ok i had a look at your datalog and your cal. Your calibration looks pretty fucked. Your trying to boostcomp your car but your boost correct fuel table is setup wrong. It looks like you tried to make the slope of the graph go thru the vertex, but thats not the case at our elevation. So problem #1.

Second, this car was never setup with boost comp properly in the first place. Your "VE" graph doesn't resemble anything close to the ve of any motor. Its basically set flat across all RPM and load points.

This car needs to be on a dyno, have 1 load column tuned to a specific afr for all rpm breakpoints. THEN you copy that to the rest of the map to obtain your "VE" Problem #2.


It looks like your were able to tune the car to around avg of mid 12's under boost but your accel fueling looks like it needs alot of work, your hitting mid to high 13 afr's every time you are building boost. So your running super lean for 0.5 to 0.75 seconds, boost between 4 to 6psi, when you stomp on it. Problem #3 (this one is veddy bad! http://www.telusplanet.net/public/legend88/Smiles/nono.gif)

You did no fine tuning of your ignition values at all! They are directly off the aem basemap! Even the advanced ignition trims haven't been touched.


Finally, i'm not sure which wb02 you have but your o2#1 gain is set at .95 which tells me there is probably some difference from what YOUR o2 sensor (and logs) say to what your actuall afr's were. If you could tell me how exactly you calibrated your wb and what brand you have.

Also, you have internal logging turned on, it be nice if you pulled that off as soon at the motor let go after that run at bud park.

Thanks for looking at the cal and log, I calculated the boost comp table based on the following equation (14.7psi=1atm absolute) 1atm absolute + 1atm of boost= 200% fuel from 0psi relative. All values were based upon that premise. Eg. -13.9psi is my first value in my boost comp table. -13.90/14.7=-0.9456 or -94.56% fuel. All boost comp table values are considered to be linear so why should it not pass near the vertex? If at our elevation my map sensor reads -1.8psi load with engine off, and this value doesnt change, how would this affect the linear scaling of the boost comp table?? When I first tried inputing max and min values for the table -94.56%(@-13.90psi) and +74.41%(@10.94psi) and hitting the auto calculate function, the values it gave were incorrect, in my mind at least, so I calculated each one manually.

I know the VE map looks flat, I was thinking that too I believe it has to do with the injector sizing. Due to the size of the injectors (1000cc) the injector duty was very low. I'm not even sure the fuel line pressure to the injectors, but I suspect it to be higher than stock due to the walboro HO pump. With smaller injectors for example, half the size, every change in VE would be doubled as injector pulse width would need to be doubled to match the output of the larger injector.

From looking at that log I saw it leaning after the shift while boost was building too, I believe that has a lot to do with the accel/decel values as you were mentioning.I could be wrong but it go leaner than 13.5 when I shifted with the DFCO. I know for a fact now that the default DFCO values are to extreme and it cuts fuel far too quickly under boost.

As for ignition values, that I admit is where I need a lot of work/help. Yes they are of the turbo 440 basemap, but this was after testing some other values, and watching knock. The stock map seemed to work just fine, and the motor felt smooth. EGTs with that ign map at cruise were 700-800*F and under boost 1000-1100 max. Which as long as there was no significant knock, was acceptable to me.

The AEM uego is what I have, Wb o2 gain was setup correctly, with sensor cable disconnected, as instructed by AEM.

Edit:
One other thing to add to the VE map looking flat, my rpm breakpoints are non equal distribution, this adds to the impression it is flat. The original map had a max breakpoint of 10000rpm.. and I had no intention of revving that high, so I added resolution in an area that needed it. Secondly my Microsec/bit was lowered to increase resolution.

[legendboy]

Its been a while.... but when calculating your boost correct table you take the min & max readings at full vac and wot and interpolate between thoes. Your linear graph will not pass thru the vertex... not here. (i'd have to go back and look to be sure of this)

You should definatly change your microbit/sec value if thats causing your map to look like that. I had rochester 900cc injectors and I had no problems with resoloution like you do. The ve map should resemble the smooth shape of your engine ve, if it doesn't boost comp won't work correctly.

Your accel fueling problems were suficient enough to cause detonation and cracked ringlands, but i suspect your engine mishap was a couple different things working together.

If you are not having any ground loop issues at all with your aem uego you should be using a gain of 1, meaning the value off the orange wire with the sensor unpluged will exactly match your aem pro parameters o2#1 raw.

