I would like to know which type of mainteanace strategy is popularly used for these "in tank" and "pot mounted" pumps. Many times vibration probes are not working in this extremely cool environment and onset of mechanical failures are impossible to diagnose, unless full blown performance problem like flow, power or pressure is surfaced out.... Does any body use scheduled overhaul or sort of preventive maintenance ?

Comments from LPG, LNG, NH3 and Ethylene pump users appreciated..

This message has been edited. Last edited by: RR,

Are you refering to submissible pumps? Are they installed as 2x100% units? Any pump switching policy?

Yes, these are product lubricated, submerged pumps, with 2 X 100% spares. Spare pump is tested every month.

I guess the pump switching policy should be optimised. How big are these pumps? Are they expensive? What materials of construction are used? What are the operating parameters such flowrate, temp, press, etc? Do you have the operating history to get its MTBF?

This message has been edited. Last edited by: Josh,

Hullo Raj,

Canned LPG/LNG pumps run reliably for very long periods as long as the original assembly and installation work is properly done. As you point out, their design does not make it easy to assesss their condition easily. The heavy insulation, usually 6" or more in thickness, dampens vibrations and noises, so even indirect measurements are difficult.

Note that the internal environment is clean and at a constant temperature, and it is well shielded from the external environment. As long as the pump is operated within the design envelope, its operating reliability should be pretty close to its intrinsic or design reliability, and this would mean failure free runs of 25000-50000 hours or more.

I would advise you to
1. Do as little intrusive work as possible on a preventive basis
2. Use the vendor's engineer any time you do any work on them
3. Use a duty-standby cycling
4. Use the motor current signature analysis and operating parameters to understand how the pump is behaving

As far as LNG tank pumps are concerned, the electric cables can get snagged in the suspension cables in some designs. Check with the vendor on the performance of their design in other locations.

Vendors of these equipment are often willing to share information about other customers, so you may be able to do some performance benchmarking with them and share best practices.

V.Narayan.

That sounds like almost a run to failure or breakdown maintenance strategy which we use for submissible or sump pumps which are of small or medium size. However, not sure how to do that MCSA. Who should do this, Electrical or Mechanical engineers because the motor is normally under Electrical?

Josh,

Step back a moment to cpnsidere why we maintain equipment. Degradation is caused by a number of factors, including fouling by particulates, wear, corrosion, erosion. misalignment, unbalance etc.

In my earlier post, I said we should assemble and install the equipment properly. Implied in this is that the item is designed properly. These steps should minimize initial imbalance and misalignment. This also means bolts are tight and jointing is done properly. If the joints are designed and made up properly, they will remain so for very long periods.

Operating the item within the design envelope and the clean environment reduces wear, imbalance, and joint loosening. Proper design minimizes corrosion damage.

With canned pumps, the pump and motor are on the same shaft, so coupling issues are eliminated. The bearings are product lubricated, and as long as the vendor is present during reassembly, bearing clearances and liquid flow paths will be close to optimal. So lubrication failures are minimized.

The motor is cooled and lubricated by the product, but it could fail due to electro-mechanical stresses after some time. These should be picked up by electrical signatures. As to who actually measures them in your company, that is purely an organizational decision. As long as the technician is competent and analysis is done properly, it should be possible to 'catch' potential problems. Operating parameters are also useful in checking the status.

I am sure you appreciate that it is not necessary to 'pro-actively' open up equipment because it is fashionable. Chances are we will cause failures in perfectly good equipment. These equipment are usually high-cost items in critical services, where downtime should be minimal.

We should maintain to arrest or recover from degradation, no more, no less.

V.Narayan.

Raj, how are your submissible pumps now? What maint strategy do you use now?

Josh, We may finalise, working hours based overhaul (40000 hrs).

40k/8k per yr = abt 5 yrs overhaul. In between overhauls, what strategy do you use to keep them at tip top conditions?

Josh, These pumps are fitted with online vibration monitoring system, so we depend on that.

Josh,
In an earlier note, I explained why we have to focus on degradation management. Canned LNG pumps that are designed, installed, commissioned and operated properly DONT need a lot of intrusive work between major overhauls. This is because:
- the product is super clean
- the operating temperature is rock steady
- there is no mechanical seal to worry about
- lubrication is provided by product stream
- the motor is cooled by product stream

The motor can be damaged, e.g., by frequent starts or by continuous overload; good operating practice will minimize this possibility. Internal screws or bolts may work loose, but these are hard to detect in the best of circumstances; any intrusive work is likely to make the situation worse, not better. External insulation can be damaged, causing hot spots which may resultin structural distorsion. In turn this can cause pump failure. This fault is easy to detect with thermography (or even visually, as we can see ice patches on the surface).

So there is really no need to do any intrusive work between overhauls. Using a duty/standby or 90:10 switching, at least one pump will see minimal wear, minimum damage to windings etc., so if and when we do overhaul the duty pump, we will have a reliable spare unit.

Raj, while I support your 40-48000 hour overhaul policy as heading in the right direction, I suggest the trigger point should not be age but operating performance. So you may decide to do it earlier if the performance begins to drop steadily or if not, postpone it even further. At the risk of repetition, please make sure you have the vendor specialist when you do the overhaul. It may cost a lot, but that would be well worth it. Certainly it will be cheaper than pulling the pump apart beacause it seems fashionable.

Do you do MCSA in addition to online vibration monitoring?

Points noted: We should not overhaul as it is fashionable and breakdown maintenance strategy is not necessarily bad.

This message has been edited. Last edited by: Josh,

Thanks Mr. Vee for suggestions and Mr. Josh for making the topic interactive.

We do not have Motor Cusrrent Signature Anlysis.

What we have planned is every 40,000 hr, rotating engr. will access the need for overhaul based on vibration and hydraulic performance. Just to share, cost of new pump can be 500,000 US $ or more and hence breakdown is not affordable.