testing for poor rings

2533 Views 16 Replies 8 Participants Last post by Alfapaulic, Oct 10, 2007

Andrew

Discussion Starter · Oct 4, 2007

I did a compression test on a friend's Giulietta Sprint Veloce yesterday and got 20, 100, 120, 100. Not great. I can see a badly burned exhaust valve in the 20# cylinder, so that's clearly got to be fixed.

Hoping to avoid a full rebuild, I wonder if we can do this: pull the head off and do a valve job. Then bolt the head back on with the cams not installed (and not even install the chain), and do another compression test. Knowing the valves at that point are good, that should quickly tell us how the rings are. I don't expect the rings to be perfect, but if I can get all readings over 120, I think the engine can last a few more years before a full expensive rebuild.

Any reason this wouldn't work?

Andrew

1 - 17 of 17 Posts

Tifosi

· #2 · Oct 4, 2007

It should work, but if after all that you find the rings are bad, it'll have to come back down again anyway. (how many head gaskets do you want to buy?)

Before you tear it down, remove the cams, do the compression check again and record the #'s, then add about a tablespoon of motor oil to each cylinder and repeat the test.

If compression magically goes up, the rings are well on thier way out anyway and all the head work in the world won't bring them back.

Andrew

Discussion Starter · #3 · Oct 4, 2007

I figure the labor involved to drop the head back on (it has to come off either way) is not signficant just to torque it down with a cheapie head gasket for test purposes. Not having to set the valves, time the cams, install everything, means I see this as only a 1-2 hour project once the head comes back from the machine shop. So it may be lost time, true, but not much.

The chunk missing from the #1 exhaust valve is so huge (it's like 1/5 of the valve) I think there's no chance a teaspoon of oil will make any difference on the compression reading. I can't believe it made any compression at all. But it can't hurt to try that before we take it apart.

Andrew

Tifosi

· #4 · Oct 4, 2007

Oh, huge chunk then.

Yeah, oil isn't likely to make a difference then

Still, there might be concern over whether that actually burned away (intake leak is the usual culprit and it generally burns away a bit of piston crown too) or that it broke off and you're seeing the burned edges where the leak it caused when it broke superheated. (where did the chunk go, and did it damage the piston or liner when it went?)

crashmctavish

· #5 · Oct 4, 2007

Fix the head & valves use the old gasket & retest, if its ok fit a new gasket & keep driving. Regards

Richard2

· #6 · Oct 4, 2007

I agree with crash on the head gasket issue. Compression readings on a cold engine will be lower than normal. I don't know if 120 is a realistic low number on a cold engine or not. Certainly you'll not see 190+.

Mine was running 125-135 on a hot engine before the Berlinas to Bovine trip. I haven't checked it since the return.

Andrew

Discussion Starter · #7 · Oct 5, 2007

The missing chunk is so big I do worry that it was some kind of catastrophic event, rather than just burning. Again, hard to really know what's going on until the head comes off, which has to happen whether we do just a valve job or a total rebuild.

Andrew

John M

· #8 · Oct 5, 2007

The oil will seal the rings. So if pressure comes up with oil on any cyl significantly then go ahead and rebuild the entire thing. IF you find an impact site on top of the "burned valve" piston/cyl....go ahead with a full rebuild. Something that size rattling around in the chamber would most certainly damage the journal bearing and perhaps bend the rod.

Best Regards,
John M

crashmctavish

· #9 · Oct 5, 2007

If a valve has hit the piston gently deburr the scars on the crown of the piston in situ, then do a dye penetrant crack test. Pistons will cope with quite a lot of grief. If a rod was bent you will see that piston lower than the others on TDC. The best form of measurement is use your eyes it will tell you pretty much what happened. The broken valve should be in the exhaust pipe.

Andrew

Discussion Starter · #10 · Oct 5, 2007

This car has been running this way for at least 10 years and perhaps 15,000 miles. It has good oil pressure and no scary noises, so I think if there was other major damage we'd know by now.

My guess is the valve burned/broke a long time ago (without damaging anything major), and it's been getting progressively worse as the valve continues to erode away, and probably the valve guide and seal too. The owner these 10 years says it has never been particularly fast, but it also has never had an overnight change of performance either.

Andrew

John M

· #11 · Oct 5, 2007

crashmctavish said:
If a valve has hit the piston gently deburr the scars on the crown of the piston in situ, then do a dye penetrant crack test. Pistons will cope with quite a lot of grief. If a rod was bent you will see that piston lower than the others on TDC. The best form of measurement is use your eyes it will tell you pretty much what happened. The broken valve should be in the exhaust pipe.

I will agree with you that you can polish out the piston, magnaflux, etc. But you will not be able to discern whether or not you have damaged the rod journal bearing. The rods are strong as well. The bearings are not. And it may not necessarily show any difference in oil pressure or noise until its been driven for awhile post damage. IF I found that the piston was scared on the crown and the top of the combustion chamber too...at minimum before even considering trying to get away with not rebuilding the bottom end, I would pull the pan or pans as necessary and inspect that rod bearing. I say this from bearing hammering experience...unfortunately.

Only takes a piece of metal about the size of a finger nail clipping hammered between the piston and the head to fubar a bearing. Of course the tale tale sign will be imprinted on the crown of the piston and the top of the combustion chamber. If no sign of that...then you are good to go. IF not...pull it apart.

Best Regards,
John M

See less See more

Andrew

Discussion Starter · #12 · Oct 5, 2007

Agreed. We're not trying to be utter cheapskates. It'll be up to the owner to decide once/if he and I pull off the head. However, he may just decide to go for broke and give it to one of the local Alfa shops to rebuild. Dunno yet.

Andrew

ToonRboy

· #13 · Oct 5, 2007

Sounds reasonable - maybe, but I wonder if the engine will turn over with all the valves closed. I suspect she may not breath enough through the spark plug holes to get proper readings Andrew. It'll certainly be more difficult to turn it over once the head is torqued back on + the cylinder you're going to check will have the gauge in that hole - where is it gonna get air from?

I'm curious if that'll work, but I would also try to determine why that exhaust valve went - Bad injector, improper lash (too tight) etc...

Keep us posted.

Andrew

Discussion Starter · #14 · Oct 5, 2007

Good point; I didn't think about that. I had presumed to leave the plugs out so it would turn over easily, but you're right that it won't be able to draw any air. Thanks for helping me think this through.

This is a dual-Weber carbed car, the engine's history of which we don't know. It came from Montana, to Seattle, to Calif 10 years ago. It does have a leaky carb mount on #4, but not on #1, the low cylinder.

Andrew

Kjell

· #15 · Oct 5, 2007

You might also ask owner what the oil consumption rate was.
anything worse than say 400 miles per quart might suggest poor
ring seal (yes I know can also be valve guide seals). If it were mine,
I'd just pull the thing and rebuild (assuming the rest of the car is in good
shape of course.....). I've seen this movie, doing a head job and then ending
up doing a rebuild soon there after, a bit of a PIA.

crashmctavish

· #16 · Oct 5, 2007

leaky carb mount on 4 will make it run weak ping & possibly take the valve out. I assume by use of the trade name magnaflux you mean the dye not mpi which obviously would not work with ally. Regards to all I think we have done this one to death, time to get hands dirty.

Alfapaulic

· #17 · Oct 10, 2007

A compression test won't work with no valve action. Just overhaul the whole thing and be done with it.

1 - 17 of 17 Posts

This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.