7 replies to this topic

[raghav8931]

Dear Che elites,

I have been confronted with a waste heat recovery problem. Details of the problem are as follows:

In my refinery aromatics complex, we have 3 Raffinate columns where ~75 Gcal/hr of latent heat is lost through an overheat condenser at 145 deg C. We are looking for means to save at least a small portion (30-40%) of the bulk energy.

From the column overhead, we can generate ~150 TPH of steam at 3.0 kg/cm²a which we don't have a direct utility . We are looking for an option where we can generate a small amount of power from it using a turbine whose outlet conditions would be vacuum (0.15 kg/cm² a). We have a turbine in CPP which is expanded from 17 kg/cm² a to vacuum condition. But I am not aware if such a low pressure steam can really drive a turbine and generate electricity.

The whole idea would is to generate electricity using total steam (approx. 13 MW potential based on 70% isentropic efficiency) or divide the steam into two parts to generate electricity from one part and upgrade the other part to MP steam (at 17 kg/cm²)-[potential for this is 66 TPH of MP steam based on rough estimates).

Please share your experience in similar heat recovery. what would be the approx. capital requirement (FOB cost) for such a turbine.

Regards,

[ankur2061]

Raghav,

3 kg/cm²a is too low a pressure to be useful to drive any steam turbine. You may check these out with reputed steam turbine vendors and they are probably going to tell you the same thing what I am telling. There just isn't enough energy in 3 kg/cm²a steam to extract electricity out of it.

You might require to boost the pressure of the steam ten-fold to have an effective steam turbine for power generation. Using a pressure booster such as a thermo-compressor will in itself require high capital investment as well as be energy intensive (OPEX) for such a large capacity of 150 TPH steam.

The more effective solution for utilizing the LP steam could be as follows:

1. Identify process heating which requires heating medium with a temperature corresponding to 3 kg/cm2a and use this recovered steam

2. Identify all heat tracing requirements in the plant / complex and use this LP steam for heat tracing purpose

3. Identify combustion air pre-heating requirements for fired heaters, furnaces and boilers and use this LP steam for preheating the combustion air.

4. Identify HVAC requirements for the plant / complex and utilize this LP steam for HVAC purpose.

5. Identify heat conservation requirements for storage tanks holding viscous materials (crude, furnace oil etc.) and utilize this LP steam for providing heating for these tanks to maintain temperatures above pour point and ensure flowability.

These are some of the ideas I have provided. There could be several more ideas coming in from other members. Also, some new applications for this LP steam could be identified by you depending on your plant / complex requirements.

Hope this helps.

Regards,
Ankur.

[raghav8931]

The ideas you have provided are very traditional and most of them are already implemented in our refinery. Looking at the extent of energy (75 Gcal/hr), it is clear that we need heat sinks which can accept considerable amount (each to the tune of 20-30Gcal/hr) which is very difficult to find.

Precisely, I want to convert to into a bulk utility like MP steam which is easily absorbed in the refinery and reduces fuel firing in CPP.

I've attached a Mitsubishi document which details the standard turbine frames (condensing) which operate at 4kg/cm2 a. I understand that volume required would be very high at such low pressure and hence capital cost. But would like to know the approx cost and any previous experience in similar heat recoveries.

Attached Files

•  Pages from at_turbine.pdf   67.64KB   25 downloads

Edited by raghav8931, 07 February 2012 - 02:46 AM.

[Art Montemayor]

Raghav:

What Ankur has outlined in not only traditional methods to exploit LP steam sources, but also practical methods.

If you have an assigned project from your engineering lead, then it is a serious attempt to identify what you are seeking. Therefore, the most practical and efficient method for you to get the information you seek is to go to the source of those that claim they have the means to carry it out - Mitsubishi. Their capital estimate(s) will not only come from specific experts, but they will be very credible in your final report. I doubt if you would prefer to use our Forum members' estimates in your report. If you are to report your findings you must furnish credible and expert information that can be researched, verified, and relied upon for a relatively large capital investment - like this one.

I recommend you contact and work with Mitsubishi on this application (if they are interested) in order to establish the basic details that confirm what they publish and claim as being able to supply. If the application can be done, then a capital estimate from Mitsubishi will be the most reliable and credible piece of information. I can tell you from personal experience that there will be very specific and rigorous turbine specifications that must be adhered to and a tight performance guarantee due to the large amount of capital that must be invested.

[R NESAMANI]

Dear Ankur,

Refer point 4, How the LP steam can be used in HVAC....? kindly elaborate.

[ankur2061]

Nesamani,

LP steam is used for humidification operations for air conditioning. Refer the links below:

http://www.armstrong.../humidification

http://hpac.com/humi...ation_reducing/

Regards,
Ankur.

[pavanayi]

Raghav,
This might be a stab in the dark as I am not aware of a lot of things. Even then, what is the lowest pressure steam (LPS) header in your plant?

In addition to what Ankur has said, I can suggest something that can be looked into. If you analyse your steam balance, you might find that there might be steam let down into the LPS from higher pressures to meet the demand.This might be either through the extraction (in some places, its referred to as pass-out)/ backpressure turbine or through a let-down station. If you can part-fulfill the demand of the LPS through the new steam generation (which will normally be heating demands similar to those enumerated by Ankur), you might find that you can reduce the letdown / extraction flow and instead put more of the higher pressure steam through to condensing turbines.

This will improve the total power generation of the plant by diverting the higher pressure steam to generate more power in your existing turbines rather than let-down to be used for heating demands.

Again given the max temperature is only 145°C, that might not be readily feasible. One other option to look maybe is to pre-heat the BFW that is used in the existing boilers, thereby using all available energy for increasing steam generation in your existing system.

[raghav8931]

Art Montemayor,

I totally agree with you. Right now I am trying to make a kind of survey of past experiences and looking only for rough figures and not for my final work. I would be eventually going to vendor for estimates once I confirm myself that it is worth pursuing.

Payanai,

As I have said earlier, the amount of heat is huge (~75Gcal/hr) which cannot be disposed of as sensible heat to BFW.


I am also looking to possibilities of Organic Rankine Cycle and LiBr/water refrigeration system (to make chilled water). Please share past experiences in such heat recovery if you have any. Thanks