Underwater Oscillating Seal Failure
Greetings, First Post.
I have a persistent seal failure problem on an oscillating underwater seal in seawater. The motion oscillates in simple sinusoidal motion, back and forth, through an angle of +/- 50 degrees. The motion is very slow, on the order of 1 Hz. The shaft is about 8mm (0.315" dia), 17-4ph stainless heat treated to HRC44 and polished to 1.5 Ra microinches. The only motion is oscillatory, i.e. there is no axial motion. Two lip seals have been tried, both are PTFE based and are proprietary to the seal manufacturer. In the case of the first seal, the material contained graphite additives. Because of heavy shaft scoring, and reports that the graphite additive could be abrasive, and because of galvanic potential differences between the graphite and the shaft, we changed seal designs from the graphite-filled, spring-energized lip-seal design to a molybdenum disulfide filled PTFE with no energizing spring. There is about .060 of interference or "squeeze" on the shaft at installation.
Both seal designs are resulting in heavy scoring of the shaft at the seal lip, which eventually is leading to accelerated seal wear and leakage, in as little as 50 hours of service. The tests are being run in fresh water at a depth of less than 1 ft.
I have performed optical microscopy and scanning electron microscopy on the failed shafts and see very deep and sharp tracks in the shaft which go all the way around the circumference. A single particle lodged in the seal lip would be expected to result in only a single score mark of 100 degrees due to the oscillating motion present, yet I am seeing continuous, nonuniform scoring around the entire circumference. This leads me to believe that a bunch of particles could be contaminating the seal lip, becoming embedded and resulting in the heavy shaft scoring.
Since I don't fully understand the failure mode, I am plating the shafts with a process (name undisclosed) which will reportedly take the shafts to HRC70-72. Because hydrogen embrittlement bake-off is a problem as a result of a piece molded to the shaft which can't take the bake temperatures, I am using this proprietary process instead of hard chroming the shaft. This process advertizes a lower coefficient of friction than hard chrome, and is also harder (HRC 70-72) vs. hard chrome in the low 60s. Both of these should help my problem.
Can anybody help me out on this one? Is there something inherently difficult about oscilllating rotary motion? My best guess is contamination from some as yet undetermined source becoming embedded in the seal lip and leading to shaft scoring, and I'm running further tests to isolate the failure mode. Just thought someone who's "been there done that" might be able to help me shed some light on this.
Thank you for your consideration.
-Noah
Mechanical Engineer
|