5 replies to this topic

[Amit Sonpal]

We have two identical (same capacity) plants # A & B. In both the plants, we are contineously running 2 same capcity pumps in parallel to transfer hydrocarbon. One pump is steam (MP --> LP) turbine driven and another is motor driven.

We are facing problem of low suction pressure in plant # A steam driven pump when discharge valve is opened fully whereas suction pressure of motor driven pump is maitained. Vibration in steam driven pump suction line is also observed. If discharge valve is throttled turbine driven pump suction pressure is maintained.

In plant # B no such problem is observed. NPSHa is sufficient in both the cases (NPSHa is higher by ~ 2 meter to NPSHr). Furthermore, suction piping geometry, elevation of both the plants are identical.

Is it typical problem with steam turbine driven pump ?

What could be probable reasons & solutions of the problem ?

Any article / literature / link on this topic?

Thanks for help in advance.

[pavanayi]

Amit,
I did not fully understand the problem of low suction pressure. Could you explain a bit more maybe with a rough sketch of the system?
Please show instrumentation locations etc (on both steam side and proces side) along with the history of the problem? When was it first observed? how long has it been a problem?

As a first thought, have you compared the RPM of both turbine drives?
Have you measured/compared the MP and LP header pressures near the inlet and exhaust of both turbines?

Edited by pavanayi, 05 October 2011 - 07:57 AM.

[kkala]

I would agree with pavanagi that a reason for low suction pressure (and subsequentl trouble) in pump of plant A can be: increased rotational speed (RPM), due to failure of rotational control of its turbine (www.nema.org/stds/aboutstds/upload/FAQ-Turbines.doc). This can be checked by measuring RPM of the two identical pumps of A (when problem exists) and B. Discharge pressure and flow in A may also be lower during problem, compared to B.

Editing note 6 Oct 11: "cavitation" of steam turbine is not a proper characterisation, hence the point above was "corrected". The term seems to be used, http://commons.wikim...Revolutions.jpg, but it is not like a centrifugal pump fed with liquid.

Edited by kkala, 06 October 2011 - 11:21 AM.

[Amit Sonpal]

Thanks for prompt response.

I am attaching rough sketch of pump P & ID and schematic of turbine control system. Following information will help u to understand the problem better -

• For this pump, turbine RPM is fixed (~ 2950 rpm) i.e. steam flow will adjust to maintain same rpm for change in pump duty.
• No indication of steam flow, steam pressure & individual pump discharge flow is available. From turbine side only parameter available is turbine rpm. (I have MP/LP steam pressure / temperature at unit battery limit).
• As I mentioned earlier Pump "y" discharge isolation valve is throttled to maintain suction pressure.
• As per plant operator feedback, turbine rpm is NOT increased even if discharge valve is fully open but suction pressure goes down which results in suction line vibration.

If RPM of the turbine is increased for fully open discharge valve than it is clear case of higher steam flow in turbine & malfunctioning of turbine governor. But operator’s feedback denies this possibility.


Is there any way to measure performance of turbine at throttled & full discharge valve opening (with limited availability of data)?

I am planning for pressure survey during fully open discharge valve at various points of suction loop like XV upstream, Strainer upstream & at pump suction (refer attached P & ID). I hope this will throw some more light on the topic?

Anyone faced such problem before?

Thanks

Attached Files

•  P & ID.pdf   60.28KB   28 downloads
•  Turbine Schematic.pdf   26.23KB   28 downloads

[pavanayi]

Amit,

You have not explained about the history of the problem.

From my personal experience, more than 90% of plant operating problems occur due to instruments not showing the values they are supposed to do correctly.
I would get that speed transmitter checked, even if it has been calibrated recently. If you can use a separate tachometer to cross check, I would suggest that too. The plant operator will not be able to know if the speed displayed/transmitted by the instrument is the correct speed of the pump shaft and that the resultant governor control is accurate.

You have only mentioned about low suction pressure. What happens to the discharge pressure of the pump when the above mentioned problem occurs?

[mav9rick]

Amit,
It appears the the turbine driven pump is conveying more flow than the motor driven pump. Due to higher flow in the pump Y, the pressure drop in the suction piping is larger than that for pump X which caused lower suction pressure. By throttling the discharge valve, you are artificially inducing discharge head and hence lowering the flowrate through the pump. Can you check teh following please:
1) Check the strainers for both pumps. It could be possible that the pump Y strainer is clogged. What sort of difference in pressure are we talking about anyway?
2) The discharge seemd to have a minimum spill back arrangement. With valves fully open, is there is a difference in the discharge flowrate measured by FT?
3) Check the original pump datahsheet for possible difference in the design operating flow and the rated capacity. It is possible that motor driven pump has lost some performance if it had been used as a duty pump for long time. This would mean that the pump is not operating on its original curve and hence achieving a lower flow than originally specified yet suitable for your current application

Let us know how you go.
cheers