sancho

The leak-down test results...

<b>
________________________________
CYL # PSI IN PSI OUT %DIFF |
----------------------------------------------------
1 95 75 78.95 |
3 102 98 96.08 |
5 96 82 85.42 |
7 94 84 89.36 |
2 100 98 98.00 |
4 96 88 91.67 |
6 94 47 50.00 |
8 100 97 97.00 |
----------------------------------------------------
</b>

<i>[I have attached these same results as a formatted TXT file for easier viewing, if needed.]</i>

I know enough to know that those are not entirely impressive results. I guess what I ultimately want to know, then, is what these results mean with regards to the problem that I've been trying find for several months now: engine overheating (and quickly, too).

The biggest thing that's staring me in the face are those results for cylinders #1 and (especially) #6. I could hear considerable more "hiss" comming from the engine on those cylinders, and it sounded like it was comming form around the lifter valley (the intake is off). So I got a lenghth of PCV piping and listened around; much of the leakage for #1 sounded like it was comming around the lifter valley and the intake port, and all of the hiss for #6 was definately comming from out of the EGR duct in the head (covering the hole would mute the sound). On #6, I thought that if the hiss was comming from the EGR duct then the exhaust valve might not have completely closed (although both lifters looked to be completely down). So I tried moving the crank a bit and then testing again, and I only got the same results.

So, again, I'm just looking for some opinions on what these results might mean for an engine that has not left the driveway since a complete rebuild over seven months ago, and what my next steps should be in diagnosing this problem.

Thanks.

sancho

I'm just gonna throw this in, just in case. Here are my compression-test results...
<b>
DRIVER'S SIDE
1. 193 PSI
3. 192 PSI
5. 189 PSI
7. 202 PSI

PASSENGER'S SIDE
2. 200 PSI
4. 200 PSI
6. 200 PSI
8. 212 PSI
</b>
Looking the results of both tests, I think it's safe to say that the problem isn't going to be something as easy as a blown head-gasket...

F-BIRD'88

Possible cracked cylinder wall.

Repeat the leak doen test on the low cylinders but this time
clock the crank with that cylinders piston at TDC, then BDC
then Half way inbetween. If the % of leakage is a lot les with the piston locked at TDC then when the piston is locked further down the bore then the air is leaking out of a crack in the cylinder wall. This would explain the overheating too.

Vader

Sancho,

You seem to have conflicting results. By the leakage test, cylinder 1 appears to be a little worn, and 6 looks like it has a serious problem. Hoever, you dynamic compression results seem to contradict that, and both 1 and 6 appear to be just fine.

I would perform the leakage test again, but be certain you have the cylinders at TDC. As for "clocking" the cylinders for the test, be very careful to avoid having the compressed air drive the piston down in the bore while you're in there. Unless you have some means of securing the crank in place, you might not have much success with that method.

Since you have conflicting results and apparent leakage from a couple of exhaust valves, try removing the rocker covers and rocker arms. While the suspect cylinder(s) is under compression, lightly tap the top of the valves in that cylinder with a rubber mallet or light hammer and block of wood (to protect the valve stems). This may help seat the valves a little better if there is any carbon or debris near the valve faces. I've been able to get them to seal and perform a better leakage test by this method.

My hypothesis is that you may have valves adjusted too tightly, bent valves, carbon, or some other valve train issues. Leaking valves will cause low power output and overheating as well.

sancho

Ahh... the valves. Just what I was thinking (and hoping) it might be. It makes more sense to me that it be the valves than something between the piston and the cylinder wall because any leakage was best heard at either an intake port (#1) or an EGR/exhaust port (#6).

I'll try your suggestion, Vader, and post back here when I have results...

sancho

Well, I just pulled the rocker arms from #1 and ran the leak-down test on it, and I got the exact same reading that I posted (so it's not leaking because the rocker-arms were too tight). After that, I left the pressure on, got a piece of wood and a rubber hammer, and then tapped both valves several times (got a loud "pop" each time from the compressed air). When I looked at the gagues after doing that, I still had the exact same reading.

I doubt any significant carbon buildup would have occured to be causing valve leakage, as these are somewhat-freshly rebuilt heads with only a couple of hours (mabye) of running time.

