Incompetent valve job?
April 27th, 2013, 03:54 PM
#1
Registered User
Thread Starter
Join Date: Aug 2009
Posts: 333
Incompetent valve job?
Went out to put new valve springs on my A block 425. I decided to do just the springs and seals first, before replacing the cam and lifters, just to make sure I didn't need to send the heads in for machine work.
Naturally, I ran into something weird.
Working on the even bank first, 2-4-6 went just fine. Used compressed air to hold the valves in place while I swapped the springs. Bit of a pain getting the new seals on, but otherwise no issues.
Started #8, and immediately noticed that when I applied the air, there was noticeably more noise - as in air escaping from the cylinder. I disconnected the air, rotated the engine to make absolutely sure I was on the base circle of the cam, hooked up the air again, no change. Noticeable whoosh.
Loosened the rocker arm nuts, the noise went down to what it was like for the other cylinders, and the intake valve visibly closed more. Tightened the nuts, the valve opened slightly, whooshing started again.
My best guess is that at some point in the past, someone did a valve job and didn't ensure the stem height was correct. The seats were ground down, the valve sits deeper, the stem was not ground to match, and when the rocker is torqued down it opens the valve slightly.
Does this seem like a reasonable guess? I've been thinking of putting a straightedge across all eight valve stems to see if they are the same height. I don't have a good machined straightedge, but I would think an aluminum 90º measuring stick would be good enough for a quick check.
The car had been sitting for 3-4 days before I started, so I can't imagine the lifters are pumped up tight. From the normal short-lived clatter on startup, I'd say they bleed down pretty rapidly when the engine is shut down.
Any other thoughts? I hate to put a brand-new cam in without chasing this down, and I also hate the thought of yanking the heads and taking them in for a full workup if they don't actually need it...
Naturally, I ran into something weird.
Working on the even bank first, 2-4-6 went just fine. Used compressed air to hold the valves in place while I swapped the springs. Bit of a pain getting the new seals on, but otherwise no issues.
Started #8, and immediately noticed that when I applied the air, there was noticeably more noise - as in air escaping from the cylinder. I disconnected the air, rotated the engine to make absolutely sure I was on the base circle of the cam, hooked up the air again, no change. Noticeable whoosh.
Loosened the rocker arm nuts, the noise went down to what it was like for the other cylinders, and the intake valve visibly closed more. Tightened the nuts, the valve opened slightly, whooshing started again.
My best guess is that at some point in the past, someone did a valve job and didn't ensure the stem height was correct. The seats were ground down, the valve sits deeper, the stem was not ground to match, and when the rocker is torqued down it opens the valve slightly.
Does this seem like a reasonable guess? I've been thinking of putting a straightedge across all eight valve stems to see if they are the same height. I don't have a good machined straightedge, but I would think an aluminum 90º measuring stick would be good enough for a quick check.
The car had been sitting for 3-4 days before I started, so I can't imagine the lifters are pumped up tight. From the normal short-lived clatter on startup, I'd say they bleed down pretty rapidly when the engine is shut down.
Any other thoughts? I hate to put a brand-new cam in without chasing this down, and I also hate the thought of yanking the heads and taking them in for a full workup if they don't actually need it...
April 27th, 2013, 06:10 PM
#2
Registered User
Join Date: Nov 2012
Location: Deland Fla.
Posts: 535
Do you have factory rockers? If so it is critical to have the valve stem height correct and if not that is one of the symptoms also it could be a slightly bent valve in any case if you are installing a new cam kit just yank those heads and fix them right and you will be much happier! I have a saying when looking at all the extra work there is only ONE WAY TO DO IT and I HAVE TO DO IT RIGHT!! and one of my freinds always says NOTHING TO IT BUT TO DO IT!! Also before tearing it all down check the pushrods on a sheet of glass or a perfect flat surface to make sure the pushrod isnt bent slightly as I have had this happen also and that could save you from pulling the heads. Good luck
April 27th, 2013, 06:14 PM
#3
Registered User
Join Date: Nov 2012
Location: Deland Fla.
Posts: 535
One more thing if you have the factory rockers check the rocker stands on that cylinder to make sure it isnt cracked. I have used a small piece of shim material under one of my stands to correct the same problem. I used .010 shim stock from brass and shimmed it until it quit hissing. Hope this helps?
