264 replies to this topic

[PingPong]

So the edited value of 203.0289 kW for the steam in the process is correct then..... is it ?

Yes and No.

Remember that you will not mix all HPS with the gas feed.

How do we move further once the BL conditions are known ?

See my message #121.

Why do we find the enthalpy of steam at 25°C and ideal gas state ?

For the same reason you calculated (in your spreadsheet) all gas enthalpies relative to IG@25C before. I explained that weeks ago.

[ravindra@096]

The battery limit conditions specify the temperature of natural gas at B.L is 40 degrees Celsius and the pressure is 36 bara.

 

The natural gas and H₂ mixture temperature is calculated by the formula m.C_p.(100-T) = m.C_p.(T-40) ,T being the temperature of the mixture. The mixture is assumed to behave ideally.

C_p values are calculated at respective temperatures of 100 and 40 degrees Celsius for H₂ and natural gas. Natural gas mixture C_p is calculated by adding the product of mole fraction of constituent gases and their C_p at the given temperature.

 

The temperature of the mixture of recycle H₂ at 100 degrees Celsius and natural gas at 40 degrees Celsius is calculated as 42.2384 degrees Celsius.Please have a look at the attached thumbnail.

 

How do we preheat the mixture of natural gas and H₂ to the specified HDS inlet temperature ? 

 

Attached Files

•  mixture.PNG   33.69KB   1 downloads

[PingPong]

That is the amateur way to estimate the mixture temperature.

 

More professional is to calculate the enthalpy of both gases, add them together, and determine what temperature of the mixed gases gives the same enthalpy. Similar sort of calculation as was used to determine WGSR inlet temperature.

 

How do we preheat the mixture of natural gas and H₂ to the specified HDS inlet temperature ?

That is for you to determine later.

But first you need to determine how much preheat duty is needed.

[ravindra@096]

The temperature of the mixture of natural gas at 40 degrees Celsius and H₂ at 100 degrees Celsius comes out to be 42.24639 degrees Celsius using enthalpy difference method. 

 

The natural gas preheat duty comes out to be 89.03128805 kW (92.69120381 - 3.659915768).

 

Now what I reckon is the mixture after passing through HDS and ZnO bed moves to the convection section of the  reformer furnace to achieve the XOT of 550 degrees Celsius which is the reformer inlet temperature.

Attached Files

•  NG preheat duty.PNG   35.76KB   0 downloads

[PingPong]

89 kW to preheat up to 370 ^oC is correct.

 

But that is only part of the total required preheat, you still need to calculate the part from ZnO to XOT.

[ravindra@096]

The enthalpy of mixture at 370 degrees Celsius is 92.69120381 kW and at 550 degrees Celsius is 155.6321235 kW their difference being 62.94091969 kW. That means we need total 152 kW (89 + 63) in order to preheat the mixture from the 42.2463 degrees Celsius to XOT of 550 degrees Celsius.

[PingPong]

That is not correct.

 

I suggest you read my messages on page 5 once more.

[ravindra@096]

The ZnO bed effluent at 370 degrees Celsius is mixed with the required amount (61.47 kmol/hr) superheated steam at 385 degrees Celsius and the mixture (whatever the mixture temperature is) is again heated to XOT of 550 Degrees Celsius. It is the mixture of process steam and natural gas which has to be heated to XOT and not the natural gas (mixed with recycle H₂) alone.

 

Is any temperature drop across the HDS and ZnO bed required to be taken into account or the temperature be taken as 370 degrees Celsius at the ZnO bed outlet ?

[PingPong]

If everything were perfectly heat insulated there will be no temperature change over HDS & ZnO as there is in this case no enthalpy of reaction.

 

In reality however all piping and equipment will loose some heat. That is not only the case for the HDS & ZnO but also for the WHB, the WGSR and all other equipment and piping. Also the radiant and convection sections of the furnace loose some heat. Moreover there is a small leakage of cold ambient air into the furnace that affects the real flue gas temperature.

 

For the moment I suggest you neglect all that. You can try to estimate that later when finalizing the process scheme and heat balance.

[ravindra@096]

The enthalpy of Natural gas at 370 degrees Celsius is 92.69120381 kW and the enthalpy of 61.47 kmol/hr superheated HPS at 385 degrees Celsius is 974.84995104 kW.

 

Now the problem is, the HPS cannot be considered to be in an ideal state at 45 bara whereas the natural gas mixture at 370 degrees Celsius  can be. If the HPS is considered ideal and the Enthalpy difference and iteration method used to compute the mixture temperature, it results in a very high mixture temperature(exceeding 1000 degrees Celsius).

So I reckon there is another way to calculate the mixture temperature since one of the mixture constituent (HPS) is non ideal at the stated conditions.

[PingPong]

the enthalpy of 61.47 kmol/hr superheated HPS at 385 degrees Celsius is 974.84995104 kW.