I tuned my car againsed a motec, fjo and a techedge wideband and i was consistantly leaner by around .6 - .8 afr! If your gain is other than 1 it could be off by more!

Aem uego :thumbsdow

If you look up my username on the ems forum i had mega problems with my uego but finally figured it out completely. Something to look at.

Always dyno tune!

[Primer_Drift]

Originally posted by legendboy
Its been a while.... but when calculating your boost correct table you take the min & max readings at full vac and wot and interpolate between thoes. Your linear graph will not pass thru the vertex... not here. (i'd have to go back and look to be sure of this)

You should definatly change your microbit/sec value if thats causing your map to look like that. I had rochester 900cc injectors and I had no problems with resoloution like you do. The ve map should resemble the smooth shape of your engine ve, if it doesn't boost comp won't work correctly.

Your accel fueling problems were suficient enough to cause detonation and cracked ringlands, but i suspect your engine mishap was a couple different things working together.

If you are not having any ground loop issues at all with your aem uego you should be using a gain of 1, meaning the value off the orange wire with the sensor unpluged will exactly match your aem pro parameters o2#1 raw.

I tuned my car againsed a motec, fjo and a techedge wideband and i was consistantly leaner by around .6 - .8 afr! If your gain is other than 1 it could be off by more!

Aem uego :thumbsdow

If you look up my username on the ems forum i had mega problems with my uego but finally figured it out completely. Something to look at.

Always dyno tune!

I used the interpolation (calculate feature) when first trying to setup the table but the values given were inconsistant with real world math, for reasons unknown. From what I have read up on in the AEM forums, altitude should not affect the boost comp table. The table will still read from -100% (net fuel value change) at -14.7psi, 0% at 0 psi(100% of column value) , and +100% fuel at 14.7 boosted psi(200% of column value).

Remember, atmospheric change is factored into MAP reading, meaning MAP values are absolute values, with the adjustment for altitude pressure change. What you are possibly thinking of is the difference between what a boost gauge (relative) and a MAP sensor (absolute) will read. My relative pressure will read higher than my absolute because of our altitude.

How to set O2 gain -->http://forum.aempower.com/bbs/viewtopic.php?t=6221 followed to a T.
O2 calibration Gain is adjusted for the input to the AEM EMS unit. The AFRS on the gauge matched my AFRs in the software. Since the gauge uses its own power to calculate the input from the sensor, and then outputs the analog signal to the EMS, I know these values to be correct to the gauge. Its possible you had a bad unit, or the entire line is bunk, I will be testing the possibility when I get the chance too with my own AEM wb vs fjo or similar wb.

[Primer_Drift]

Do you have your cal still by chance? If so please email me it back and I will take a look at how it is different, thanks :)

[legendboy]

I'll have a look for my old cal. Its v1.03 tho.

The way your boost comp table is isn't right but I can't remember exactly what the calculation is (i don't really feel like going and diging thru aemforums posts :) ).

Your uego is the gauge type, mine was the non gauge type. So with the sensor unpluged you should be seeing 4v (or whatever corrosponding afr) at the Vout wire. This has to match the aem pro o2#1 raw voltage or the afr's won't be correct. I know js said you just adjust the gain to be 4v in aempro but if you do some searching you will find this isn't exactly the way it needs to be done. It should say what i'm saying in the user manual for your uego.

[Primer_Drift]

Yep it did, at first it was slightly over 4 volts if I remember, so dropping it to .95 was how I got 4 exactly.
It also required setting up the o2 sensor Cal table (in the instructions), essential to get the right AFRs in the software. Thanks for your input Corey, I'll do some searching in the AEM forums for this calculation you refer to.

[Dj_FL!PSW!TCH]

if all four are like that then I doubt its the tune that ultimately caused that. I"ve never seen all 4 ring lands go at the same time. definately something in the build up, Wana sell me the block? ;)

are you sure those are overbore pistons in a over bore block? lol :rofl:

[Primer_Drift]

:rofl: No, block alone is worth more than my car..thats without the internals. Spend your own damn fortune making one! :whipped:

[rage2]

Originally posted by Dj_FL!PSW!TCH
if all four are like that then I doubt its the tune that ultimately caused that. I"ve never seen all 4 ring lands go at the same time. definately something in the build up
Good point. I didn't realize all 4 cylinders were scored like that until I re-read the original post. Thought it was just 1.