April 27th, 2013, 06:26 PM
#4
Registered User
Join Date: Dec 2011
Posts: 7,286
I say not likely. You have about 0.100-0.150" of travel available in the lifter to accommodate manufacturing tolerances. That's x1.6 for a measurement at the valve- 0.160" minimum. way over an eighth inch. That would be obvious w/o a straightedge.
Now, if slackening the rocker closes the valve then one of two things- you were confused, and you were NOT on the base circle of the cam, or yeah, there is something WAY out of whack and you need to verify all your valves' lifter preload is correct. I have thousands of 0.020" shim washers for under the rockers, if needed, will send as many as you want for the cost of postage.
Now, if slackening the rocker closes the valve then one of two things- you were confused, and you were NOT on the base circle of the cam, or yeah, there is something WAY out of whack and you need to verify all your valves' lifter preload is correct. I have thousands of 0.020" shim washers for under the rockers, if needed, will send as many as you want for the cost of postage.
April 27th, 2013, 06:27 PM
#5
Registered User
Thread Starter
Join Date: Aug 2009
Posts: 333
Thanks, Dave.
So far, it looks very much like an untouched factory engine. Just pulled the water pump, and it has a 384783 casting number on it, which appears to be a factory number. If anyone had pulled it apart, you'd think they would have changed the water pump...
I'll check the pushrods (good thought) when I put a straightedge on the valve stems. It's getting dark now, and sadly I can't fit a car in our teeny, tiny, crammed-full with kilns and other assorted crap garage, so I'm stuck in the driveway.
The valve guides seem very tight; no slop by hand at all (yeah, not much of a test, but I had a pair of Chevy heads once where after I removed a spring, I could grab the stem and rattle the valve in the guide like the little ball in a can of spray paint...). As far as a bent valve, it seems like that would prevent the valve from ever closing, instead of only when the rocker is torqued down, doesn't it?
So far, it looks very much like an untouched factory engine. Just pulled the water pump, and it has a 384783 casting number on it, which appears to be a factory number. If anyone had pulled it apart, you'd think they would have changed the water pump...
I'll check the pushrods (good thought) when I put a straightedge on the valve stems. It's getting dark now, and sadly I can't fit a car in our teeny, tiny, crammed-full with kilns and other assorted crap garage, so I'm stuck in the driveway.
The valve guides seem very tight; no slop by hand at all (yeah, not much of a test, but I had a pair of Chevy heads once where after I removed a spring, I could grab the stem and rattle the valve in the guide like the little ball in a can of spray paint...). As far as a bent valve, it seems like that would prevent the valve from ever closing, instead of only when the rocker is torqued down, doesn't it?
April 27th, 2013, 06:34 PM
#6
Registered User
Thread Starter
Join Date: Aug 2009
Posts: 333
Octania, the first thing I thought was not being on the base circle. I went so far as to set up a dial indicator and rotate the engine by hand to make certain I was clear of the lobe. Then I borrowed a cheap HF leak-down tester and after figuring out how to run the thing, verified that at TBC compression there was 15-20% more leakage with the rockers torqued than without. Definitely something hinky going on.
I'll do some research and figure out how to check lifter preload. Thanks for the shim offer, if it turns out I need some I'll holler.
I'll do some research and figure out how to check lifter preload. Thanks for the shim offer, if it turns out I need some I'll holler.
April 27th, 2013, 06:39 PM
#7
Registered User
Join Date: Nov 2012
Location: Deland Fla.
Posts: 535
Your right on the bent valve if it quits hissing when you back it completely off then your ok, Octania is right with the lifter preload and using shims is a way to compensate and fix the problem. Oldsmobiles have good valve guides they seem to last forever! Its good you checked the guides and it may not be precise but it works. I think your fine on the push rods I think its in the rocker stands and like I said I have done alot of shimming in the old days. Get a fresh start tommorow and you will have it fixed!!
April 27th, 2013, 10:05 PM
#8
Registered User
Thread Starter
Join Date: Aug 2009
Posts: 333
Found a method from Crane, I think. I'll try it out tomorrow, see what happens.
