Failure of Mechanical Seals 3

[gikian]

I am a student doing internship in a chemical industry<br>
One of the major problems they are facing is the failure<br>
of the mechanial seals.My assignment is to improve<br>
the design of mechanical seal.or bring some<br>
good changes.<br>
Therefore i need information inthis regard<br>

 

[djk]

You have your hands full. When you can answer the following questions you will be on your way. <br>
<br>
What form does the seal take? Ball? Poppet? O-Ring?<br>
What are the seal materials?<br>
What is the working fluid?<br>
What are the working pressures?<br>
How is a failure defined?

 

[GregRupert]

I have seen and trouble-shot many a mech-seal. Rarely does the solution "jump-out." My suggestion: start by talking with rotating equip. specialists and read a good mech seal vendor manual (my favorite: Dura Mechanical Seal Handbook). Talking with vendors is helpful, but keep in mind that seal sales=business=$$.<br>
<br>
Investigate easy failure causes/solutions 1st and eliminate and progress to the more difficult:<br>
1. Seal type correct for service? Use KISS principle...Single mechanical, double, tandem, etc.<br>
2. Check fluid pressures...a stuffing box P w/ pump discharge blocked &gt; sealing fluid P will wreck havic & cause a seal to "fail" (leak). As djk mentioned, define what is called a failure.<br>
3. Check MOC of stationary and rotating faces & compared to <br>
what the process fluid is...<br>
4. Temperatures of process fluids OK for seal?<br>
5. Solids present in seal flush fluid?<br>
....<br>
<br>
Hope this helps and gets you off on a good start...

 

[VibrationSpecialist]

Failure Analysis for mechanical seals are either in installation, operation or life time. I guess you should concentrate on the first two. Decide what type of mechanical seal you want to study (contacting or non-contacting, i.e dry).<br>

 

[manian]

Analysing the failure of seals requires detailed study and some experimentation. For example,<br>
<br>
- What is the type of arrangement <br>
- Static sealing ?<br>
- Dynamic Sealing - Rotary, Reciprocating ?<br>
- What is the process fluid ?<br>
- What are the present sealing arrangement and materials <br>
involved ?<br>
- What pressures and temperatures are involved ?<br>
- Study closely the Clearances between the shaft and <br>
housing. Higher pressures would mean lesser clearances <br>
to avoid any extrusion and blow out.<br>
After you collect these details and have your info ready speak to some good seal vendors - some of the pioneers in this area whom i came across are Parker, Green Tweed, EGC and many others are also there. You can refer their catalogs and speak to them. Most of them have their websites just search in the engines. One good book i came across was 'Seals and Sealing Handbook' by R H Warring. i hope you would find this information useful.<br>

 

[Tom584]

I am in engineer now, but spent over 20 years as a knuckle buster. The main causes of seal failure that I have seen, assuming the pump and seal designers did a good job, are improper handling of the seal at installation, improper installation of the seal or improper operation of the seal. These are also the easiest to identify and correct. <br>
<br>
<p>Tom Worthington<br><a href=mailtoworthi@astro.as.utexas.edu>pworthi@astro.as.utexas.edu</a><br><a href= > </a><br>

 

[stonerabroad]

being a beginner at seal engineering might i suggest a website for reference.

http://www.mcnallyinstitute.com

when i was starting out, there was tons of info here that brought me up to speed quickly.

 

[assco]

I deal with mechanical seals everyday. GregRupert gives some good advise. I would add the following:
Material compatability
Heat- including seal generated
API piping plans
Do the seals fail on start-up?
Stuffing box pressure
Operation-Does the equipment cavitate? NPSHA

 

Some very good advice from assco, manian and rupert, add may be:
- Correct material selection for the combination of rotating and stationary faces.
- Multi spring or single spring arrangement (depends on how dirty is the medium)
- Gasket material
- Shaft run out
- Vibration status
- Shock loading
- Solid impurities (scorring of faces)
- High temperature causing evaporation of film between faces and thus luck of lubrication.
- Thermal shocks
- Seal face surface fatigue

It is true you have a hand ful'upfuLL!!!
Good luck!

 

[JulieM]

One problem that I know was mentioned when I was first learning about mechanical seals is that they are often not installed correctly. This is due to the fact that the seal manufacturer provides installation instructions that get thrown out by the mechanic as soon as they arrive. One of the most basic causes of a face-to-face seal failure is introduction of grease (some mechanics think you should grease the faces of all seals) or just scratches, fingerprints (oil or grease), or dirt. For a seal that is supposed to form a chemical interface of its own accord, you've basically got a failure before it's installed.
Check the installation instructions then watch the process to make sure it's happening right.
I'd also suggest the McNally Institute for information and even classes.

 

[kimbo1]

well you have received some useful answers!
I would like to add that big name (burgmann, crane etc) mech seals are actually extremely reliable when not mis-applied into the wrong application or mistreated by turning off cooling water etc.

 

[blueprint]

It is highly unlikely that you are going to design a better seal but you may be able to upgade the seal standards on some of their pumps.
Unlike like stuffing boxes mechanical seals are very sensative to the product and pump specifications.

I would go to the original specifications of the pump and check what the specs. were for the seal when the pump was installed. Check to see if a more up to date seal is now available. Also check if the capacity of the pump has changed (during a turnaround) and if the seal specs have been updated to match.

Also study the manuals of the seals used on the plant and then sit down with the chief mechanical fitter(not the engineer) and let him explain the problems to you.

 

[sprintcar]

One failure that I didn't see was Dry Start - even when the pump and system are prefilled, there can still be an air pocket near the seal. This happens typically in bottom horizontal discharge pumps but is present any time you can't bleed the pump cavity. The pump can only fill to the level of the suction pipe diameter. Some seals can't take even momentary dry running.

As mentioned above, get the seal vendor involved. You might be able to build a better mousetrap, but there's going to be a lot more mice by the time you get done. Keep the wheels on the ground
Bob
showshine@aol.com

 

[dougholmes]

Interesting. I don't find any reference to pumps by the gikian. Until it is known the problem is in a rotating shaft seal, rather than a static seal or a sliding element seal, it is hard for a person to know which of the good ideas presented here are most appropriate. It is even possible that this is not a pressure containment problem. Can we get more info on the nature of the problem?