Random misfire at idle 454 vortec

[elevatorman]

Compression out of the hose, and then only to the top of that stroke, not all the way around to align the damper with the timing indicator.

I am freakin' amazed that when you turn the crank "too far" to align the damper and indicator, that when you finally apply pressure to the cylinder, it doesn't spin the crankshaft.

There's two reasons to have the piston at TDC when doing a cylinder leakdown test:
1. The piston is at the top of the bore, which traditionally has been the most-worn part of the cylinder, leading to highest indicated leakage, and
2. The connecting rod geometry means there's no rotational force on the crank. Get the rod/rod journal offset from TDC more than a few degrees, and pressure on the piston tends to turn the crank so the piston goes down to BDC.


There are some folks who want to do cylinder leakdown testing at ~12 degrees after TDC. They figure that's about where peak cylinder pressure occurs, so that's where testing should occur. But they have to mechanically lock the crankshaft in place or the pressure will spin it, and the piston heads down the bore to BDC. I cannot be bothered to do it this way. Never, ever, not once.

Hazzard Fraught used to sell a "leakdown tester" that according to the instructions, was to be used with the piston at BDC, and 15 psi. Fookin' junk, from a seller famously known for selling Tool-Shaped Objects that aren't actual tools.

Dont be amazed lol i was holding the engine back with a crank shaft tool. And im also using a OTC leak tester

 

[elevatorman]

If you don't think that being a few degree's before or few after TDC matters , hold your hand right here while I bump this propeller a few degrees forward. It WILL hit you hard. If it goes in reverse, the trailing edge is sharp, it WILL send you to the the emergency room.
You got all plugs out, turn the crank by hand, use the hose off the compression tester to feel when it coming up on TDC. Spinning with starter waiting for the whoosh it's already past TDC. There's hardly any crank resistance with all the plugs out.
Using the starter just complicates it. Every time you hit the starter, i.e., fuel pump and injectors ? Coil grounded ? All that excess fuel vapor + wild spark = a very bad day.

I was using the starter by jumping the relay and ive had the fuel pump relay removed while doing this at this point i am rotating by hand with both valve covers off trying too do this as correctly as possible

 

[elevatorman]

Elevatorman,

THANK YOU for following through and capturing the requested data for both
the Compression Test *and* the Leakdown Test. Listen, I'm flirting with running
late for an appt. at the VA hospital, but thanks to the results of your leakdown data
I think I've had a troubleshooting Eureka moment. So to get this out asap for the
GMT400 community to review, here's my thoughts, bullet style:

* The *camshaft* is the mechanical computer that sets the firing order for the engine.
* The PO installed a camshaft featuring the "4-7 swap". (This is popular with the outer orbit motorhead devos for reasons we don't care about right now.)
* The factory firing order that we all know, love, and can recite anytime, anywhere: 1-8-4-3-6-5-7-2.
* Hot, hip 4-7 swap firing order: 1-8-7-3-6-5-4-2. (Note: It does remove the "5-7" adjacent firing & moves it to 4-2? Taming 5-7 hot spot is claimed?)

* A compression test will NOT find a firing order change.
* A leakdown test performed sequentially (every 90° of crank rotation) will uncover a non-standard firing order.
* And if the cam has the 4-7 swap, you will end up with *exactly* the test results that you documented, for you
will end up attempting to test cylinders #4 and #7 in the vicinity of the Exhaust/Intake overlap {360° out} instead
of the both valves closed top of the compression stroke. (!)

While I'm taking care of business, check out something I drew to help you follow the troubleshooting theory:

You must be registered for see images attach

NOTE: I've attached a copy of your test results so that you can use them against the chart above.

Q1: How are your spark plug wires run? I'm assuming the factory firing order to match the cap?
Or did you replace them one at a time, and maybe the PO followed the new cam's firing order?

Q2: Is the wiring harness for the 8 sequential fuel injectors still stock, or did the PO attempt
to modify/hack the wiring harness to follow the new camshaft firing order?