Fast and Easy Way to Check Hydraulic Lifter Preload when using Non-Adjustable Rocker Arms
With the cam, hydraulic lifters and pushrods in place, install your rocker arm assembly. Use the prescribed method in your repair manual and torque down all the valve train bolts in the proper sequence. Pick a cylinder that you are going to check. Hand rotate the engine in its normal direction of rotation until both valves are closed. You are on the compression cycle for that cylinder. (At this position the valve springs are at their least amount of tension making the job a little easier to do.)
Wait a few minutes, allowing the lifters to bleed down. Now, lay a rigid straightedge across the cylinder head, supporting it on the surface of the head where the valve cover gasket would go. Using a metal scribe and the straightedge, carefully scribe a line on both pushrods. Now carefully remove the torque from all valve train bolts, removing any pressure from the pushrods. Wait a few minutes for the pushrod seat in the hydraulic lifter to move back to the neutral position. Carefully scribe a new line on both pushrods.
Measure the distance between the two scribe marks, it represents the amount of lifter preload. If the lines are .020" to .060" apart you have proper lifter preload. If the lines are the same or less than .020" apart you have no or insufficient preload. If the lines are further apart than 060" you have excessive lifter preload. To bring your preload into tolerance, use one of the methods described in the next section if necessary, or call the Crane Tech Line for assistance (386/258-6174).
With the cam, hydraulic lifters and pushrods in place, install your rocker arm assembly. Use the prescribed method in your repair manual and torque down all the valve train bolts in the proper sequence. Pick a cylinder that you are going to check. Hand rotate the engine in its normal direction of rotation until both valves are closed. You are on the compression cycle for that cylinder. (At this position the valve springs are at their least amount of tension making the job a little easier to do.)
Wait a few minutes, allowing the lifters to bleed down. Now, lay a rigid straightedge across the cylinder head, supporting it on the surface of the head where the valve cover gasket would go. Using a metal scribe and the straightedge, carefully scribe a line on both pushrods. Now carefully remove the torque from all valve train bolts, removing any pressure from the pushrods. Wait a few minutes for the pushrod seat in the hydraulic lifter to move back to the neutral position. Carefully scribe a new line on both pushrods.
Measure the distance between the two scribe marks, it represents the amount of lifter preload. If the lines are .020" to .060" apart you have proper lifter preload. If the lines are the same or less than .020" apart you have no or insufficient preload. If the lines are further apart than 060" you have excessive lifter preload. To bring your preload into tolerance, use one of the methods described in the next section if necessary, or call the Crane Tech Line for assistance (386/258-6174).
April 28th, 2013, 04:44 PM
#9
Registered User
Join Date: Dec 2011
Posts: 7,286
Mathematically, it can be figured out pretty easily. The Olds rocker ratio is 1.6:1, and the pedestal bolts are 5/16-18 or 18 threads per inch or 1/18 inch per turn.
1/20" is 0.050" 1/18 should be about 2 out of 20 or 10% different
Excel agrees- 1/18" or 1 turn is about 0.056" at the pedestal bolt, which is about 0.089" at the pushrod. [yes, I know that the 1.6 ratio is not from there to there, but, mathematically, it turns out to be the same].
SO
1 turn at the pedestal bolt = 0.090 or so at the lifter. You can go a little more than that, but probably should not. Minimum acceptable is probably 1/4 turn at the pedestal bolt.
1/4 to 3/4 turn, I like to see.
one 0.020" shim changes the preload by about 1/2 turn at the pedestal bolt.
so...............................
get on the base circle for any pair of rockers.
slacken both pedestal bolts
wait for the lifter to expand- a few seconds should do it
I generally do this as part of the engine assembly, and I can SEE the lifter plunger move, but you can do it by feel also. Or watch for pushrod motion.
Do one rocker at a time.
Using fingertips on the tool for a better feel, and a small ratchet like say 1/4" drive, or just the extension, remove all the slack from a rocker until the lifter plunger is just ABOUT TO begin to move. You can back off and re-start over and over until you get the feel for it.
After this, a non ratcheting tool such as a small breaker bar will help you track position of the tool/ screw.
Now slowly tighten the pedestal screw until the pedestal bottoms out, noting exactly how far you have to go from "lifter plunger just starting to move" until "pedestal against head". You do not need to torque it to exactly 25 ft-lbs. Hand tight is fine.
