Mechanical Face Seal Leak, Ingersoll Rand Flowserve SMP-1000 Pump

[Papabyrd]

Hello,

Very stuck with this issue and curious what to do. It has to be something simple.




I've used these pumps (Ingersoll Rand Flowserve SMP-1000 centrifugal) before, 4 times, in a similar application with great success. Now I've installed 5 units where the seal begins to leak within 5 days of operation.

It's pumping Mobile DTE-25 oil. Around 5 gpm. The tank above the pump is ~10 feet and pumping back to it. The last unit we purged all air from the casing from the bleed port, let it sit over the weekend and rebled before operating. When pump/seal was disassembled the components were debris free other than the chewed up rubber.

Seals are viton on the last unit that failed.

The impellers are trimmed down from 4" to 3" at the local distributor shop.

The only other items that come to mind,

a max NPSH, where the above tank pressure overwhelms the mechanical face seal.
The construction of the pump, its also hard to check runnout behind the impellar once installed.
The pump capacity is 30 GPM, running below 5 gpm perhaps doesn't push enough fluid by the seal to cool it.
The hydraulic suction hose is somehow collapsing inside and reducing NPSH.
Seals are confirmed to be installed dry and the oil wicks/capillary into the seal face upon startup. Have not considered prelubing as thast not what is supposed to be done.
Good impeller balancing to be confirmed.

Trying to work with the distributor and manufacturer for support now. It's been going on for 14 months. Frustrated.... Thanks for any thoughts on this.

 

[shaggyco07]

Looks like the pump application/selection is wrong. PD pump would be a better pump selection.

Mobil DTE-25 has kinematic viscosity of 44.2 mm^2/s at 40C, which you have a max of 35C so it will be thicker.

This SMP pump is for water and the flow rate range on the curve only goes down to 10 GPM as a minimum.

 

[Papabyrd]

Hi Shaggy,

A positive displacement gear pump is what came out. Those had their own issues. With the trimmed impeller the flow rate can be lowered. The stock charts are for the 4 inch impeller.

You are correct that this pump is listed for water service only.

 

[TugboatEng]

5 days of run time and the rubber is chewed up? It sounds like you're getting EPDM trimmed seals and not Viton.

 

[Papabyrd]

Hi Tug,

Originally it was Buna-N/ Nitrile, the new failed Viton unit is still in service. Spoke the the distributor earlier and he clearly confirmed these units fine for oil service....

Perhaps pushing more flow will better cool the rubber seal elements.

 

[Artisi]

Of course the pumps are fine for oil service but seems the seal selection is not, have you contacted /involved a seal company to assist with sorting out your problem.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Papabyrd]

Currently attempting the work it out with the distributor and manufacturer rather than trying to reverse engineer the product with a seal company...

 

[Artisi]

Why is contacting a / seal manufacturer reverse engineering - considering the seals were more then likely sourced from a seal manufacturer initially. When I was involved in pump manufacturing, anything other than an obvious problem - the seal manufacturer was the very next call.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[TugboatEng]

Viton should easily tolerate temperatures up to 300°F and 400+ for only 5 days.

You say the impeller was trimmed. Was the rotating element replaced during the trim? That would introduce a chance for the wrong sealing element to be introduced.

Buna-N/Nitrile has very poor life expectancy above 150°F but will last thousands of hours before failure.

Did the seal for the stationary element show similar deterioration on the exposed edge?

 

[Artisi]

If the suction hose is collapsing, the NPSHa is extremely negative that could result in cavitatition across the seal-faces.

First step is to assertain the failure mode, is it degradation of the elastomer of an introduced hydraulic problem.


It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[Papabyrd]

The pump trimming happened at the distributor which I was not apart of. The impeller can be removed without touching the mechanical face seal. The stationary element did not appear damaged for the one pump that was rebuilt.

The suction side hose will be replaced with all rigid pipe and a pressure gauge on that side of the pump.

 

[TugboatEng]

Rigid pipe is a pain. There are suction rated hoses.

 

[electricpete]

I'm a little bit confused about the type of seal. In my limited experience, when I think mechanical seal, I think of hard materials at the mating face surfaces like carbon and silicon carbide, and flexible materials like viton etc are only used at parts that don't have a lot of relative motion. Is the face material viton here... or just viton used somewhere other than the face?


=====================================
(2B)+(2B)' ?

 

[TugboatEng]

This is an elastomer bellows type seal. The rotating element consists of a rubber bellows that sticks to the shaft and drives the rotating carbon or silicon carbide face. The bellows allows for some axial dimension changes and seals from leakage between the carbon/sic face and shaft. The rotating carbon face is pushed against the stationary ceramic, sic, ni-resist seat which is sealed in pump housing with a rubber cup seal or o-ring.

 

[82ghost82]

What materials are the rings of the seal?
If I understand correctly you described that other than the "rubber" al the components of the seal were intact, but from the picture you attached the rotating ring seems broken.
I had an issue with a single type mechanical seal Sic/Sic (where the oring of the stationary face got damaged and also the rings broke): this was due to high temperature of the rings: as you may know, mechanical seals must be lubricated; single type mechanical seals (as yours) needs process fluid to go between the faces to cool and lubricate (and that's ok, you have oil!); if no fluid is between the faces the temperature of the rings will rise and they will break.
As in your case I also noticed the damage after few hours of operation.
If you want to use a single type mechanical seal think about:
1- if you have hard/hard material rings eventually replace one ring with graphite (graphite has better lubrication compared to ceramic)
2- eventually install a valve on the outlet of the pump to let operate the pump at a higher pressure, so you will have more process liquid going between the faces to lubricate and cool them

You also may think about changing plan and use a double type mechanical seal (this was what I did, with clean pressurized barrier fluid), more expensive and may not apply to your case because of required installation space.

Daniele

 

[goutam_freelance]

Since you have trimmed the impeller from 4 in to 3 in, the discharge press will reduce by (0.75)^2=0.5625. This lower pressure may not be adequate to maintain a cooling and lubricating oil film at seal face. Pl check.