QuestionQUESTION: Hello,
thinking I had a valve stem seal problem (light blue exhaust smoke)I pulled the head, and installed some new, special (perhaps snake oil?) seals. While I was there I hand lapped the valve seats as well; the intakes looked essentially new, and the exhausts had some wear but did not appear to be burnt, etc. The head and block measured flat with straight edge. I made sure the head and block were clean and dry, put on a brand new Payen head gasket correctly, and torqued the nuts in the proper sequence up to the specified 50 ft.lbs. The engine started right away, but there was a huge coolant leak between the head and block. What did I do wrong, or what can you suggest?
Also, please comment on these compression test results. These were the numbers BEFORE I pulled the head and did the above work: dry 135/132/135/125 wet 155/140/150/150
Thank you in advance! :-)
Dan
ANSWER: Hi Dan,
It sounds like you used the correct procedure down the line. You said coolant leak, but didn't say where. Your wet and dry tests indicate you are on the edge but the rings should be usable (15%+).
Payen is a good brand of gasket so I would have thought there would not be a problem with the head sealing as you used the procedure I always used in the dealerships I worked in.
Is the leak between #2 And #3 cylinders on the spark plug side? and did the leak only show up after the engine warmed up?
Let me know,
As for the valve seals, the MGs didn't have a stem seal per-say. They just had an "O" ring at the bottom of the keepers to stop puddled oil from running down between the keeper halves.
Howard
---------- FOLLOW-UP ----------
QUESTION: Hi Howard,
thank you for your prompt response! Here is some additional data:
Initially, the coolant leaked from the back of the head/block, (firewall side) on the right hand side. Unfortunately of course one cannot observe the left hand side head/block joint, being obscured by carbs/header. After the engine got warm the leak did migrate down the head/block joint all the way to the # 1 spark plug.
Looking at the oil sample on the dipstick I was unable to see any frothing as you would expect from water getting into the oil. I just saw dirty clear oil.
The new seals I installed were called "viton" seals, purchased from whom I believe to be an earnest eBay seller. These things look like minature tea cups with a blue material and a circular spring. You invert these tea cups and tap them gently onto the valve guide on the top of the head.
Today I rechecked the compression test. (1-4) dry 130/135/125/150 wet 132/150/135/150 The numbers do not seem to agree with my previous findings; one wonders if my technique is variable... :-)
One bit of information I need to include: When I initially pulled the head I was aghast to find that valve #7 (intake for #4 cylinder) had one valve keeper broken, and the other one dislodged such that the valve spring was not as compressed as its neighbors. I thought that was the root of the oil fouled spark plug, but as it turns out it was not. After running the engine and getting the coolant leak I pulled all the plugs and all are oil wet.
Interestingly today I fashioned a crude cylinder pressure tester--unfortunately it leaks at the spark plug adaptor--so I wasn't really able to assess much, but what I did notice is when I applied pressure to cylinder #2 I could notice oil from the compression test leaking out by the #2 and #3 spark plugs. I have plans to improve this device, simply for my own edification.
In the Moss Motor catalog in the engine section the idea is mentioned of the cylinder head studs tending to pull up threads out of the block. This probably could explain the leak I have. Does this idea have merit? When I looked at mine I didn't think much of it, but maybe I do have threads sticking up; I am planning on re-pulling the head anyway...
Thank you in advance for any and all thoughts! You can be sure I will pick on you further as I uncover more issues with this car!
Best,
Dan
ANSWER: Dan, in the dealerships we always considered a wet/dry compression test a good 95% accurate test of rings. There is a 100% test of valve leakage and head gasket leakage. Most tool trucks (Snap-on, Cornwell and the other tool trucks that visit shops) have an adapter to screw into the spark plug hole to connect the line air pressure hose to. You just put that cylinder to be tested on exactly TDC and put the car in 4th gear and brake on hard. Then connect the shop air hose to it. Open the throttle and listen in the carb for a hissing sound and do the exhaust by listening at the tail pipe. No hiss at all is acceptable, it is a sign of a leaking valve.
Then remove the radiator filler cap and watch for any sign of the coolant level raising at all or any bubbles. This test will pick up the smallest head gasket leak. Even one that does not even make that engine over heat.
Howard
---------- FOLLOW-UP ----------
QUESTION: Hi Howard,
some additional information:
I was able to fashion a cheap cylinder pressure tool using my air compressor and one of those high pressure tubes with the spark plug adaptor, that you would use to pressurize a cylinder when you want to remove the valve springs with the cylinder head still on. This set up can only manage about 80 psi, but I figured it was better than nothing...
1. The number 1 exhaust valve leaks.
2. The number 3 cylinder when pressurized will bubble coolant out of the radiator, and also at the head/block joint by the number 3 spark plug.
I just wish I had an idea as to why my new cylinder head gasket installation failed...
Regards,
Dan
AnswerGood Question Dan, what did you torque the head to and what sequence did you use? Did you inspect closely for a crack in the head? If it showed up leaking at 80 PSI you have a bad leak. When you get it off recheck that area (block & head) with your straight edge too. If you can't see any thing you may have to have a machine shop magnaflux the head. Also look closely at the block too.
Howard