Q3: Could you attempt to prove/disprove the firing order that's currently installed in your truck?

****

More later, gotta show up for my Dr's appt.

Best of luck figuring out exactly where and how much total 'value-added' the PO left behind in the
engine bay besides the aforementioned inline resistor to one of your sensors.

Cheers --

Ive just performed testing again on 4 and 7 and was able too successfully do first try now that i fully understand the process. As far as your theory i dont believe that to be the case.
*ive changed plug wires multiple times too factory firing order.

*timing chain and sprockets were changed appeared factory gm stamped sprockets i find it hard to believe someone would do a cam without a chain.

*Been through the whole injector harness no hacking or changes found.

That all being said ive got the updated data for 4 and 7 below i did not retest all the other cylinders. My theory being if being not tdc on a few would make the test results worse and to me everything except number 1 appears in range. Which im confident that cylinder was done correctly as with 4 and 7 now.
But im here for help and im open too suggestions and i do appreciate all the input.
The way i see it number has been my problem child it always shows up first on the misfire graph before any other cylinders start coming and going every so often, which i think is because the computer is scrambling trying too fix 1.

 

[Schurkey]

I suggest you run the entire engine again--following the correct procedure with piston at TDC-compression. Call it "extra practice".

Any valve that has leakage, tap with a small, plastic-tip hammer. See if the leakage disappears.

 

[Road Trip]

Im back with an update have had the rona and plumbing issues so i am back on the truck.
Ive logged all the data i can and ive noticed something weird with number 7 and 4 any time i try too bring the cylinder up tdc the exhaust valves open befor i get there. Ive repeated both of those cylinders numerous times with the same results.
My data below check mark means air heard. X nothing heard.

Hello elevatorman,

I've been pondering your results. At this point, my problem definition is as follows:

* 'Good' compression test results: 142 - 147 - 148 - 150 - 150 - 153 - 154 - 155
* All nearly within +/-5%, median = 150, arithmetic mean 149.8

* Cautionary leakdown test results:

* Good: 7 of 8 cylinders 9% or less total leakage.
* Worst cylinder: (#1) 12% leakage. Both intake and exhaust valves audible leakage.
* Best cylinder: (#5) 3% leakage. No intake or exhaust valve leaking noted.
* Worrisome: 7 out of 8 cylinders have audible exhaust valve/seat leakage?
* Focused issue: 7 out of 8 cylinders have good intake valve no audible leakage? (#1 is the outlier)

Leakdown testing results analysis:

Perfect leakdown test results are when all 8 cylinders present ZERO audible intake or exhaust
leakage, combined with single digit piston ring/cylinder wall leakage.

Cylinder #5 is the only one to meet this criteria. With no audible valve leakage, then all
leakage is then assumed to be primarily piston ring <> cylinder wall leakage. In unusual
cases piston ring <> piston groove leakage can be the root cause. Stuck/frozen or broken
rings (think motor sitting for a long time in uncontrolled storage) will leak like a sieve.
In terms of total leakage, less than 15% is considered acceptable, less than 10% is good,
and less than 5% for a street motor is excellent.

Using cylinder #5 as an example, your test setup recorded 97/100, which is 3%. (Relative
to itself, not absolute from tool to tool per previous conversations here & elsewhere in the
forum.) But I for one am willing to take this at face value.

I am going to go so far as to say that it looks on balance that you have a combination
of excellent ring seal combined with marginal exhaust valve sealing, with the #1 intake
valve also failing the check.

In English, from here it looks like your short block is muy bueno, but for the smoothest
operation your cylinder heads could use a valve job.

A little more food for thought concerning the leakdown test results vs rough running.
Think about how a 4-stroke engine goes through a complete cycle. After the power
stroke, we immediately go into the exhaust stroke. Towards the very end of the
exhaust stroke, the piston is actually chasing the exhaust valve up towards the
cylinder head.

