Fixing poor leakdown, Cyl 1 leaks at 50%, rest are cool. Options

[Jake Raby]

Use an open flame around the sealing surface to check for leaks... When the flame flickers you have a leak....

You said this was a second hand kit of mine, who did you buy iit from and what serial number was the kit???

[VaccaRabite]

I got the engine from Evan Thompson. I havn't got a clue what the SN is.

Zach

[VaccaRabite]

I pulled the heads, and then made some spacers out of some 1 inch square tube steel stock i had laying around, and used them to re-tighten the cylinders down. I did not tighten them to spec, as I was concerned about breaking a cooling fin, but I got them to where the cylinders could not move.

I put a level with a good straight edge across the cylinder bank and looked for gap. If the level was square on the #2 cylinder, then there was a bit of gap on the #1 cylinder. This was somewhat expected, all things considering.

Using feeler gauges, I measured the gap. Where the bubbles were forming (#1 cylinder, facing the #2 cylinder) there was a .007 inch gap. On the opposite edge (facing the flywheel) was a .005 inch gap. In my leakdown-bubble tests, in the spot with the .005 gap showed no leak (no bubbles).

I don't think I can manipulate my calipers accurately enough to measure the lip of the inside of the #1 cylinder.

It seems pretty clear to me that the cylinder is either sitting slightly cocked, or is not squared off at the top. I'll have to measure that tomorrow.

Zach

[VaccaRabite]

Okay, so I measured deck height today with Josh's help.

#1 was the same as #2. I don't have to get my case decked!

Its a problem with how the heads were cut. The heads need to be recut, taking as little material off as possible. I took off the tin and all the other bits today, the heads will come off tomorrow after I do some baby shopping with the missus.

Zach

[orange914]

Okay, so I measured deck height today with Josh's help.

#1 was the same as #2. I don't have to get my case decked!

Its a problem with how the heads were cut. The heads need to be recut, taking as little material off as possible. Zach

how many cc's are they now? are you running carbs or f.i.?

mike

[degreeoff]

Just dont forget to swap the heads and check to see if the difference is in the fact that the heads was swapped ?? as there was a CHT hole where there shouldnt have been one. It'll take 10 min to do as an xtra precaution and settle the matter once and for all.

Josh

[Borderline]

I pulled the heads, and then made some spacers out of some 1 inch square tube steel stock i had laying around, and used them to re-tighten the cylinders down. I did not tighten them to spec, as I was concerned about breaking a cooling fin, but I got them to where the cylinders could not move.

I put a level with a good straight edge across the cylinder bank and looked for gap. If the level was square on the #2 cylinder, then there was a bit of gap on the #1 cylinder. This was somewhat expected, all things considering.

Using feeler gauges, I measured the gap. Where the bubbles were forming (#1 cylinder, facing the #2 cylinder) there was a .007 inch gap. On the opposite edge (facing the flywheel) was a .005 inch gap. In my leakdown-bubble tests, in the spot with the .005 gap showed no leak (no bubbles).

I don't think I can manipulate my calipers accurately enough to measure the lip of the inside of the #1 cylinder.

It seems pretty clear to me that the cylinder is either sitting slightly cocked, or is not squared off at the top. I'll have to measure that tomorrow.

Zach

Okay, so I measured deck height today with Josh's help.

#1 was the same as #2. I don't have to get my case decked!

Its a problem with how the heads were cut. The heads need to be recut, taking as little material off as possible. Zach

OK, with the risk of being chastised by the Cap'n, I'll stick in my $.02:
How did you come to the conclusion that the head needs machining? You found the problem: the top surfaces of the two cylinders are not even. You found gaps of .007 and .005. Now you have to find out what is causing that variation. Just because the deck hts are the same, doesn't mean the problem is in the head. The top surfaces of the cylinders must be even. I would check both cylinder lengths and check the top ends for flatness. You may need to have the case decked. How did you measure the deck hts? If you want things to be right, you've got to measure everything until you find the source of the problem. The person that assembled the engine may have shortened one cylinder because the deck hts weren't even, not thinking about the heads sealing. You're on the right track, just keep measuring..

Just trying to help.

[Dr Evil]

I agree that you need to have the cylinder lengths checked. It sounds like the one may have an uneven head surface. If this is the case you can have that fixed and then place a gasket under the cylinder to even up the heights. You should also have the head surface check, of course. It sounds like this is what you are going to be doing. IIRC, you checked the registers using a straight edge and going at it from various angles.

[VaccaRabite]

Bzzt. I'm wrong again.

Now I think that the heads are fine...

I swapped the heads (put 3&4 bank on the 1&2 side, and vice versa) and re-did the leak down. So the #1 cylinder head was now on the #4 cylinder, and the #4 cylinder head was now on the #1 cylinder.

I re-did leak down on Cylinder 1&2&4. I skipped doing leakdown on three.

Cylinder 1 (Head 4) still leaked at 55%.
Cylinder 2 (head 3) was cool at 7%
Cylinder 4 (head 1 - which I thought was the leaky one) was also cool at 7%

I measured the #1 cylinder from behind the shim to the mating lip, but was not able to get a measurement I was able to replicate from time to time.