I understand that the purpose of a leak-down test is to determine the condition of the cylinder walls and rings, and so my results for #1 and #6 should suggest something being worn between the rings and/or the cylinder walls. However, I have noticed that, with both of these cylinders, any audiable leakage clearly comes from the EGR ports on the heads (I rechecked #1 again and the leakage is definately comming from the EGR port--not the intake port). On both #1 and #6, the hissing sound of air leakage is muted when I block the EGR port with my hand. I would think that leakage past the rings would only be audiable from within the crankcase and not at the heads.

So, what that roughly suggests to me is that the bad results for #1 and #6 have something to do with the heads and not the cylinder walls or rings. Anyone agree/disagree with that?

I'm basically trying to decide if I can just get away with pulling the heads (i.e. they are the problem) or if I need to yank the whole engine (i.e. because the problem is somewhere on the bottom-end).

Thanks for the help.

F-BIRD'88

A cracked valve seat could explain the leak and the overheating.

Loosen the rockers right off,
Pour some water down the intake port that is leaking.
Fill up the port.

A leaking valve or seat will bubble a lot.
Do the same on the one that the egr/heat raiser is hissing.
loosen off both the valves on the center two cylinders and fill the
egr/ heat raiser port with water. Again the water in the full port will bubble if the valve is leaking.

Be sure to purge the cylinders of water by kicking over the starter with the plugs out before restarting the motor
as a cylinder full of water will bust a piston. (Hydro lock).
I think your going to be tearing this one out.
( multiple issues here)

Kingtal0n

my engine was doing that, but only on #8. for some reason, ##8 always hs the most problem for me. It turned out to be a leaky exhaust valve. REAL leaky. when i took the heads off, i turned them upside down and put gasoline into the combustion chambers, filled them right up. then i felt through the intake/exhaust ports for fuel leaking onto the valves... sure enough.... gasoline all over the exhaust valve from underneath.

Vader

Sancho,

A leakage test is to determine overall chamber condition. While that includes the rings and cylinder walls, it DOESN'T exclude the valves, gaskets, and heads.

The real advantage of a leakage test over a dynamic compression test is that you have some clue as to where the leaks are occurring by the source of the sound. By your decription of the leakage, I think it's a safe assumption to say that your problems are not in the case, but somewhere in the heads.

F-Bird may be on to something. Are your heads aluminum? If so, either a cracked seat insert or loose insert could be the cause. It is also possible that you have a cracked head, warped head, or leaking gasket, but those are less likely. However, you could still have a burned or bent valve.

Another "trick" you can do is to remove air pressure, then try to rotate the valves slightly by grabbing the retainers with a pair of Channel-Lock pliers and turning carefully. Make sure the valve stem is turning as well as the retainer. Go about a quarter-turn, then reapply the test pressure and "pop" the valves again to see if results change. That usually helps seat valves that are carboned, but won't help a bent or burned valve.

It would be interesting to try that, but it seems that you will probably be removing the heads again anyway to repair whatever you find.

sancho

That's about what I thought it was...

I'm going to yank those heads sometime this weekend and see what I can see.

The heads are the stock iron heads that came on L98 engines and, though they are rebuilt, they had >160K miles on them before the rebuild.

I like F-BIRD'88's idea, except that I *really* don't like the thought of pouring water into my engine--not so much because I think it will hydro-lock (can't with the plugs out), but more because I'm afraid of what the rust inside of the cylinder will do. I might try Kingtal0n's suggestion just to see if there's any ovbious leakage, but the heads are comming off either way.


Now I face a potentially new challenge ... I plan on taking these heads back to my machine shop which rebuilt them and having them take a look at them. The thing is, if the source of the overheating was, in fact, cracked/warped heads, then they must have been cracked/warped when I got them back from the machine shop. And, to my knowledge, they were supposed to have magnafluxed the heads and found this out before they even rebuilt the heads.

Or mabye they don't magnaflux heads... I don't know.

Either way, I'll keep you guys posted; thanks!

sancho

Hmm... looks like I'll have to wait to pull those heads. I spent the better part of today pulling the driver's side head, and that much was pretty straightforward.

The passenger's side, however... there are bolts on the exhaust manifold that I can't even get my hand on, let alone getting any wrench/socket that I can get torque on. You guys up north without air-conditoning and those of you with carbed engines probably can't relate (I'm *so* jealous of you guys when it comes to engine work), but between the A/C and emissions junk, I think that my time would be better spent just pulling the whole engine, else that head is going to be a <i>PITA</i> to get off.

