Operating Conditions for Mechanical Seals

Good questions to review. Another question is what has changed over the last 2 years? Seal supplier the same? Different mechanic installing the seals? Has the process varied? Any change in process operation? Change in staff? Is the shaft or seal chamber properly cleaned and in good condition?

Something is different now than when the seal life was 9 monthes.

Hi Ried,

It is curious, but the maintenance staff has changed. The last seals were installed in the 1st week of january 2010. Bewteen january 20th and 23th both seals show leak into the process, around 0.2 - 0.3 liters/h. barried fluid is TEG. Process specification has not change, Reactor level and pressure and temperature are the same. I need to check the method used has changed, I assume that mainttenance guys are cleaning shaft and chamber properly, but I need to check it also. Seal supplier is the same.

Hi Tornado, thank you for your answer. The barrier fluid is TEG (triethilene glycol). TEG is going into the process. Initial temperature and pressure were 110°C and 12 bar when the leak started; then the temperature was increased to 150°C and the leak decreased from 0.3 liters/hour to 0 practically, but we have seen this before a major seal failure.

From: Tim Hawley Master Mech.

Hello Peter67mx,

I'm not very familiar with monomer, I assume it has the same properties as polymer.

Ask the pump manufacturer if they can provide the seals made of Viton material. Its has better properties than most rubber seals and will not deform of swell.

Also check the cartridge seal mounting or face surfaces for a worn groove or nick causing wear to the seal surface. You may need to replace additional parts or replace the pump.

Take a good look at this link for more details:

http://www.pspglobal.com/

Good Luck,

Best Regards,

Tim

Hi Tim, I'll check the pump cartridge. As the pump is very expensive, I'm worried about a pump replacement.

Filter the barrier fluid better?

We have thermosiphon in barrier fluid. Filtering requires a pumping system. It is not a bad idea, this pumping system has a low operation cost compared to the price of the seals.

Hi peter67mx,

You have given a lot of information, but to answer your questions you have to give us a bit more. In order to calculate (estimate) seal chamber pressure you need to tell us what kind of pump it is. I guess single stage overhung, but we also need to know if the impeller is balanced or not (holes in the impeller at the suction eye).

What is the barrier fluid?

Plan 53 A or B or C?

As asked by others, which seal fails, the product seal or the atmospheric seal? Do you leak barrier fluid to the deck or into the pump? What damage do you see to the seal faces and or to the Orings etc. If you have the information, what are the balance ratios of the product and the atmospheric seals? (If your seal vendor does not know the answer to this then start looking for another vendor).

What kind of pumping ring is installed on the seal and is it uni-directional (pumping the right way)?

If this is a single stage overhung pump with discharge wear rings and a balanced impeller then your seal chamber pressure can be estimated by

where Ps and Pd are suction and discharge pressure respectively. (This is an intelligent approximation, for a real figure a pressure gauge works wonders, or ask the pump vendor for his educated guess).

Your barrier fluid pressure should never be allowed to get lower than 2 bar higher than the maximum seal chamber pressure achievable under any circumstances. (So use shut off head and maximum suction pressure for your calculations).

In order to get a decent interval between topping up of the barrier system if you are using Plan 53 B you would then need to have the barrier fluid re-fill pressure about 5 bar higher than your minimum pressure.

By the way, in a modern well designed petrochemicals plant with double seal systems one would expect a seal lifetime of around 5 to 6 years.

Hi Kaisan, Thank you for your comments. I don't have all data at this moment, I'm still waiting for some information based on your replay. Here are some data:

The barrier fluid is TEG (TriEthylene Glycol), the product seal is failing, so the TEG is going to the process, some times the operation people increase the temperature from 100 to 140 or 150 °C and the leak decrease a lot. Plan 53 is a thermosyfon presurized at 12 bar. The seal is Chetra brand type 821. The average time life of the mechanical seal in our installation is 10-12 months, but recently in one plant is around 3-4 months. Some guys think that we should try a different barrier fluid like silicon food grade.

Thank you a lot guys for your comments. I have a lot of work to do in order to improve the life time of our system. All your comments are very significant for our team.

I want to check for any problems with piping and pump installation as thermal expansion, as well as alignment techniques, seal installation procedures, training degree people, seal components (materials), etc. We also want to ask the vendor for recommenations to solve the problem and, based on the las Kaisan replay, the capabilities of the seals we are buying.

I'll let you know what we found.

Thanks!