If it's more than 3/4 turn, note how many shims to add. Should keep each pair within one 0.020" shim of alike, unless maybe you have the steel strap 3-pc pedestal pair set, then a little more discrepancy can be tolerated.
Do the adjacent mate valve alike, then install the needed shims and re-check for 1/4 to 3/4 turn at the pedestal bolt. Then do the rest of the valve pairs. "In firing order" is one way, but you can go any order you want if turning the engine is easy.
I have seen as many as 2-3 shims required, or on my 403, ASSembled by a dolt previous owner, there was "less than 0 turns" of lifter preload- they used fatter head gaskets, did not shave the heads, and ended up with slack in some of the rockers. This had to be fixed to run right, of course.
1/20" is 0.050" 1/18 should be about 2 out of 20 or 10% different
Excel agrees- 1/18" or 1 turn is about 0.056" at the pedestal bolt, which is about 0.089" at the pushrod. [yes, I know that the 1.6 ratio is not from there to there, but, mathematically, it turns out to be the same].
SO
1 turn at the pedestal bolt = 0.090 or so at the lifter. You can go a little more than that, but probably should not. Minimum acceptable is probably 1/4 turn at the pedestal bolt.
1/4 to 3/4 turn, I like to see.
one 0.020" shim changes the preload by about 1/2 turn at the pedestal bolt.
so...............................
get on the base circle for any pair of rockers.
slacken both pedestal bolts
wait for the lifter to expand- a few seconds should do it
I generally do this as part of the engine assembly, and I can SEE the lifter plunger move, but you can do it by feel also. Or watch for pushrod motion.
Do one rocker at a time.
Using fingertips on the tool for a better feel, and a small ratchet like say 1/4" drive, or just the extension, remove all the slack from a rocker until the lifter plunger is just ABOUT TO begin to move. You can back off and re-start over and over until you get the feel for it.
After this, a non ratcheting tool such as a small breaker bar will help you track position of the tool/ screw.
Now slowly tighten the pedestal screw until the pedestal bottoms out, noting exactly how far you have to go from "lifter plunger just starting to move" until "pedestal against head". You do not need to torque it to exactly 25 ft-lbs. Hand tight is fine.
If it's more than 3/4 turn, note how many shims to add. Should keep each pair within one 0.020" shim of alike, unless maybe you have the steel strap 3-pc pedestal pair set, then a little more discrepancy can be tolerated.
Do the adjacent mate valve alike, then install the needed shims and re-check for 1/4 to 3/4 turn at the pedestal bolt. Then do the rest of the valve pairs. "In firing order" is one way, but you can go any order you want if turning the engine is easy.
I have seen as many as 2-3 shims required, or on my 403, ASSembled by a dolt previous owner, there was "less than 0 turns" of lifter preload- they used fatter head gaskets, did not shave the heads, and ended up with slack in some of the rockers. This had to be fixed to run right, of course.
Last edited by Octania; April 28th, 2013 at 04:49 PM.
April 28th, 2013, 08:32 PM
#10
Registered User
Join Date: Nov 2009
Location: Red Oak, Texas
Posts: 2,999
Mathematically, it can be figured out pretty easily. The Olds rocker ratio is 1.6:1, and the pedestal bolts are 5/16-18 or 18 threads per inch or 1/18 inch per turn.
1/20" is 0.050" 1/18 should be about 2 out of 20 or 10% different
Excel agrees- 1/18" or 1 turn is about 0.056" at the pedestal bolt, which is about 0.089" at the pushrod.
1/20" is 0.050" 1/18 should be about 2 out of 20 or 10% different
Excel agrees- 1/18" or 1 turn is about 0.056" at the pedestal bolt, which is about 0.089" at the pushrod.
1 divided by 18 =.0555 or .056 thousands per turn.
for 5/16-24 bolts it is 1 divided by 24 = .0416 0r .042 thousands per turn.
