Interesting Problem

[RickSE]

Here's an interesting problem. I was changing my valve springs this weekend using compressed air in the cylinders, got to #6 cylinder and a ton of air was blowing out the exhaust port. I tried a couple of things like opening and closing the valve and adding oil to the cylinder, nothing worked. When I added oil to the back side of the valve I noticed the air seems to be leaking around the entire circumference of the valve. Has anyone ever experienced this before? My feelings are that it's either deposits on the valve or seat, holding the valve open; or a serious problem like a cracked seat or bad grind. The good news is that the engine is still under warranty!! http://www.donzi.net/ubb/biggrin.gif I'm just glad I found this now if it ends up being serious.

[This message has been edited by RickS (edited 07-10-2000).]

[Forrest]

Yea, in my case I needed a valve job. Try lightly wacking the top of the valve stem with a soft mallet few time with the spring and retainer still installed, then check the leakdown. If it still leaks, pull the head off, remove the valve and take a look. Worn valve guides can make the valve not seat perfectly everytime too. Good luck.

------------------
Forrest

[Jamesbon]

Try this one if the valve isn't bent. Put a little rubbing or grinding compound on the valve through the exhaust port, attach a drill to the top of the valve, spin the drill/valve in it's seat while "pulling" on the drill to apply pressure between the valve and valve seat. You're basically lapping the valve and seat. If you're dealing with minor imperfections, this may seal up that valve, if not..........

[PaulO]

Just be careful getting abrasive lapping compound into the cylinder. Don't want to trade valve problems for ring problems.
PaulO

[RickSE]

I'll try what you said Forrest, I'm all for trying the simple things first. Lapping the valve will probably be the next step if I don't find any obvious problems. This engine by the way only has 40 hours on it.

If this engine takes a while to fix I'll probably pull the 400 SBC out of my Camaro and use it for the summer. Does anyone know if there is a limit for camshaft overlap that will cause revision? This engine has a Comp Cam 292 deg. duration w/.501 lift, I need to look at the cam card for the overlap though.

[AVickers]

Oil in the cylinder won't help a valve -- it'll temporarily seal off a ring problem. If, during a compression test, pressure goes up w/ oil in the cylinder, then you know you have a problem w/ the rings. If the pressure is still low, then it's valves (and maybe rings in addition). That's why it's such a good diagnostic -- it'll mask ring problems, but valves still leak.

Which is what you got going on...warped or bent something or worn seat, or face, or both.

Or worse, an inserted seat is loose or leaking. I've been told not to use inserted seats on raw water cooled engines 'cause with the radical temperature cycles, they'll eventually come out -- even if staked in well (I don't remember who told me that, but it made sense at the time...).

The only other thing it could be is that there's a bit of gunk in the guide holding the valve open a bit, but a smack or two on the end of the valve to let the springs snap it back up should eliminate that as a problem.

Good luck!


[This message has been edited by avickers (edited 07-10-2000).]

[AVickers]

With a new engine, you probably have a valve that's not properly seated. Are you getting any compression on that hole at all? Doing nothing may be the way to go -- run the engine for a while without loading it up too much and the valve may lap in in its own time.

If you have to do something short of taking her down, I agree that you should try valve grinding paste first... Just a little bit of the fine variety applied w/ a Q-tip. When you're done, you can use a clean Q-tip to remove a lot of it. The rest will blow out rather quickly when you fire the engine up.

In fact, a drill motor may be overkill...and it might sling the paste around in the cylinder. Do you have an old woodworking brace? You could grip the end of the valve with it and turn the valve more slowly than with an electric drill motor.

The other option would be to use a piece of 5/16" fuel line slipped over the end of the valve and held in place w/ a helical clamp. You could then rotate the valve like you do with a lapping tool -- only difference is that you'd be pulling instead of pushing to lap the valve in.

As I said earlier, the conservative option would be to put it all back together and hope the valve laps in on its own before it burns. It is more likely to burn if you really load the engine up. This 'cause if it's not making good contact with the seat, it can't cool off...and valves in loaded engines really get too hot to dissipate all their heat through the guides alone.

In any case, if you do anything, I'd be sure to have the piston mostly up in the cylinder so if you drop the valve, it won't fall all the way out of the guide.

[rayjay]

Rick,

You did a basic "leakdown" test, and found a problem. Having done the lapping compound down the port trick when racing, I would advise you to pull the head and get a good look see at what is wrong. Yes, it could just be some carbon build up, but 40 hours would seem too little for that unless you're just idling along for hours (which doesn't sound like you). The seat could be coming apart, or the valve margin could be too narrow and breaking down or not have been hardened properly, or the valve stem could be starting to crack by its head. All of which can cause some very expensive problems later. rj

[RickSE]

I just finished talking to mechanic at the dealership, unfortunately he's 350 miles away. Anyway, he feels could be one of two things,
1) Valve was to tight in guide and burned.
2) Valve seat may be or have been loose.

He wants me to do a leak down test and then remove the head and inspect the valve and seat. Also check for bent pushrod.

I did do a compression test a while back, but I messed up and forgot to open up the throttle blades. All cylinders were within 5 psi when I did the test, 110-105 psi at 7000ft. elev.

I agree with Rayjay, in that I do not want to try and lap the valve with the head on. I'll take the head off after I do the leak down test.

[This message has been edited by RickS (edited 07-10-2000).]

[This message has been edited by RickS (edited 07-10-2000).]

[RickSE]

Anybody know how to build a leakdown tester? Nobody has one here in town. I would think two gages and a valve, one gage to watch cylinder pressure and one for line pressure. Use the valve to shut off the air and record how fast the cylinder pressure drops.

I found a picture of one on the net, think I know how to do it now.

