Driving it off the cliff

[blue62]

Originally Posted by Stl-986

Just had a better look at the plugs. Makes no sense. 1-3 look good/normal to me. #4 though looks almost new.

Yes....made sure I am working on the correct sides.
1-3 being passenger, 4-6 being driver.

They look about right considering random misfires on bank one.
Shows that they are firing most times.

I was just looking at the fuel trim data you posted in post #2 I saw something odd.
If I am reading it correctly RKAT is the idle range.
FRA is upper idle to mid range RPM..
Data is showing around 700 RPM
So looking at fuel trim I would expect to see positive fuel trim (RKAT idle range) for the bank with the misfires.
But you have a negative fuel trim for bank 1
and a very strong positive fuel trim for bank 2
Bank 2 trim should be very near 1 for trim it is at 3.42
Is there any possibility that the bank1 and bank2 pre cat O2sensor wires are crossed???

[Stl-986]

With this car, anything is possible, but that would require flipping the entire engine harness I think. The wires are only so long. dont see how I could even connect the front to rear, rear to front let alone bank 1 to bank 2.

[blue62]

Originally Posted by Stl-986

With this car, anything is possible, but that would require flipping the entire engine harness I think. The wires are only so long. dont see how I could even connect the front to rear, rear to front let alone bank 1 to bank 2.

Ok thought I would ask
your fuel trim data on bank 2 is what I would expect to see for bank 1.
When you have all three cylinders on the same bank misfiring your O2 sensor is going to see the unused oxygen from the misfires. not the unburned fuel.
So the DME/ECU will see that as a lean condition and add fuel.
But your car is taking away fuel at idle on bank 1
At the same time it is adding lots of fuel to bank 2
Which has no misfires.
Very odd.

your misfires are mostly at idle correct?????

[Stl-986]

well, take a look at this. Got it all back together. all this time I had been focusing on bank 1. After pulling coils/plugs for the compression test I went ahead and put the new coils & plugs in for bank 2. After seeing what plug #4 looked like had me scratching my head.

Still not seeing the vacuum I should but for the past 20 minutes this is what it has looked like



Idle is too high, saw it shoot to 1012rpm, but....no misfires at idle. Car is up to temp too

[blue62]

Originally Posted by Stl-986

well, take a look at this. Got it all back together. all this time I had been focusing on bank 1. After pulling coils/plugs for the compression test I went ahead and put the new coils & plugs in for bank 2. After seeing what plug #4 looked like had me scratching my head.

Still not seeing the vacuum I should but for the past 20 minutes this is what it has looked like



Idle is too high, saw it shoot to 1012rpm, but....no misfires at idle. Car is up to temp too

Ok now we are talking fuel trims look like they the should.
You have excellent compression.
Is this car a tip trans. Automatic transmission??? what ever they are called???
Idle RPM is usually higher then a standard trans.
Do you know the idle spec???
I am going to view you video a few more times then get back to you on vacuum testing.

[ike84]

For compression my understanding is multiply atmospheric pressure x compression ratio, so for our cars it should be on the 160 range. But, conditions vary as does equipment precision, so I always look for consistency rather than raw numbers. This is one reason why leakdown tests can be more helpful. Either way though your numbers look great.

Now that everything is back together you have strong vacuum and your fuel trims are spot on. Looks like you fixed it!

With the idle, it takes a few minutes after a big change for it to level out. Was your ac on though? Idle without ac should be around 650 but idle with ac on is much higher, closer to 1k.

Sent from my SM-G970U1 using Tapatalk

[blue62]

Originally Posted by ike84

For compression my understanding is multiply atmospheric pressure x compression ratio, so for our cars it should be on the 160 range. But, conditions vary as does equipment precision, so I always look for consistency rather than raw numbers. This is one reason why leakdown tests can be more helpful. Either way though your numbers look great.

Now that everything is back together you have strong vacuum and your fuel trims are spot on. Looks like you fixed it!

With the idle, it takes a few minutes after a big change for it to level out. Was your ac on though? Idle without ac should be around 650 but idle with ac on is much higher, closer to 1k.