That is not correct.

 

Read my message #124 again.

[ravindra@096]

The enthalpy of natural gas at 370 ⁰C = 92.69120381 kW

The enthalpy of HPS at 385 ⁰C and 45 bara = 191.2848688 kW

The addition of above enthalpies = 283.9760726 kW

 

The mixture (HPS + NG) temperature at which the enthalpy summation equals the enthalpy of the NG at 370 ⁰C plus enthalpy of HPS at 385 ⁰C = 380.4741 ⁰C

 

The enthalpy of natural gas at 550 ⁰C = 155.6321235 kW

The enthalpy of HPS at 550 ⁰C and 45 bara = 310.1592558 kW

Heat duty for heating the mixture from mixture temperature to 550 ⁰C = 465.7913793-283.9760726 = 181.8153 kW

 

[PingPong]

The mixture (HPS + NG) temperature at which the enthalpy summation equals the enthalpy of the NG at 370 ⁰C plus enthalpy of HPS at 385 ⁰C = 380.4741 ⁰C

That is not correct.

The mixture of natgas + recH₂ + HPS with an enthalpy of nearly 284 kW will have a much lower temperature.

 

The enthalpy of natural gas at 550 ⁰C = 155.6321235 kW

The enthalpy of HPS at 550 ⁰C and 45 bara = 310.1592558 kW

It is not clear to me what you are doing here.

The enthalpy of the reformer feed at XOT has already been calculated by you weeks ago. Look that up.

Edited by PingPong, 21 February 2018 - 10:40 AM.

[ravindra@096]

How we do we calculate the mixture (HPS + recycle H₂ + NG) temperature then ? The NG plus recycle H₂ is at 370 degrees Celsius and the HPS at 385 degrees Celsius.

 

Is the (NG+ recycle H₂) enthalpy at 370 degrees Celsius correctly calculated ?

Is HPS enthalpy at 385 degrees Celsius not correct ?

[PingPong]

The 284 kW for the mixed feeds is correct, but the mixed temperature will be much lower than 380.47 ^oC.

 

I can't see what went wrong in your calculation of the mixed temperature.

[MurtazaHakim]

The mixture temperature is 354.01 degrees Celsius when the temperature value is varied in the table created weeks ago for calculating the reformer inlet enthalpy at 550 degrees Celsius (478 kW). However when we calculate the enthalpy separately at a common temperature x for natural gas and HPS (from steam table) and add them later,we do not get the desired mixture temperature.

 

The heat duty obtained for heating feeds from 370 to 550 degrees Celsius is 478.5397221-283.9757117 = 194.5636495 kW

 

The enthalpy of mixture at 550 degrees Celsius has been calculated very early in the heat balance however when we calculate the same enthalpy at 550 degrees Celsius separately for NG and HPS (from steam table) and add them,the resulting summation is around 465 kW and not 478 kW. What could be the reason for that 13 kW drop in enthalpy of the mixture at 550 Degrees Celsius

 

What would be the next move if the above reported feeds mixture temperature and heat duty are correct ?

 

EDIT: The figure in bold is corrected.

Edited by MurtazaHakim, 22 February 2018 - 08:28 AM.

[PingPong]

Once again: HPS at 385C/45bara is not an ideal gas.

As you can verify in a steam table it has the same enthalpy as steam at 346 ^oC and nearly zero pressure (ideal gas).

 

If you would let the HPS down to nearly zero pressure its temperature would drop to 346 ^oC. That is the Joule-Thomson-effect.

 

For the natural gas at 370 ^oC it feels like it is mixed with steam at 346 ^oC and therefor the mixture temperature is 354 ^oC (assuming the natural gas and the mixture were ideal gases). The real mixture temperature is more like 365 ^oC because due to the high steam content in the mixture it is not really an ideal gas. In other words: if you would let the mixture at 365 ^oC down to nearly zero pressure its temperature would drop to 354 ^oC (JT-effect).

 

HPS at 550 ^oC is not an ideal gas so if you heat NG and HPS separately to 550 ^oC and then combine them the mixture temperature would be lower than 550 ^oC. To bring the mixture to 550 ^oC would require additional duty, roughly that 13 kW.

The heat duty obtained for heating feeds from 370 to 550 degrees Celsius is 478.5397221-283.9757117 = 193.5636495 kW

You mean 194.56 kW.

What would be the next move if the above reported feeds mixture temperature and heat duty are correct ?

The preheat duties that you calculated are good enough for the moment. Later you can try to include effect of non-ideality if you would still have time and energy for that.

 

Now you need to look at available waste heat in flue gas and WGSR effluent, and determine how you can use that for feed preheat, steam superheat, BFW preheat, additional HPS generation for export (and consequently more BFW preheat) and, if you have waste heat left, consider air preheat.