Right in the middle of all this, as the piston is slowing to a complete (momentary) stop,
the intake valve starts to open. This means that for a short period of time, both the
exhaust and intake valves are open. (This is the overlap period, very short on a smooth
idling stock cam, quite a bit more on a high rpm performance cam.)

Right in this time period, the exhaust valve closes while the intake valve continues to
open and the piston starts to accelerate away from the cylinder head. IF the exhaust
valve/valve seat is leaky, guess what happens during this period? Think of the extra
exhaust gas dilution occurring in this high vacuum environment as an overactive EGR
event on top of the EGR already being commanded by the ECM upstream in the
intake manifold. (!)

To finish explaining why (slightly) leaky exhaust valves can cause rough running, but
only during light throttle situations, let's let the intake stroke complete, and start
up the compression stroke. Most folks think that the intake and exhaust valves
are only held shut by the valve spring above.

But there is a 2nd force acting upon the valves, and that's the increasing cylinder
pressure pushing on the circular face of the valves themselves, providing dynamic
additional closing force. (Total square inches of valve face area x cylinder pressure
in PSI.) Especially with a successful cylinder firing, the pressures are way up there.

BUT, if too much exhaust gas dilution from the leaky exhaust valve seal coincident with
running 14+ inches of vacuum in the intake manifold (because the throttle blades are
barely cracked during cruise/coasting driving) causes there to be nothing burnable
in the vicinity of the spark plug gap when it fires...then we end up with a misfire.

But guess what? IF we step into the gas, and the ratio of uncontrolled EGR
(from the leaky exhaust valve) is now being overwhelmed by a rich, dense
intake charge, now the cylinder is going to easily fire with authority.

Apologies for that dense passage. And the fact that you may have to read
it a couple of times in order for it to make sense.

The point I'm trying to make is that back in the pre-emission days IF we got
a compression test with a +/- 5% spread and we were getting a rough idle or
light throttle roughness? We'd simply fix the issue by cranking out the idle
screws a 1/4 or 1/2 turn, and all would be right with the world. (In extreme
cases, we might also read the bone white spark plug porcelains & rejet the
primaries to a little less stingy size. :0)

Because we'd cover up this uncontrolled EGR situation without even really
knowing exactly why. (Back then, most compression testing on customer
cars was simply to quantify gross failures in order to sell a valve job, and I
remember reading about how only the hottest circle track dudes fussed around
with exotic procedures like leakdown testing.)

But today? With the O2 sensors holding the truck to a strict 14.7:1 air/fuel
ratio, even at idle, up through part throttle (ie: not full power open loop) ...well
now we can no longer cover this up. And these big bore big blocks will oftentimes
run dead smooth under load, but be a little sloppy at idle and light loads.
(With optimal quench this issue isn't as bad, but with normal tolerances
there are simply a lot of square inches in the combustion chamber where
there's no spark plug to light off the one good spot of combustible mix
thanks to the marginal exhaust valve sealing.)

And all this can occur with perfect spark plugs, wires, cap, rotor, coil, etc.,
*if* you have exhaust valves where you can hear the 100 psi air escaping
out the tailpipe at TDC on the compression stroke.

To recap, your test data suggests to me that you have a great short block,
the intake valve in cylinder #1 is tired, and a total of (7) exhaust valves are
leaky. A valve job would go a long way to laying the foundation for a 454
that runs as smooth as new at all expected speeds and loadings.

More in a bit, for I want to ask for some additional data from you if you
are still interested in pursuing a fix. I appreciate your willingness to dig
deeper than most, and being so patient with the process.

Cheers --

 

[Road Trip]

Im back with an update have had the rona and plumbing issues so i am back on the truck.
Ive logged all the data i can and ive noticed something weird with number 7 and 4 any time i try too bring the cylinder up tdc the exhaust valves open befor i get there. Ive repeated both of those cylinders numerous times with the same results.
My data below check mark means air heard. X nothing heard.