Sent from my SM-G970U1 using Tapatalk

that formula for compression does not work
My 1969 Jaguar E-type XK engine has a 9:1 CR factory spec.for compression is 175 PSI
My 1960 Austin healey BT7 engine has 9:1 CR factory spec. for compression is also 175 PSI

[ike84]

Originally Posted by blue62

that formula for compression does not work

My 1969 Jaguar E-type XK engine has a 9:1 CR factory spec.for compression is 175 PSI

My 1960 Austin healey BT7 engine has 9:1 CR factory spec. for compression is also 175 PSI

How is that possible? Not questioning your experience, but the math there doesnt add (or multiply) up lol.

Sent from my SM-G970U1 using Tapatalk

[Stl-986]

Originally Posted by ike84

How is that possible? Not questioning your experience, but the math there doesnt add (or multiply) up lol.

Sent from my SM-G970U1 using Tapatalk

off topic, but, I know what you are saying here, but.....his numbers are right. I thought the same about the multiplication you mentioned. Even our cars the math doesn't work cause these are 11.1 compression rate which means they should top out at 163.17. Not the case though. This isn't my video but one I found while going through the process today: https://www.youtube.com/watch?v=cznw9idwpkU & this: http://986forum.com/forums/boxster-general-discussions/35231-compression-test-results.html

in my own words....no clue how it's all calculated. Too much work to figure it out.

[blue62]

Originally Posted by ike84

How is that possible? Not questioning your experience, but the math there doesnt add (or multiply) up lol.

Sent from my SM-G970U1 using Tapatalk

Well you can start with things like valve overlap.
Valve timing. air temp.
The simple math formula of multiplying atmospheric air pressure by CR assumes things like.
Intake valve wide open exactly at BDC and closed at exactly TDC with no exhaust valve overlap.
Piston cranking speed is another example. Watch the needle on the vacuum gauge in the video's in this thread.
As RPM increases vacuum increases. = increased cylinder fill.

So the simple math formula of multiplying atmospheric air pressure by CR will not work to give you cranking PSI.
Look up any engine specs. you can find that show CR and cranking compression PSI
That simple math formula will not give you stated cranking PSI

As for leak down tests being more helpful than compression tests:
A leak down test is a "static test" of the cylinders ability to seal. (hold pressure)

A compression test is more of an operational test of volumetric efficiency (cylinder air filling ability) which is directly related to vacuum
and the ability to "build pressure"
Very closely related but you can get a little different info from each test.
So it depends on what info your after on how helpful a given test method is.

[blue62]

STL-986

When you get time graph your O2 sensor "voltage signals" for all 4 O2 sensors at idle.
and RPM. Then post the graph so I can have a look see.

They are in actual values under the engine module.
Before you do the graphing go in and click on each O2 sensor "voltage signal" with key on but engine off.
That should show you the "BIAS" voltage on the signal wire.
It should be 0.45V That will test the signal wiring between the DME/ECU and the O2sensor.
Then click on the O2 sensor heater signal again with key on but engine off. This will test the wiring and voltage to the O2 sensor heater circuit. I think the spec. is 4.5V or close to that.

Then get the car up to operating temp. and graph the 4 O2 sensor voltage signals and the RPM.
Just let it idle during the graphing

[Stl-986]

That's just it, dont have strong vacuum at idle. I have great vacuum when the rpms are high, but think I absolutely have a vac. leak.

I can get vacuum up to 20hg around 3k rpm, let off the throttle and watch to drop slowly. Seems to drop faster when all the fans kick in.

Misfires did come back after an hour. They reset themselves to 0 and then when the front fan kicks it they start to climb up again. They are not happening anywhere near like they were before though.

Didn't grab a chart, but I started to graph voltage the last time and noticed that it is around 13.5 and then when the engine fan kicks in it goes to down to 13.2 for a bit, then when they front cooling fans kick in it goes down to 12.8 but does start to climb back up to around 13.4.