[This message has been edited by RickS (edited 07-10-2000).]

[PaulO]

RickS,
As far as my performance marine small block chevy book says, the magic number for overlap is 12 degrees.
PaulO

[oldLenny]

as for a leakdown tester, here is an option that I think would work well. In my cabinet shop we have one way valves before the pneumatic machinery so that we can drain lines/modify air feeds etc and not lose pressure to clamping systems and air rams on various equipment. The guages are on the trapped side of the air lock. When the lines are drained the air pressure is still as it was prior to this and there is no leakdown. It will hold for days like this. If you were to do the reverse of this, that is a spark plug fitting and hose (we'll call this the compressor in the shop in this case) feeding into a "T" fitting (3/8" npt) with a 0-150psi air pressure guage, large dial,then to the one way gate valve followed by another large dial 0-150psi guage. The first guage along the "air compressor line" will be to monitor/time leakdown and the second one on the other side of the one way valve will keep record of the maximum inlet pressure that was recorded without releasing any for a very long time. So long as the rubber seal at the plug is well seated. When you are happy with the results you can depress the "button" on the gate valve and release the pressure stuck between the gate valve and the end pressure guage. I buy Lincoln or Parker Johnston air parts here and they have performed flawlessly for years. I would guess the guages are 20$ each US, the valve about 20$ and you would have to modify the sparkplug hose connection. I would get about 6 feet of hose swauged onto a barbed and threaded transition, and then do your other connections. This way you can get away from the engine and turn over the motor and monitor the progress. We buy all
of our components at a specailty hose and fitting company that caters to hydralic hoses and aircraft lines etc for the heavy equipment sector. Maybe things are similar in your neck of the woods.good luck.

[RickSE]

Found a good site for leakdown testing, http://www.briggs-racing.com/LEAKD.HTM . It's done differently than I thought, actually look at pressure drop with air on to determine leakdown rate. I've already put together the parts and will try it tonight.

[AVickers]

One thing w/ leakdown testing is that you need to make sure your piston is at TDC compression stroke, or you might turn the engine. On a boat this won't be so bad 'cause there isn't a fan to hang on stuff, but something still might get caught up in a belt or pully.

I have a leakdown tester (admittedly primitive) that screws into the sparkplug hole. It has an inlet side that comes from a two-stage regulator that I pre-set to 100psi. It has an outlet side that is attached to a 125psi gauge. Both hoses attach to a yoke that goes through the plug into the cylinder. I install the plug and turn on the air. Knowing that I have 100psi going in, I read the outlet side pressure. If it's 95lbs, then I have 5% leakage in the cylinder. If it's 50lbs, I have a problem. If it's rings, I can hear the air through the oil filler cap. If it's blowing by a valve, I can hear it through the intake or exhaust port.

I've been told a 10% leakdown is at the outer limits of acceptable for a standard engine with a 4" or larger bore. Race engines will leak something like 2% to 4%. A fresh automotive engine will leak something like 5%.

Here's a URL on a tester kit... http://wiwaxia.cc.swarthmore.edu/tech/wrench/leakdown.html

[This message has been edited by avickers (edited 07-10-2000).]

[Forrest]

Lenny, Victoria, BC! Man, that and The San Jaun Islands are my favorite places to visit. Do you see many Classic Donzis up there? Last time I visited Victoria, I watched a 40-some-odd-foot Scarab get pitched around like a cork in 8 to 12 foot seas in the Strait of Jaun de Fuca while riding the ferry over from Port Angeles, WA. Oh boy! No light two-foot chop out there! I'm sure there must be *much* lighter seas elsewhere around there. In fact, it was smooth as glass the last time I went on a whale watch out of Friday Harbor north towards Saturna Island. That water is cold, deep, and clear. A very beautiful part of the world - A big change in enviornment for this Mid-Atlantic/North Florida boy. I think that Friday Harbor would be a fantastic place for a Donzi get together.

------------------
Forrest

[rayjay]

Rick,
I'm glad you will take the head off. A leak down tester is a great tool to have, and can help you locate problems so you will know what you need to do.

When you use it, remember it is a measuring tool that gives readings that are really only comparable to itself, just like a dyno. Good units should all give similar readings, but can/will vary from each other by 1 - 2%. The readings on all your cylinders should be close to one another, idealy within a 0 - 2% range from low to high, with the lower your leak down rate the closer your readings should be. Try to test the rings a little before or after TDC, and as close to the middle of the stroke as possible. These are the two critical or problem areas for ring sealing and bore wear. The engine should not turn over especially if you leave the spark plugs in on the cylinders you are not testing. If it does you probably are using much too high an air pressure setting or the engine is so "loose" all the clearances are way out of wack. Good luck, and let me know if you need any help using the leak down tester or interpreting the findings.

As a "general" rule, bigger cams (longer duration) need bigger lobe offset, like 112 to 114 degrees, instead of the more common 110, to make any bottom and mid range torgue. Of course with saying that I am sure that someone can come up with a combination that will break that "general" rule, and big blocks can go "bigger" in their cam and still make a lot of bottom end power.

Makes me wish I could afford your 18 footer!! rj

[This message has been edited by rayjay (edited 07-10-2000).]

[RickSE]

I made up a leakdown tester and checked the engine last night.

Cylinders #3,5,7,2 & 4 read 98-99 psi. w/100 psi. supply air, therefor 1-2% leakdown.

Cylinders #6,8 & 1 read 95-96 psi. @ same 100 psi. supply air, 4-5% leakdown.

To me this is good news that there is no serious problem but there is definitely something going on. Cylinders #6 & 8 have always seemed to run rich and I always have more soot on that side of the boat. More than likely I will remove the heads and lap the valves, it can't hurt.