Pressurized dual seal arrangements 2

[gilus02]

As I know there are 3 plans for Pressurized dual seal arrangements (53A, 53B and 53C). We usually use these plans when we deal with a corrosive fluid, (not suitable for flushing).
The question is what are the advantages and disadvantages of these three plans? (initial cost, long term cost, maintenance, etc)
Thanks for you help!

Mohammad

 

[Reifleman2]

53A: Most basic and common form of a pressurized dual seal support system. Easy to support, fill and monitor (via sight glass).

53B: Typically used when a refinery (or other manufacturer) does not want inboard seal leakage to go to flare or sump. When the inboard seal fails, instead of filling up the seal barrier fluid pot and then discharging to flare, the excess fluid dumps into the accumulator and registers on a pressure indicator. Seal pressure builds and a lot of bad things can happen if the failure is not noticed and pump shut down. A 53B should only be used when a process absolutely cannot be allowed to go to flare or sump. I don't like 53B's unless they are absolutely necessary.

53C: Introduces a piston accumulator to regulate fluctuations in stuffing box pressure. The piston provides a variable pressure on the barrier fluid and (in practice) ensures that the barrier fluid pressure is always a certain psi over the stuffing box pressure. These are used in applications where large swings in pressure can occur. It saves a lot of money and effort on barrier fluid loss. For example: A pump sees stuffing box pressure ranging from 150-200 psi. In standard plan 53A, the N2 pressure would be set at something higher than the max pressure the seal can see, something like 220 or 230 psi. In a 53C, the piston is set to track at 30 psi above stuffing box pressure, effectively lowering the loss of barrier fluid when the pump is operating at the lower pressures.

A side note: Piston regulators are (IMHO) a slightly older technology. Many companies make much better N2 regulating systems, and specifically I have experience with a company called Tescom. Tescom regulators are awesome and can help save a lot of time and money when dealing with barrier fluid filling and support.

Finally, a plan 54 is another form of pressurized barrier support systems (and kind of a catch all). This used for more extreme or hybrid cases where a full gas panel, boosters, or different regulators are needed. For example, I have been involved with a seal using 1300 psi barrier fluid. It has a plan 54.

Hope this helps.

 

[gilus02]

Thanks a lot for your help. You answer was invaluable for me but it raised some questions as following:

- You mean 53A is also equipped with regulating systems and is able to track a predetermined pressure?

- In API-682 no pressure relief valve is considered for 53A. It means that there is still possibility for pressure increment in the fluid reservoir?

- The pump pressure shall not reach to a pressure higher than Shut-Off pressure? (is it right?) So if the pump vendor provide 53B plan with a bladder charged 10psi more than shut off pressure, is there still possibility for dumping fluid in accumulator?

 

[Reifleman2]

Happy to help.

1) Sorry for the confusion on the 53A, it is not attached to a tracking system. It is simply dead headed with a predetermined amount of pressure (through a regulator) and kept constant. A 53A will not be able to track up and down with a varying stuffing box pressure. For example, the stuffing box pressure may vary between 150-200 psi, but the N2 pressure will stay constant above the max pressure the box will see, so 220-230 psi.

2) Your correct about the 53A lack of a pressure relief valve. Ill describe the 53A in more detail: The idea behind a 53A is that there is no need for lots of extra piping and support for a set of seals that may not be capable of receiving an on-sight flush of clean fluid. Instead of running a clean fluid all over the plant, a 53A allows a self contained pot to be installed at the location of the pumps and then be periodically checked for barrier fluid levels. A 53A pot should always have clean barrier fluid in it at a higher pressure than the process pressure, and therefore (in practice) should never see back flow from the process side and no need to vent.

Now, is that what always happens? No (we don't live in a perfect world). I have seen two primary failure modes for these 53A's. The first is loss of barrier pressure (N2 system goes down or tanks are not recharged). Once pressure is lost, the inboard seal no longer has a higher barrier fluid pressure than the stuffing box, and will begin to run hot on the nasty fluid being pumped. If not caught fast enough (with low pressure alarms or something similar) the inboard seal will fail, back fill and pressurize the clean barrier fluid pot and begin ruining the outboard seal. Not much fun to clean up.

A second failure mode is inboard seal failure over time. This is inevitable and will happen as the seal wears out its life. The barrier fluid will begin to flow more freely into the process and N2 usage will go up as well. This is when equipment should be shut down and seals changed. Unfortunately, many people will simply continue filling the barrier pot with clean fluid more often and dumping more and more barrier fluid into the system. Inevitably the inboard seal will fail totally, and who knows what will happen then (I've seen some crazy stuff). You now have a failed inboard, no clean barrier fluid and N2 fighting the stuffing box fluid along with the outboard seal running in some form or fashion.

3) Again, sorry for the confusion on this one. I focused more on the parts I didn't like about the 53B than the accumulator and bladder (recent bad experience with a modified 53B). There are definitely times when a 53B is the best solution to a problem. It's really just a 53A with a membrane between the N2 and the barrier fluid to prevent gas entrainment. You are correct in your assertion, the bladder is designed to be charged at a higher pressure than the seal will ever see, and if it does see that pressure the entire system should already be shutting down. Same as the 53A, but with the bladder instead of gas directly on the barrier fluid. The failure modes for these can get a lot stranger than a 53A, but they usually fail in similar fashion to a 53A (loss of N2, inboard seal wear).