April 29th, 2013, 05:42 PM
#13
Registered User
Thread Starter
Join Date: Aug 2009
Posts: 333
What I found yesterday was one stem (exhaust, I think - I'd have to go doublecheck to be certain) is higher than the others, but it's #4 cylinder. All the others have the same stem height, as best I can tell with a non-machined straightedge (aluminum carpenter's tee). With the other stems, I can fit at most a .003 feeler gauge between the straightedge and the stem with difficulty, then the #4 exhaust is maybe .010 higher than its neighbors. #8 cylinder has about .070 preload on the intake, about .065 on the exhaust. Other valves tested show preload in the high .050 range, just barely within the .020-.060 tolerance.
So I believe I'm looking at one of three scenarios: 1) the engine had the valves done at some time in the past, the stems were not shaved properly for valve stem height and the assembler did not ensure proper preload, or 2) the seats are getting hammered and have sunk slightly, or 3) a combination of both.
I talked to the machinist I've used in the past for minor stuff; owner/operator at the same location for 30 years or more, but he has not done A head Olds before. He is, however, meticulous with specifications, so I'll make sure he has good information to work with. I'll drop the heads off Saturday, and we'll see what he finds. He charges $90 per head for an ordinary 3-angle valve job, plus any necessary parts - which shouldn't amount to anything unless the seats are sunk, as the compression is good and I've already got springs, seals, etc. If the seats are not sunk, I don't even think I'll have hardened seats put in, as it won't be a daily driver, just a fair weather cruiser, and it's not a radical cam.
Then I can make sure the preload is correct with the new head gaskets (if it currently has a steel gasket the replacement Fel-Pro gasket will be thicker, which will change the preload, but I'm not sure in which direction...better or worse?) and hopefully I won't have to ever go through the process again.
Octania, I suspect that what I have for rockers is the three-piece design. Each valve has a stud, a rocker, a fulcrum, then a strap that runs over both assemblies, and a nut for each stud. I tried to take a picture, but either my wife's camera is squirrelly or it is just smarter than me. Whatever they are, I don't want to have to fiddle with it multiple times, so it's off to the machine shop I go. Depending on what happens with the heads and the gasket thickness, I might need some shims.
So I believe I'm looking at one of three scenarios: 1) the engine had the valves done at some time in the past, the stems were not shaved properly for valve stem height and the assembler did not ensure proper preload, or 2) the seats are getting hammered and have sunk slightly, or 3) a combination of both.
I talked to the machinist I've used in the past for minor stuff; owner/operator at the same location for 30 years or more, but he has not done A head Olds before. He is, however, meticulous with specifications, so I'll make sure he has good information to work with. I'll drop the heads off Saturday, and we'll see what he finds. He charges $90 per head for an ordinary 3-angle valve job, plus any necessary parts - which shouldn't amount to anything unless the seats are sunk, as the compression is good and I've already got springs, seals, etc. If the seats are not sunk, I don't even think I'll have hardened seats put in, as it won't be a daily driver, just a fair weather cruiser, and it's not a radical cam.
Then I can make sure the preload is correct with the new head gaskets (if it currently has a steel gasket the replacement Fel-Pro gasket will be thicker, which will change the preload, but I'm not sure in which direction...better or worse?) and hopefully I won't have to ever go through the process again.
Octania, I suspect that what I have for rockers is the three-piece design. Each valve has a stud, a rocker, a fulcrum, then a strap that runs over both assemblies, and a nut for each stud. I tried to take a picture, but either my wife's camera is squirrelly or it is just smarter than me. Whatever they are, I don't want to have to fiddle with it multiple times, so it's off to the machine shop I go. Depending on what happens with the heads and the gasket thickness, I might need some shims.
May 5th, 2013, 03:25 PM
#14
Registered User
Thread Starter
Join Date: Aug 2009
Posts: 333
Today's update with a question or two:
I pulled the heads, pan, and timing cover.
The valves did not look as if they had sunk in the seats, though I should find out tomorrow what the actual case is.
The head gaskets are factory steel shim gaskets, original nylon timing gear, absolutely no ring ridge. Pistons are low-compression dished (which I had assumed, as per factory 2-bbl specs). Everything looks very clean, except the bottom of the oil pan is somewhat gunky. Not horrible, just not perfect.
The timing chain is definitely worn, with a fair amount of slack.
I'm about as happy as I can be, considering I bought this car just because it was cheap and I wasn't going to be destroying an original classic.
