Leakdown during rebuild
 

I would search on this but I know that 'leakdown' and 'rebuild' would give me a lot of results noise. I have not read a question like this, so I will risk asking it.

I'm about ready to install the remaining rockers during my rebuild. Lacking the confidence of a prior 911 rebuild (but the experience of a VW rebuild with a head leak) I am curious if there is any information, utility, harm or embarrassment in doing leakdowns. I figure a leakdown would test valve sealing (did I bend any with those gentle bumps during clearance testing), head gasket leaks and stupid ring problems. And it would be so easy to figure out where the leaw was coming from with the engine bare.

Am I nuts??

Sounds like a good plan to me. The time to find a bad mistake is when the engine is apart, not in the car running.

It's only embarrassing if you screw up tell everyone.

I would say go ahead, but don't expect the same results as a broken in engine. It might prevent you pulling the engine if you make a big mistake, but once you've assembled it and set the cam timing I doubt you'll have an issue. It ain't rocket science.

I don't think you will get meaningful results. Cold engine and unseated rings...numbers are gonna be all over the place.

After I get my timing where I want it I will give it a whirl. I swear my rings are fully seated from all the engine turning during timing. If you don't hear from me, assume the worst.

I worked with a guy that did an apprenticeship at the Porsche factory. He did a leakdown on every engine that he built. He said it's a lot easier if you find a problem when it's on the engine stand than when the engine's in the car.

I leak down all my motors when new just for giggles. The worst I have ever seem is 10% on a new never turned over just assemabled motor.....

I use it more to make sure everything is consistant. With the motor on the stand it takes about 30 mins.....


Cheers

For another data point, I performed a cold leak down test on my rebuilt engine before installation - no more than 6% on all.

Sherwood

Did you do another leakdown after you ran the engine? Just curious what the difference might be.

Yeah. That would be the other part of the equation. Nope. The tester was borrowed. I'm left with a compression tester. That wasn't a very good ending to a teaser, was it? Sorry.

Sherwood

My Harbor Freight tester ends up using very low pressure. Good and Bad.

Good - I was able to leave the flow going while I cycled the piston through a full 4 strokes. I got appropriate noise, flow etc. My readings were 20-25% leakage and it seemed to be all ring flow. I am using a set of P&C's put together from 2 engines and eBay - the right bank has a tighter fit and had consistently less leakage. My ring gaps were 0.4-0.5 mm which leaves a pretty large hole. At actual RPM there really isn't time for much to sneak through that path it seems.

Bad - higher pressures would have tested the rings better. And there would be more opportunity to pneumatically drive the engine and injure myself. No good stories - sorry!

Thanks for the feedback and encouragement. I'm in the process of getting the fan and topside stuff back on. Need to find me some SSI's ( thin flange, no sensor - scratch and dent fine, but would like shine!)

You need to replace the right gauge with a better 0-100psi pressure gauge.
Then you can bump up the inlet pressure without pegging the right gauge.

I have the same gauges and with the stock gauge, my inlet was something like 10psi and I was also getting like 20% leakage #'s.

After changing gauges and upping to 50psi inlet, leakage dropped down to 10% and less. Plus with the higher pressure you can more easily hear where the air is coming out.

I performed a leak down test on my 3.6 after putting on the heads in my top-end/re-ring rebuild. All cylinders were <5% leakdown.

It's a good check to make sure nothing is wrong before you to to far. For grin and giggles, I also did another leakdown after the engine was assembled. Same result. Also provided some comfort that there is no obvious/glaring mistake.

I've just remembered a good example of why this is a good idea. My old boss was assembling a 2.2S when he was called away to the phone. He returned and installed the heads, cam towers, etc. After the cams were timed and all of the valves adjusted he did a leakdown, one cylinder was at about 50%. the air was coming out of the bottom but it didn't sound like it was out of the exhaust. He blew it apart and found that the #2 headgasket had slipped down out of the groove in the cylinder a little bit and wasn't sealing.
It was a matter of filing a small burr off in a couple of spots and replacing an $8.00 gasket and retiming one cam. Way better than pulling the engine out again and all the crap off of the outside again. If he had been a little bit taller he would have had the engine tilted a little more vertical, or if the phone hadn't interrupted him, or if he'd paid more attention as he was torquing the heads he would have noticed the difference.
So often it's a series of small "if's" on the way to disaster.

I started to do leak down test on new engines on the stand years ago because of my fears of breaking a ring on cylinder assembly. At that time I was putting the piston with rings on the rod and then installing the cylinder which can be difficult especially when dealing with a mag case with case-savers that are not exactly installed square to the case. I now install the pistons in the cylinders on the bench and then onto the case. The numbers I expect from a cast biral cylinder honed using a 800 grit stones is 8% to 12%. What I also found was cylinder head leakage between the cylinder and head joint. Where the heads have been surfaced and installed on new cylinders I will lap the cylinder into the head with valve grinding compound to solve the leakage. When that does not stop the leakage I would use a VERY thin coat of locktite. I use a 100 lbs of air pressure and care has to be taken to keep the engine from turning over. On new Nicasil cylinders the leak down numbers will drop below 5%. All these numbers are with lightly oiled cylinders. The hone coarseness has a greater affect on these numbers than ring gap. When I finally went to Scotch Brite pads to hone used cylinders the numbers dropped to 3%. A well known engine builder never hones used Nicasil liners and gets 2%. I am not comfortable with not honing the cylinders. Over the years this method has found sealing problems I would have missed until the engine was installed so it has paid for the time invested. In one case particularly I found a low spot in the top of a new Nicasil cylinder that I missed when I was checking with a dial bore gauge. I suspect it had been dropped at some point because it was .008 out of round at the very top for about 1 inch in circumference where the top ring seated causing a leak down of 25%.
I always check the heads with clean solvent for valve leakage before I put the head on the engine. I pour the solvent down the port and inspect the combustion chamber for leakage. This procedure has also saved me from grief over the years.

aws

That was a very useful post, thank-you for the info.

Glad I came across this thread... I have the harbor freight leakdown tester too, and my 930 tested between 20%-23% on all 6.
The motor was cold because starting it is not an option at this time...

Seems like a decent made set up for $30, but it only lets you put in around 12-13psi before pegging the right side leakdown gauge.

Where have you found a decent but inexpensive 100psi gauge replacement for the right hand gauge?

www.mcmaster.com
Look up 3847k71

$8.50 plus shipping.

thanks!!

Jim

Just added this thread to the PelicanWiki. Thank you.