Definately an issue with the buildup... someone provided you with improper specs maybe?

[Dj_FL!PSW!TCH]

Originally posted by Primer_Drift
:rofl: No, block alone is worth more than my car..thats without the internals. Spend your own damn fortune making one! :whipped:

take your engine to another shop, get it sleeved, and assemble it yourself in your kitchen y0

[legendboy]

I'm just curious why you guys would say it couldn't be caused from a lean condition if all 4 cylinders look that way? My way of thinking is the other way around.

If it was only one cylinder there is a better chance its mechanical related ie. skirt wall clearance, out of round, ring gap..etc..

Usually if someone runs lean hard and long enough all 4 piston ringlands will be damaged! I have pics to prove it :rofl:

[rage2]

Because:

a. It's in the same place in all 4 cylinders
b. No mention of head gasket issues.
c. No mention of any other damage.
d. your 4 ringlands did not break in the exact same place did it? ;)

There were a lot of guys on the 944 forums that had issues with their 2.8L builds because they got bum information from JE on their pistons (bad ring gap specs) so pretty much all of them failed with the same damage in each cylinder within 5000miles.

Of course, I could be completely wrong. Won't know until he tears it all apart.

[Zero102]

Yep, I remember reading about that issue. There was a huge fit around JE's responsibility in the matter IIRC.
I believe it listed the ring gap too low??

[buh_buh]

If your at an 87.5mm bore, you definately have more room, so that shouldn't be a concern. I've seen people go as high as 89mm.

[Primer_Drift]

I'm going to finish pulling the motor tomarrow. Hopefully I'll have some answers on friday.
Some extra info I did not mention in my original writeup (but meant to) is that before tearing it down I did a WARM compression test. #1 cylinder (pictured at the top) had normal compression #2 was down 5-8psi, 3 was at 130 (down 25), 4 was at 90psi (-65psi). Plugs 1 2 4 appeared oily, while 3 appear normal, if not slightly ashen (coolant?) but not white like a detonated plug. Headgasket appeared to be ok, but trouble signs on a MLS could be elusive if it has just started to leak.
Could be compression rings or the headgasket with those kinda
numbers.


Originally posted by Dj_FL!PSW!TCH


take your engine to another shop, get it sleeved, and assemble it yourself in your kitchen y0

Its got a Darton MID sleeve kit installed y0 :)

[Dj_FL!PSW!TCH]

Original Post Removed. (Please read the Forum Rules and Terms of Use (http://forums.beyond.ca/articles.php?action=data&item=1) before posting again, or risk getting banned).

[legendboy]

Here is a picture from the mike laskey racing team in cali and a blurb of information that might help explain things.


http://members.cox.net/elaskey/p2x.jpg

"You scarred up the skirt of the piston, messed up the surface of the cylinder wall and maybe even egg shaped the cylinder. New pistons are tapered smaller on the top to larger at the bottom of the skirt. Your piston to wall clearance is measured at the bottom of the skirt. As the engine warms up to operating temperature, the upper portion of the piston begins to expand slightly. The bottom of the skirt does not expand much. When you boost in a lean condition, the upper part of the piston expands quickly. Since the ring land area is cut smaller than the tapered skirt below it, the first part of the piston that pushes into the cylinder wall is just below the oil ring. Thus you will see the worst scarring on your piston right under the ring lands where the excess heat is the highest. The more heat that is generated, the harder the piston pushes into the cylinder wall. The uninformed would blame the piston damage on bad piston to wall clearance. Untrue. If that were the problem, the damage would show up at the very bottom of the skirt. What has happened is that you have expanded your piston to the point that it has just ground itself into the cylinder wall. Keep expanding the piston by super heating it and it will push your cylinder egg shaped and maybe even balloon out the cylinder slightly. At the same time this is happening, your ring lands will begin to distort to where they will never seal properly again. Sometimes after doing this, the engine will still run but it will be a smoker. This all happens in a few seconds of high boost with a lean air fuel ratio. Also it can happen from 500 freeway miles of driving where the tune up is off enough to build excess heat at a slower rate, thus doing the same damage over a longer period of time…but the end results are the same. Death to your pistons and cylinder walls. "

[Primer_Drift]

Great info :thumbsup:
I'll compare my pistons to those when I take them out

[Zero102]

Very helpful information.