In the mean time, I suppose I should be thinking about what heads I ought to get. I would love to get a set of Vortecs, but then I'd have to get a new intake manifold (I want to keep the TPI setup)...

Kingtal0n

can you afford some nice aluminum aftermarkets? if you got some AFR's or somthin you would never need new heads for the street.

On the iron side cant you just re-tap the vortechs to work? I thought the angled bolt holes were the only things stopping you (any TPI owner) from doing just that. Do a search on it, i think youll find it easier than you think.

or maybe not.

F-BIRD'88

Vortec heads also have raised trapazoid port openings.
Won't match up to a Stock TPI intake.

You can buy a Vortec TPI intake from Scoggin Dickey
http://www.sdpc2000.com/

But a Ramjet intake would be the real killer set up with
vortecs.

F-BIRD'88

I would look into a set of Pro Action heads if you want to replace your heads.

www.protopline.com
www.proshaver.com

Kingtal0n

darn. i knew there was a catch!

Well theres always pro-action... or AFR.... or brodix... or AFR... or.. um... dart... or AFR...

Did i mention AFR?

sancho

Thanks for the suggestions, F-BIRD'88 and Kingtal0n; I'll look those up and see how they compare (no promises that I'll know what exactly to look for, though! ). About the only techincal thing that I know that I want to maintain is the compression ratio (i.e. I don't want heads that will give me a lower CR; a little higher is okay). I believe that means that I want ~64cc heads, right?

The thing that I should have mentioned before I asked for head suggestions was that this car is going to be a daily-driver; I don't have any plans for this car to see a track, or least of all participate in a street race. I have nothing against it being "fast", though, and I'm always welcoming anything that will get me more power--I just don't want that power to kick-in up too high.

Also, the new cam that I put in during the rebuild is supposed to be a stock-replacement (I picked that only because I don't know a lot about cams and I figured that GM tuned their cars to be streetable more than anything off of the floor). I'm sure most heads aren't going to live up to their potential with this cam, but hopefully there will still be some increase even with the current cam (and, ideally, I'd still have room to "grow" if/when I decide to upgrade the cam).

Vader

Sancho,

First, even we "guys up north" like air conditioning, too. Not many ThirdGen V-8s came without it. It just takes a bit more time to get the components off and out of the way.

As for head selection (IF you decide to replace them) you might want to be careful about port aize and runner volume. If you are planning to stick with a stock or mild cam profile, and want more of your power down low (torque) you might not want to have huge ports and runners that can be used as rental space. Smaller ports are going to maintain a higher intake velocity and promote better cylinder charging at lower RPMs. That will help give you that stump-pulling torque out of the hole, and cost you a little flow and RPM on the higher end. It sounds as if you would prefer the lower RPM torque over the high RPM power, by your description.

1990GTA

I have been reading this post and although I am not doing a leakage test I am changing my springs using compressed air, I could hear the hissing of air leaking also. Is there a certain amount of air that will leak past or should it be completely air tight. I did tap a couple of the valve stems the way you described and it seemed to help but I could still hear some air.

sancho

1990GTA: Some air will leak past the rings; I don't think it's even possible to have 100% sealing. Even on my best cylinders (i.e. those with the least leakage), I could hear some air leakage comming from within the block.

Another thing to realize is that you will likely hear more leakage when the piston for a given cylinder is at BDC as opposed to TDC (keep in mind that the piston will want to travel to BDC and will do so unless you have secured the crank at TDC). Because cylinder wear tends to increase (taper) towards the bottom of the cylinder, there is more room for air to get past. For your purposes, however, it really doesn't matter where the piston is in the cylinder since all you need is enough pressure to hold the valves up.

The only thing that should be a red-flag, however, is if you can hear the air obviously comming from the head somewhere (i.e. you can hear air comming out of in intake port such that when you block it off you can no longer hear it). Such leakage should indicate that either the head is cracked or that the valves are not sealing as they should (as explained above in one of Vader's posts).

1990GTA

It is coming from the intake ports, and as you describe, you can cover the ports and it slows down. I'm sure the head is not cracked, the car was running fine, I just have it apart for a cam swap. I hope it is just some carbon buildup as stated above, I only put air to 2 cylinders so far because I realized I didnt have the proper valve stem seals so I stopped. Out of the 2 I did one of them stopped leaking after I tapped it a couple times. If they continue to leak out of these ports , what do I need to do?, get new valves, get these valves cleaned up or what? Damn I did not want to pull these heads!

sancho

1990GTA: Well, since your car has been running fine, then you probably don't need to worry. You're likely just experiencing the carbon buildup that Vader was talking about--especially if the leaking lessened as you tapped the valve (you probably just knocked some crap off of the valve).