I'm just rambling, but the past 2 years I have been working on this issue I found 1 person who was having similar misfires and what fixed it was a new alternator. Electrical issues can cause some weird things to happen but I have kept it in my mind since seeing it

AC was off

Here is the video:
https://youtu.be/XqlX4KJZ24U

[Stl-986]

Yes its a tiptronic. Had to get the wife something she would drive

[blue62]

Originally Posted by Stl-986

Yes its a tiptronic. Had to get the wife something she would drive

Do you know the idle RPM spec???
or how far from spec. it idles at??

[blue62]

STI-986

From what I see in your video you "DO NOT" have a vacuum leak
The test for a vacuum leak is like this.
Get the car fully warmed up.
Get idle as close to spec as possible.
Hook the vacuum gauge up.
Let the car idle for 3-5 minutes and observe gauge reading and needle action.
At idle your in the green zone on your gauge. and the needle is very steady.
That is good

Now bring RPM up to 2000 RPM the needle should rise a little and stay steady.
that is what your gauge is doing at 3000 RPM.
If you had a vacuum leak the needle would fall when you held RPM at a higher level.

Do the vac test again. get the car at operating temp.
Do the above test.
Do it closer to 2000 RPM

Then let the engine idle again for a minute or two.
Then do a "snap throttle test"
Go from idle to floored to off throttle very quickly.
Needle should drop to near 0 then rebound above idle reading very quickly.
then it should settle back to idle reading.

Your going to have a little lower vacuum reading at idle then your expecting.
Your reading very very close to my 2000S
The reason is that you have things like vacuum canisters on both your SAI and your EVAP systems along with all the valves and hoses.
If they were blocked off you would probably gain 2 in.hg

Another thing that is going to affect vacuum is the Vario Cam system.
Cam timing changes at around 1500RPM so as you increase RPM the vacuum gauge readings may not be what you expect.

[Stl-986]

Looks like it's the same as Mt cars. 750 +/- 40

[Stl-986]

Watch it again, you can see it decreasing while it is idling. The 1st 25 seconds are so you can see it drop, this was at idle.

Did the snap test a couple times, never goes to 0. At the end of the video is it coming down after being held around 3k rpm and you can see it decrease.

With all that said, it is WAY better then it has ever been.

I'm letting it cool down. In the past when it would misfire like crazy turning the car off for a while and then starting it back up again would turn the MIL on. Want to see if that happens again.

My biggest thing I want to solve for now is to get it to just pass emission testing. At worse I want it to get to a point where all 5 of the tests pass even if it does throw misfire codes cause I have a waiver that will get approved if that happens.

[Stl-986]

At 1:24 is when it goes back to idle and you can see it dropping.
at 1:40 the cooling fans start to kick in and it drops even more but hard to tell.

I'll do another video in a bit. want to get things setup with the gauge by the cluster. Want to see if you can spot another issue that happens had really has me stumped (but I think it is all related)

[blue62]

Originally Posted by Stl-986

At 1:24 is when it goes back to idle and you can see it dropping.
at 1:40 the cooling fans start to kick in and it drops even more but hard to tell.

I'll do another video in a bit. want to get things setup with the gauge by the cluster. Want to see if you can spot another issue that happens had really has me stumped (but I think it is all related)

Yes watching RPM and the needle is very important.
Hard to relate everything about using that simple little vacuum gauge via the internet

[blue62]

Originally Posted by Stl-986

At 1:24 is when it goes back to idle and you can see it dropping.
at 1:40 the cooling fans start to kick in and it drops even more but hard to tell.

I'll do another video in a bit. want to get things setup with the gauge by the cluster. Want to see if you can spot another issue that happens had really has me stumped (but I think it is all related)

I turned the sound on so I could hear the engine. That little drop in vacuum when the fans kick in is due to a little drop in RPM.
That slow drop at idle looks like you could have some restriction in your exhaust.
Cats may be partially plugged which could lead to random misfires.

When you do your next vacuum test bring the RPM to 2000 and hold it as steady as you can for 2-3 minutes watch the needle. if it drops slowly you have some restriction in the exhaust.