[MurtazaHakim]

The enthalpy of WGSR effluent at 420 °C is 422.6 kW and at the condenser temperature (45 °C) is 20.2232 kW which provides around 402.3768 kW  of heat.The enthalpy of flue gas (by burning FG + tail gas) at 1000 °C is 1449.997201 kW. 

 

What would be the outlet temperature of flue gas at which we must find its enthalpy and the difference of both would then give us the available heat for fulfilling the duties ?

Edited by MurtazaHakim, 22 February 2018 - 09:16 AM.

[PingPong]

It's not as simple as that.

 

At any point in the heat integration the hot temperature has to be higher than the cold temperature.

 

You should be familiar with Pinch Technology.

[MurtazaHakim]

The pinch analysis requires the process streams to be divided into source and sink. In our case, WGSR effluent and flue gas happen to be the sources and NG+recycle H₂,mixed feed and BFW appear to be the sinks. The method I came across requires a table to be created having the stream,its supply temp.,its target temperature,the available heat it possesses and the final parameter is kW/K which is the heat needed to be delivered/received in order to achieve the target temperature. Comparing the kW/K for source and sink should give the waste heat used. If dimensional analysis is performed we get the same kW/K by having (C_p*F)/3600 where C_p is in kJ/kg.K and F is the mass flow rate in kg/hr. The dimensional analysis form can be used for computing kW/K of sink since the sink flow rates are known and the values of kW/K of both (source and sink) could be compared. The Cp value is however a function of temperature and it would thus be difficult to have an average C_p over the temperature ranging from supply to target.Please see the attached thumbnail. I have arbitrarily divided the flue gas enthalpy into 1000 and 500 kW for the stated streams since one stream (NG + Rec. H₂) had lesser flow rate.

 

 Am I heading in the correct direction ?

Attached Files

•  Pinch analysis.PNG   6.79KB   1 downloads

[PingPong]

I have arbitrarily divided the flue gas enthalpy into 1000 and 500 kW for the stated streams since one stream (NG + Rec. H₂) had lesser flow rate.

I have no idea what that means so let's assume that it is not correct.

 

You did not mention steam superheat duty.

 

If you want to recover as much waste heat as possible, as you stated a few weeks ago after consulting your supervisor, then you need to generate additional HPS and superheat it. That means also additional BFW preheat duty required.

 

You may also consider combustion air preheat.

 

How you are going to do all that is up to you. You can do it using Pinch Technology, or a simpler approach, or just using common sense. This is your project so I am going to let you figure it out yourself for the coming week or so.

In any case it is handy to make graphs with T on horizontal and H on vertical axis for individual streams. Or composite curves if you use Pinch.

[MurtazaHakim]

The flue gas is utilized to heat two streams,(NG+recycle H₂) from 43 to 370 degrees Celsius and mixed feeds from 365 to 550 degrees Celsius. How would the available heat in flue gas be distributed in that case ? 

 

We are interested in heat integration using pinch analysis. How do we begin the pinch analysis ? I mean how do we generate the composite curves ?

[PingPong]

I mentioned Pinch because I thought you might already be familiar with it. The internet is full with info on it, for example google: pinch technology composite curves. Use also google images and youtube.

 

However if you are not at all familiar with Pinch, then I think you better use common sense and trial-and-error to figure out how to make best use the available waste heat. In that case, to keep it simple, use the WGSR effluent only for BFW preheat so that you can concentrate on flue gas heat for all other streams.

Some time ago I posted here a link of another SMR topic that may give you some ideas.

[ravindra@096]

For the cold stream, should we select 65.244 kmol/hr BFW at 30 degrees Celsius supply and 242 degrees Celsius target temperature (saturated liquid at both T_s and T_t) for receiving the heat of WGSR effluent ?

 

The 65.244 kmol/hr was calculated on the basis of estimations followed in the preceding discussions related to SD and WHB.

 

Should the flue gas be considered as a hot stream (its target temperature is unknown) ? We have so far considered two hot streams namely reformer effluent with T_s of 850 °C , T_t of 353 °C and WGSR effluent with T_s 420 °C, T_t 45 °C.

[PingPong]

For the cold stream, should we select 65.244 kmol/hr BFW at 30 degrees Celsius supply and 242 degrees Celsius target temperature (saturated liquid at both T_s and T_t) for receiving the heat of WGSR effluent ?

Keep BFW preheat using WGSR effluent for last. Until you decided how much additional HPS you can/want to generate and superheat in the furnace convection section you can't calculate how much BFW you need to preheat anyway.

 

Should the flue gas be considered as a hot stream (its target temperature is unknown) ?

Yes, it is a hot stream without a real target. You can take as much heat from it as you need, but you do not want a stack temperature below say 150 ^oC.

Again: look at that other SMR topic for ideas.