But I now either need to go back to either that the cylinder needs shimmed up about .005 inches, or the case needs decked.
*edit* actually, I need to get the cylinder measured by someone who can replicate the measurement. Mine seemed to vary too much for me to put any stock in it. */edit*

I had negated my previous theory as Josh had suggested that they way I went about it was inaccurate. But, I am going to try my spacer before I deck the case. Now I just need to find a .005 inch cylinder spacer.

I also talked to Evan yesterday, and checked out a lead he had remembered, but it, much like communism, was a red herring. Would have been sweet if it worked though.

Zach

[Jake Raby]

You cannot shim just one cylinder! Both cylinders must have the same deck surface.

Pull the cylinders and put a piece of 1,000 grit sand paper on a perfectly flat piece of glass.. Use some Dykem machinist dye on top of the cylinder to use as a control..

Make a pass over the 1,000 grit and see if any portion of the seating surface appears to be even or erratic.. You may have a warped cylinder seating surface..

My bet is still a drooped case register...

[VaccaRabite]

I'm going to try this - not now but tomorrow morning. I have a lot of 1000 grit paper from painting.

Wet sand or dry? Or does it not matter? Will a sharpie marker work in-place of machineist dye?

Zach

[Jake Raby]

Dry.. The Sharpie will work if you apply it heavily.. Let it dry for an hour before the process.

[scotty b]

Zach a shim isn't the answer. Remember when we pulled the head and saw the leakage it was only in that small area. Shimming wouldn't correct a tilt. Do as Jake says. Since the head swap yielded the same results you have a register or jug issue.

[davep]

I put a level with a good straight edge across the cylinder bank and looked for gap. If the level was square on the #2 cylinder, then there was a bit of gap on the #1 cylinder.

from this statement I'd say that either cylinder 2 is too tall or cylinder 1 is too short. If the gap is not even from one side to the other of cylinder 1, then one of the two is not sitting square (register droop) or the cylinders have unequal heights, not square, or some combination. I'd pull each cylinder, place on a flat (like glass) and use a height guage to ensure (a) equal height while spinning the cylinder, and ((IMG:style_emoticons/default/cool.gif) equal height for each cylinder. Since the bottom seating surface is not the bottom of the cylinder, you have to invert the cylinder on the glass and measure the height of (a) & ((IMG:style_emoticons/default/cool.gif) to the seating surface. If there is any variation, then do as Jake suggests. If there is no variation, then suspect a drooped case register or two.

[Jake Raby]

Are these Chinese cylinders??

[Brando]

Did you ever check your piston rings?

I've seen destroyed rings yield the same type of results.

I agree it may be a bad cylinder or bad case register as well.

[scotty b]

Did you ever check your piston rings?

I've seen destroyed rings yield the same type of results.

I agree it may be a bad cylinder or bad case register as well.

Doubtful. The inside of the barrels looked fine with no signs of scoring. As to what Dave mentioned I believe that to be the problem. The blowby inside the head was confined to a smal area on the inside of #1. Area was about 1/4" wide. Were the whle cylinder off I would have expected to see blow by in several areas around the chamber (IMG:style_emoticons/default/confused24.gif)

[VaccaRabite]

My rings are golden. As on good, not gold colored. :-)
'
Are the Chinese Cylinders... Dunno, let me go see if they say anything.
Not Chinese. Cyl #1 is stamped: "039 101 ?01C 18 15 17 VW" The "?" is either a 3, 8 or 9, I can't tell.

Pistons are Keith Blacks

So, today I tried the 1000grit sandpaper on glass. There were two very shallow depressions where the ink stayed, but neither of them were where the leak was. However, I polished the cylinder until it was level, and put the head back down. 70% leak, worse not better. But that made sence as I had just removed a little metal.

I measured the cylinder and got 6.008 inches. I measured several times, and was able to replicate the measurement at different points. Good, I think I have it level now.

I pulled the #2 cylinder off, and measured it. I noticed right off that it had 2 spacers to the #1 cylinders single spacer. I measured it at 6.020 inches. One of the spacers was .010 inches, so I removed it and remeasured. As expected, the cylinder (#2) was now at 6.010 inches.

I was going to put it all back on the block and re-leak it, but my wife let me know that I should consider heading in for dinner. We had cancelled our NYE plans for the evening, so after dinner I'll head back out and put everything back together.

I guess I am going to need to re-measure deck height since I polished one cylinder, and took a base shim off another.

Or, I suppose I could buy a .010 shim to bring the shorter cylinder up, which is what I probably need to do.

Right now I think that I should be within .002 inches of one to the other, which should even out the leak down, right?

Jake, would you like me to continue posting this on your forums too?
Zach

[scotty b]

NOW we're getting somewhere !! (IMG:style_emoticons/default/smilie_pokal.gif)

[davep]

Well, it appears most of your problem was the extra base shim on cylinder 2. However, you said the gap was .005 to .007 and you took out .010 . So you should still have an error of .003 to .005 even though the cylinders are .002 different. Wow, lots of tiny errors all over.

What I would try at this point would be to clamp the cylinders down and remeasure the evenness of the two cylinders in place. If you get an error, then try switching the cylinders. One combo should be the the better of the two, so use that combo and assemble. Then do the leakdown test again.