Greetings elevatorman,

My previous reply (#75) was setting up this request for you to collect data
one more time. The good news is that your compression test results are
good to go, so they don't need to be repeated.

I do need you to run the leakdown test one more time, with the additional
constraint that you run it in the standard firing order: 1-8-4-3-6-5-7-2

There was a bit of discussion about all this, but to simplify matters, mark
your harmonic balancer every 90 degrees and add the firing order like so:

You must be registered for see images attach

Leakdown testing run in the firing order adds the camshaft firing order verification to the data set for no extra charge.


And in order to improve the communication of what you observe vs what we understand
remotely, here's a revised version of the Compression Test & Leakdown Test Scorecard:

You must be registered for see images attach

NOTE: The biggest change to the leakdown testing is substituting a '0-10' scale for the check mark & 'X' notation.

IF this time around you get the correct leakdown test results while following the firing order then we'll chalk up what we
saw in Reply #61 as pure learning curve. But IF you once again run into the same snag with #4 & #7 cylinder testing,
then this would document an additional reason why this 454 just doesn't want to run as smooth as it should.
At the very least the spark plug wires must follow the cam's firing order, and for optimal results the #4 & #7 injectors
should also be wired to follow suit.

At this point, if the #4 & #7 cylinders don't test correctly, then performing the test in the 1-8-7-3-6-5-4-2 (4/7 swap) firing
order and having all 8 cylinders test correctly would be proof positive.

To recap, in 4+ decades of troubleshooting motors, most fixed just with fresh tune up parts & no exotic theory required.
But every once in a great while you cross paths with an engine that just won't take a tune up & run smoothly. I could be
wrong about this, but in my humble opinion the results that you got in Reply #61 were not due to an error in your testing,
but it's an indication that all is not stone stock in your big block.

Thanks in advance for you checking all this out so carefully. We've got at least one other 454 that is resisting repair, and
I'd to keep a second one from this same category if there's any way we can avoid it.

Cheers --

 

[elevatorman]

Sorry for the delay ive been busy and ultimately dove into the heads over the last few weeks. What i found was a worn exhaust manifold surface at cylinder 1 and replaced the manifold.
I know some may disagree with my approach to clean the valves but i lapped the valves and installed new mls head gaskets.
My reasoning for this is i just cant justify dumping more and more money into this with only data getting better but not my problem.
So that being said the truck idles like crap still but LT fuel trims both banks are holding steady at 7% now. I also caught my egr valve fluctuating too 5% without a demand and changed it now its holding 0%. What im confused about is my iac counts at 70% at operating temp in park.
I will be doing the leak down test again my misfires have moved to cylinder 7 for the most part. I did keep track of which valves and all other parts came from so everything went back to where it came from.
This is just an update until the leak down is performed again

 

[Road Trip]

Thank you sir for the update. I 100% understand what it's like having a lot of
irons in the fire. :0)

Looking forward to your next leakdown test results. Once we fully characterize
the mechanical side of the equation then we can follow the data wherever it takes
us.

Best of luck with the hunt for the fix --

 

[Schurkey]

What i found was a worn exhaust manifold surface at cylinder 1 and replaced the manifold.

Could have allowed an air leak into the exhaust system, driving the O2 sensor to report false "Lean".

I know some may disagree with my approach to clean the valves but i lapped the valves and installed new mls head gaskets.

Were the guides wiped-out? Valve seats and valve faces still in usable condition?

MLS head gaskets were probably not the best choice--wildly expensive, and sensitive to flatness and surface finish of the gasket surfaces.

the truck idles like crap still... ...This is just an update until the leak down is performed again

OK.

 

[elevatorman]

Were the guides wiped-out? Valve seats and valve faces still in usable condition?

I could not find any extensive wear too the guides. Seats and valves looked ok for a 240k mile motor nothing stood out there was some light carbon pitting i guess you could call it on all the exhaust valves. Also super careful cleaning up the heads and block for what it was worth.