Trust me, if you say your car was running fine, then you don't have near the problems that I have been having (i.e. car not leaving the driveway since May, 2002).

..."If it ain't broke, don't fix it."

sancho

I was wondering... Assuming that my heads (or valve seats) are cracked, what would be the odds of that happening due to over/under torquing the head bolts?

I'm just curious, because while I did use a torque wrench and did torque all of the bolts in a consistient order in 20 ft/lb increments, I do have a comparatively cheap torque wrench (it was ~$20.00 no-name wrench from Harbor Freight Tools).

Just curious... I don't know how "mission critical" it is to have all of the torques exactly the same, or if a few ft/lbs of difference is just splitting hairs...

Vader

Sancho,

Even if your torque wrench is within 10% accurate, you should be fine. The torque specififation for your head bolts should be 60-65 ft/lb, or a 10% allowance already.

More likely, you have a valve that isn't sealing for another reason.

sancho

Hmm... never tried reviving a thread before, but it might work...

Just thought I'd update my progress on this and see if I can get some more opinions.

Since I last posted, my dad has taken the heads down to the machine shop to have them vacuum tested, and every cylinder (combustion chamber) tested out good. If you're not familiar with the test (I wasn't), what he does is take a suction-cup type thing, puts it on each intake and exhaust port (individually), and tests the vacuum on each. Beyond the vacuum test, the guy at the machine shop magnafluxed the heads, and he couldn't find any cracks.

So, looks like there's nothing wrong with the heads. I'm still a little confused, though, about why I heard obvious leakage from the EGR ports on the cylinders that tested poorly on the leak-down test.

But, anyway, since the heads checked out good, I now have to believe that something is wrong between the cylinder walls or the piston rings.

The shop said that I might have cracked a ring putting the pistons in, but I don't see any scores in the cylinder walls on any of the cylinders.

I'm going to take the pistons out of the "bad" cylinders to make sure they look okay. I just hope I don't have a cracked cylinder wall...

Vader

Don't overlook the valve timing and the head gaskets. Apparently, you valve and heads are sealing properly.

the following might explain your leakage results unless you had the rocker arms loose for the tests. If the rockers were left in paosition, there is a chance that one of the valves was being held in a slightly open position by the valve train. Usually, the exhaust valve will open BEFORE the piston reaches BDC of the stroke, so if you didn't test the leakage with the cylinders at TDC and the crank held in position, your results might be normal.

I know you removed the rockers on the #1 hole, but I can't recall how you did the rest of the test (if you mentioned the rockers or not). The #1 may have had a gasket problem or may have ring wear, excessive gaps, or severe taper.

Kingtal0n

yes i remember when i did my leak down test the air entering the combustion chamber was strong enough to force the piston down to BDC therefore opening my ehaust valve letting it think there was like 40% leakdown. dont overlook that! my cam open the valve 50* BBDC i beleive. thats alot of opening, almost .050 of lift at the point of BDC.

sancho

Well, now that you bring up the head gaskets... I will say that they weren't blown--that is to say, the metal rings surrounding the cylinders were in tact. Something I did notice when taking the head bolts out (specifically on the driver's side) was that at least one bolt came out with rusted threads (I know it didn't go in like that). Also, when I actually pulled the head off, part of the gasket was significantly soaked.

As to why the gasket was wet... It could have been either or both of two things:

1. Left-over coolant that leaked out of the head/block when I loosened the head (although I *did* drain the coolant before loosening the bolts). I did actually leave the head on the block overnight with the bolts out, so mabye it had time to soak.

2. The threads on the head bolts weren't properly sealed (hence the rusted bolt(s)), thus coolant slowly soaked the gasket when it was actually running.

I don't know that a "wet" gasket constitutes a "blown" gasket, or if the condition would even cause the problem that I've been having. It's just an observation that I had.

Regarding the rocker arms/timing issue: I did remove the rocker arms on #1 and re-tested, getting the same results. I didn't remove the rocker arms on #6, but after retesting it several times I decided that I would need to pull the engine regardless.

As far as the crank moving and altering the valve positions: the leak-down test was a two-man job, hence somebody was always looking at the crank/degree wheel for crank movement. I only had the crank jump on me once.