A Comparison of the Timing Chain Tensioners

Used on the BMW S50 and S38 Engines

There has been much discussion about the use of the S50 timing chain tensioner in the S14/S38 engine applications. The later S50 design has several features that make it an attractive replacement (or upgrade) to the S14/S38 parts:

1. The S38 part has a diaphragm and stack of metal seals that inevitably leak.
2. The S14 applications are notorious for having a noisy chain on startup, which can be remedied by using the S50 part.
3. The S50 part is much easier to install, due to a pre-loaded spring design.

For many users, the above reasons make the choice to use the new part an easy one. However, there are a few who have suggested that doing so will cause increased wear on the timing components, especially the chain guides. To provide a little more insight into the question of extra wear, I decided to measure the two units to see if there was really any need for concern.

Here is a picture comparing the S50 and S38 designs:

It’s pretty obvious that the S50 part (top) is much simpler to handle. It also has a spring that can’t be seen in the picture, because it is inside the piston assembly on the left. This looks very similar to the S38 piston below it, but it is really a smaller piston inside a larger piston (the green and blue inner cylinders below):

Here, the red arrow represents the spring, and the black outer cylinder is the outer tensioner housing, filled with oil under pressure, pushing on the blue piston/spring housing. The spring is compressed and retained by a locking ring. When the oil pressure behind the blue cylinder exerts a force greater than the preload in the spring, the blue cylinder starts to move. Since the green cylinder is against the timing chain guide and can’t move, the blue cylinder moves and compresses the spring, increasing the force on the chain tensioner beyond that provided by the preload.

The S38 tensioner operates in the same way, except there is only one inner cylinder, and the preload force is less, making it possible to install the part.

Graphically, here is roughly how the development of forces occurs with the two parts:

As can be seen, at warm idle (or with the engine off) the spring provides the only force on the chain, with the S50 part providing about twice the force (18 lbs.) of the S38 part (9 lbs.) At 40psi, the two parts provide about the same tension, and above this pressure, the S38 part actually provides more force, given a larger piston diameter of 19.5mm, vs. 18mm for the S50 part. At max oil pressure of 58psi (4.0 bar,) the S50 part produces ~23lbs. of force, vs. ~27lbs. for the S38 part. This difference of about 17% is probably not very significant.

At idle when warm, it is true that the S50 part provides twice the tension of the S38 part; however, at 30psi (~1500 rpm for the engine) the difference is only 4 pounds. At 40psi, they are essentially equal, and at 50psi, the S38 part provides 3.5 pounds more force.

Conclusion: I wouldn’t worry about extra wear with the S50 part; in fact, at an engine speed of 2500rpm (common cruise range) the S50 part provides LESS force, which might equal less wear.

One final note on cold startup: With cold oil, the startup oil pressure could be 50-60psi, meaning again, the S50 part provides less force. For a few seconds before the oil pressure builds, however, the increased spring force is probably responsible for the reduced chain noise.

By Kurt Lehman, with thanks to Richard Baxter for providing measurements of the S50 spring rates.