A final thought: These piping plans are all "standard reference," meaning they can be modified and adjusted to fit your needs. I've seen many different systems modified with Tescom regulators to help with N2 and barrier fluid usage (one of the more common upgrades). They are designed to be a baseline for design, and modified if needed.

Man, I should write a novel.

 

[gilus02]

Thank you a lot Mr. Reifleman! What you did is more than a reply to my post and I don't know how can I express my gratitude?

I want to ask another question about this plan and sorry for any inconvenience this may cause.

- No flushing (like plan 11) could be used along with plan 53? Is it right? Because this plan are mainly used for fluids which prone to polymerizing or being corrosive and not good for flushing.

- If we have got flare or sump drum in our site, how much higher is the initial cost for 53C than 53B?

- I have attached some data about the pumps of our project. Please provide me any possible guidance for Mechanical Seal selection as I am young engineer. It is noteworthy that the pump vendor has proposed 53B for these pumps.

Thanks a lot!

 

[Reifleman2]

Happy to help, I enjoy rotating equipment.

A plan 11 actually can be used in conjunction with a 53, but it all depends on viscosity and what the piping can handle. The whole idea behind a 53 is ensuring the seal faces run on a cleaner fluid than can be offered by the process. However, this use of cleaner fluid at the sealing surfaces may not deal with heat generation all on its own. It is very common to have a 53 system in conjunction with an 11 (12, 13, 14 etc.) simply to help with heat removal at the faces. You end up with internal fluid circulation from the 53 and some form of circulation within the stuffing box to help flow heat away from the sealing faces. Its all dependent on what the application calls for. From a sealing perspective, more flush is usually better (but not always!).

As far as costing goes, there are a LOT of things that go into a project like this, and simply perusing the spec sheet does not provide me enough information (I apologize). I am happy to help with general flush and seal information, but actual seal recommendation should come from someone far more familiar with the project (and whose neck is on the line, it makes them more motivated). I would feel terrible giving advice or recommendations without an intimate knowledge of the project, systems etc.

Good luck

 

[Reifleman2]

A small edit: a plan 53 does also help with heat generation (they include a heat exchanger), I wasn't completely clear. My point was simply that a plan 11 (or similar) can subsidize the heat removal from the faces.

 

[gilus02]

Thank you again Mr. Reifleman!
I understand that here isn't a good place for asking people to comment about a project in detail but in our project the vendor has offered 53B with tandem arrangement. No other plan has been proposed as the process fluid is corrosive and not suitable for flushing.
I first proposed 11/53B but with mentioned reason it has been changed.
Do you have any recommendation?

 

[Reifleman2]

Just glancing over the spec sheets, it may simply be that your seal manufacturer doesn't believe that the plan 11 is needed at the heat and pressures you are running. I see nothing wrong with a tandem cartridge seal on a plan 53B flush. 53A's are easier to deal with and maintain, but the 53B does guarantee isolation of the barrier fluid from everything else. I usually see 53B's applied when gas entrainment is the primary concern, but this application my be useful as well. I would check with the your seal guy and ask why they don't feel a plan 11 is justified, and see what they say. Also, why the 53B versus a 53A? They may have more insight and better reasoning than you or I. I prefer a 53A due to the more simplistic nature, but if a 53B is needed, it is what it is.

If you are fairly concerned about additional heat dispersion, you could always request the seal be outfitted with a plan 71. This is another of those "catch all" plans, and allows you to request the seal come equipped with a plan 11, but plugged for future use. If the new seal is installed and begins to have issues, it is a simply matter to pipe a plan 11 (or similar) into the plugged port on the seal gland plate.

Now, I don't necessarily recommend this, as it makes the seal more complicated and prone to mechanic/operator mistakes. Extra ports with plugs tend to make people feel like they are missing something and that it needs to be hooked up.

The path I recommend (after all that discussion): Talk to your seal manufacturer, and ask them to justify their logic. Why no plan 11? Why a plan 53B versus a 53A? I imagine they will have good reasons, but if not it opens the discussion for changes. Also, things can always be modified and fixed at a future date. Seal manufacturer's (typically) want their seals to have high MTBR, and will therefore help adjust designs as failures occur. The seal may go in and run great, or it may need adjusted flush plans down the line. Either way, a good seal manufacturer has drafting, design and engineering support to make those changes as needed.

Good luck

 

[gilus02]

Thank you very much Mr. Reifleman!
When I read your comment I feel I am reading a technical handbook which its author writes the book in the way that I want.
So I want to ask another question if it's possible
Have you ever seen 53B plan alone? I want you to describe the conditions where there is an obligation to use 53B without flushing. Regardless of polymerising could corrosion make process fluid unsuitable for flushing? Our fluid is caustic 5%-10%
I've made a mistake! The vendor proposed 53B with back to back arrangement

 

[Reifleman2]

53B with back to back arrangement is fine.

I have seen seals using only a 53B system, primarily when the service is excessively caustic or abrassive. When advantages of removing extra heat from the faces with the addition of a standard piping plan (like an 11) are overshadowed by the potential damage from the process being pumped, then the extra flush should not be used.

It all comes down to what best fits the system, and changes can always be made int he future. Seal's are an ongoing fight to find the perfect system for a given application. The initial seal may only last a few years, but as the seals degrade, the failure modes can be recognized and adjustments made to the design as needed.