Now a couple more questions:
Since today's gaskets are thicker (FelPro gaskets are .039 compressed compared to .017 for the factory steel gaskets), I'm considering having the heads milled to maintain or even gain a little compression. From what I've been able to research, every .006" milled from the head results in app. 1cc decrease in compression chamber volume.
The piston dish depth appears to be .220" (crappy tools, and I can't find the factory spec for the dish), the dish diameter appears to be 2.865" (ditto on crappy tools), and the deck height appears to be .020" (ditto again on tools, although I think I can make a cylinder strap to get better measurements with if need be). This gives me a dish volume of 23.24cc (actual volume measurements using liquid isn't really an option, since I'm not disassembling the block).
Does anyone know what the factory spec is for the low compression (rated at 9.0 to 1) dish?
I'm contemplating milling .060" from the heads. This would give me an effective decrease of .038" given the thicker gasket, and a compression ratio of 9.2 to 1, according to Summit's calculator. It would also decrease the lifter preload, wouldn't it? The pushrods and rockers are stock original, the heads are moving closer to the block, and there will be a fresh 3-angle valve job. The original preload was at or over the maximum .060, resulting in a valve or two never fully seating.
Would milling the heads .060" (resulting in total movement of .038" closer to the block) move the lifter preload back within tolerance? Move it too far in the right direction? Move it the wrong direction?
I had originally wanted to go with roller rockers, but decided the expense wasn't worth the gain. Would this setup require adjustable rockers?
Besides changing the intake manifold matching, would milling that much from the heads cause any other problems? I would have the shop mill .038" from the intake side of the heads to compensate for the manifold fit and use silicone for the front and back rails with a steel turkey pan gasket. I'm changing from an Offenhauser 360 to an Edelbrock Performer manifold, but the turkey tray worked fine with the Offy.
The valves did not look as if they had sunk in the seats, though I should find out tomorrow what the actual case is.
The head gaskets are factory steel shim gaskets, original nylon timing gear, absolutely no ring ridge. Pistons are low-compression dished (which I had assumed, as per factory 2-bbl specs). Everything looks very clean, except the bottom of the oil pan is somewhat gunky. Not horrible, just not perfect.
The timing chain is definitely worn, with a fair amount of slack.
I'm about as happy as I can be, considering I bought this car just because it was cheap and I wasn't going to be destroying an original classic.
Now a couple more questions:
Since today's gaskets are thicker (FelPro gaskets are .039 compressed compared to .017 for the factory steel gaskets), I'm considering having the heads milled to maintain or even gain a little compression. From what I've been able to research, every .006" milled from the head results in app. 1cc decrease in compression chamber volume.
The piston dish depth appears to be .220" (crappy tools, and I can't find the factory spec for the dish), the dish diameter appears to be 2.865" (ditto on crappy tools), and the deck height appears to be .020" (ditto again on tools, although I think I can make a cylinder strap to get better measurements with if need be). This gives me a dish volume of 23.24cc (actual volume measurements using liquid isn't really an option, since I'm not disassembling the block).
Does anyone know what the factory spec is for the low compression (rated at 9.0 to 1) dish?
I'm contemplating milling .060" from the heads. This would give me an effective decrease of .038" given the thicker gasket, and a compression ratio of 9.2 to 1, according to Summit's calculator. It would also decrease the lifter preload, wouldn't it? The pushrods and rockers are stock original, the heads are moving closer to the block, and there will be a fresh 3-angle valve job. The original preload was at or over the maximum .060, resulting in a valve or two never fully seating.
Would milling the heads .060" (resulting in total movement of .038" closer to the block) move the lifter preload back within tolerance? Move it too far in the right direction? Move it the wrong direction?
I had originally wanted to go with roller rockers, but decided the expense wasn't worth the gain. Would this setup require adjustable rockers?
Besides changing the intake manifold matching, would milling that much from the heads cause any other problems? I would have the shop mill .038" from the intake side of the heads to compensate for the manifold fit and use silicone for the front and back rails with a steel turkey pan gasket. I'm changing from an Offenhauser 360 to an Edelbrock Performer manifold, but the turkey tray worked fine with the Offy.
Thread
Thread Starter
Forum
Replies
Last Post
