M119 Cylinder leak down test results - lessons learned!

[JC220]

Hi all,

OK so it has been discussed here that I recently picked up a used 1992 production replacement m119 motor to go into my 1995 S500 Coupe.

Before going ahead with all new seals throughout the engine prior to install and firing it up for the first time I wanted to test the health of the motor.

It has unknown miles since it came from a 500E - but was a w140 119.970 engine (what I actually wanted / needed) sucks for him but was lucky for me in that respect.

The motor appears like it had not run in many years. And was partially dissasembled by the guy that had it presumably because he was taking the w124 exhaust manifold etc back off it. The intake manifold was unbolted and just sitting there. It was all a little debris covered but the engine itself was nice inside so I think it has not covered high miles. Appears like a 100k odd mile motor to me.

A compression test gives a limited insight into an engines condition. Aka it will read what compression the cylinders have but not assist in pinpointing where that compression is lost. This is where a leak down tester is much more intuitive.

You pump compressed air into each cylinder in turn at TDC of the compression stroke (both valves closed), take a % loss reading and listen to where the air is escaping. This could be the crankcase, inlet or outlet valves or Into the coolant system etc in the case of a blown head gasket or cracked block / head.

Good practice procedures and great tutorials on how to carry out a leak down test are on you tube for anyone interested in this.

I believe it would be preferable to do this test on a recent running engine - and at proper operating temperature. Which will be more indicative of the engines actual operating conditions VS what I was testing. A cold motor that has not run for years.....

Here are the INITIAL results

Leak down results 1992 M119 ambient temp 18c


1: 20

2: 38%

3: 32%

4: 6%

5: 20%

6: 32%

7: 8%

8: 11%

So you can see there are some high numbers here

Generally speaking it is said that for a car engine (not a race engine) leak down for a great / top condition engine should be up to 10%. A good engine will be under 20% generally. Anything over 20 and nearer to 30 / 40 etc is getting troublesome and shows high wear perhaps. If a cylinder shows near 100% leak off then possible major damage /bent valves etc. This is just very general overview in my understanding! And like compression tests you want there to be not very much imbalance between cylinders.

All engines will have leak off - expect higher numbers when carried out cold too. In my case I could hear crankcase leak off in all cylinders - and in the trouble cylinders especially high leakoff from either the exhaust or inlet valves.

So what to do - pull the heads and have them redone? What to do with such high numbers? Is my engine toast?

 

[JC220]

PS I am being quite naughty here. This is a rhetorical question since this was a couple of weeks back and I know now what had to be done. But this critical step is missing in the tester instructions, online guides and videos. My brother is a Top Fuel crew Chief and knew the answer immediately...... hence the lessons learned title. Update coming in 24 hours but I'd be interested to hear what you guys would do with readings like these?

 

[a777fan]

You sneaky guy!!

Looking forward to the solution

 

[gsxr]

It's so hard to kill an M119 that my initial reaction is no, the engine is probably not toast, unless there's something else you're not telling us. If the air leak is mostly out the valves, it could just be some carbon buildup preventing them from closing fully. Since this engine hasn't run in a long time, it's even harder to diagnose.

But there's no way I'd pull heads off unless you have lots of time and money to burn. The head gasket job is $$$$ due to the never-ending list of "while you're in there" stuff. I also wouldn't have the heads redone/rebuilt unless there was some significant reason to do so, like worn guides. At most, MAYBE, replace the valve stem seals (assuming the heads don't come off).

Remember there are differences on the early engine, for example it will have double valve springs...

 

[JC220]

Hi Dave - you are on the right tracks with your suggestions! Initially I thought I would have to pull the heads and have the valves re ground to seal properly again.

Since I was renewing the timing chain and going to do the valve stem oil seals I had the camshafts out. Which turned out handy since we had to access the tops of the valve stems to test what my brother wanted to do.

What he has seen many times on the dragsters is excessive leak down from the valves being caused by debris on the valve seats. (Foreign bodies, Carbon deposits etc) What has to be done is to use a dead blow hammer and soft object to hit the valve stem and open it a little for a split second. Thus Any trapped debris is cleaned and blown out of the way by the compressed air inside the cylinder and suddenly the reading improves and the valve seals correctly again.

This was the case in my engine too. An issue perhaps exaggerated by the time the engine stood unused. And with the intake manifold loose there was higher chance of debris getting onto the vale seats.

We used a small dead blow hammer and a tool from a PDR kit that was ideal. A nylon rod which could accurately affect the blow but also no chance of damaging or marking the valve stems.

We targeted only the higher loss cylinders to clean the valves in this manner. Note that as soon as the motor would actually run any debris would be cleared away anyway!

These were the tools used:


It was quite fun to watch my brother do this. The gauge immediately improved with each tap on the offending valves and the air leakage could immediately be heard dropping off. A few taps on each valve and the readings really came good. Here are the "real" leak down test results which may serve as a rule of thumb for someone looking for m119 data (which I could find no where BTW!)

Leak down results 1992 M119 ambient temp 18c


1: 7% (got down from 20%)

2: 12.5% (got down from 38%)

3: 10% (got down from 32%)

4: 6%

5: 10% (got down from 20%)

6: 7% (got down from 32%)

7: 8%

8: 11%

These results are very good indeed now - and in particular with a cold motor that has not ran in such a long time. For example the cylinder walls and piston rings will be dryer than normal of oil due to standing so long. So this is a great engine and this crucial trick my brother told me saved me untold unnecessary work pulling the heads etc.

 

[gsxr]

We need your brother to buy a 500E and start tinkering with it, and posting here!!