7 replies to this topic

[dsvalenzuela2]

Dear Forum Members,

Our methanol storage tank in the tankfarm transfers methanol into a smaller tank in the Plant, we call daytank. The methanol from the daytank is supplied to spray nozzles in the Oxanol Plant. We are expanding, hence constructing a bigger 10,000 MT storage tank. We are considering removing the daytank for fire safety and simplification of process and instrumentation, hence supply to the spray nozzles would now come from the storage tank that is 1.1 km away from the Plant. Does this make sense? Following would result, please add if I overlook any.
1. Methanol is class 1B flammable liquid with Flash Point of 12 degC. Hence, keeping fire hazards away from Plant.
2. Avoiding installing another (bigger) daytank in the Plant, which also require fire safety appurtenances like nitrogen blanket, level/temp gauge, vent, etc.
3. Replacing the bigger transfer pump with smaller feed pump.
4. But, the pressure in the pipeline would increase.

Thank you very much.

Dante

[ankur2061]

Due to the low flash point (12 deg C) and low NPB (64.8 deg C) you will need to ensure that your vapor losses from the big 10,000 MT tak are minimized.

There are two ways are doing this:

1. Build a tank with an internal floating roof

or

2. Provide a vent with a in-line vent condenser connected to the normal vent line which will minimize methanol losses while allowing non-condensibles such as blanketing N2 to vent off.

The option 2 with a shell-and-tube type vent condenser with process vapors on tube side and chilled water of 8-10 deg C on shell side has been personally utilized by me to minimize methanol vapor losses specifically when ambient temperatures are high (typically 35 deg C or more).

Hope this helps.

Regards,
Ankur.

[Himanshu Sharma]

Ankur Sir ,

Can you please elaborate more on the above two options , i mean indicate the capacity above/below which option 2 or option 1 is more attractive.

My observations particularly on option 2 is rather intriguing as i have absolutely no experience in designing and operating a methanol tank.

Option 2 as Vapour recovery system is both Higher capex and opex option.Chilled water at 8~10 deg c means there will be a refrigeration system where this water will be chilled (indirect refrigeration),a water drum and a pump.This whole system will be in service 24x7.
Cost of this operation can only be justified in case of high value product recovery or statutory requirement of practically zero emissions as compared to rather significant fugitive emission from a floating roof tank.

Why cant we treat mthanol in similar way as Light naphtha and store them in a convenient floating roof tank !

I remember having a small methanol tank in Hydrogen generation unit as floating roof tank.

[DB Shah]

I have designed mehtanol tanks 2400 MT capcacity and other similar tanks of 4000 MT. Normally higher capacity tanks with such materials should be floating roof, but due to some other reasons I had to design 4000 MT of fixed roof. Major issues-
1. N2/Inert gas requirement (with API 2000 VIth edition the requirement will be very high compared to Vth edition), this will need evaluation of existing inert gas capacity, line sizing.
2. Referigration system requirement
3. Exchanger
4. Kindly refer post-http://www.cheresources.com/invision/topic/13935-reduce-the-emission-from-tank-partial-condenser-vent/ where I have uploaded the schematic for the system.

To sum up, for such a high storage capacity IFR will be prefered over fixed roof.

[ankur2061]

Himanshu,

Option 1 is more atttractive for large tanks such as above 3000 m³.

Option 2 which I had mentioned does not have an elaborate vapor recovery system. It was a for a tank of about 1000 m³. The vent condenser was a vertical, self -draining, in-line shell-and-tube type heat exchanger with chilled water as the cooling / condensing medium. The key word here is self-draining. The condensibles were self-drained by gravity from the vent condenser back to the tank. The vent line from the tank to the vent condenser was designed to handle 2-phase flow. The non-condensibles were vented from the outlet of the vent condenser. In this particular case, inert gas blanketing had not been provided. Besides the normal vent line, emergency blow-off hatches had been provided for the tank. It was a very old design.

Personally I would recommend a tank with internal floating roof for a large capacity such as 10,000 MT for a low flash point liquid such as MeOH. The initial cost of a tank with IFR would be much higher than a fixed roof design but it would be more than off-set by the operating cost incurred for continuous inert gas blanketing requirement of a fixed roof tank.

Regards,
Ankur.

Edited by ankur2061, 06 February 2012 - 11:18 PM.

[dsvalenzuela2]

Thank you very much for the quick response. The fixed roof tank, 120' dia x 40' high (12810m3), is located at 8 deg N latitude with average ambient temperature of about 30degC. Following are my issues:

1. Does it make sense (or more beneficial) having no more daytank, instead, the 80 psig methanol requirement of the Plant would be supplied direct from the storage tank that is 1.1 Km away?

2. What is the venting rate to use for sizing the PVRV, nitrogen blanket, etc.?

2a) Using the API 2000-2009 formula at http://webwormcpt.bl...estimation.html (Thanks to the those who worked on this. I wish somebody to send me copy of API 2000-2009, thank you very much in advance):
V_in,air = V_pe + CV_tk^0.7 R_i = 22 + (6.5 * (12810^0.7) * 1) = 4900 Nm3/h
V_out,air = V_pf + YV_tk^0.9 R_i = 566 + (0.32 * (12810^0.9) * 1) = 2158 Nm3/h

2b) Using API 2000-1998:
V_in,air = (0.94 * 22 m3/h liquid outflow) + (0.577 * 2452 m2 shell and roof area) = 1435 Nm3/h
V_out,air = (2.02 * 566 m3/h liquid inflow) + (0.577 * 2452 m2 shell and roof area) = 2558 Nm3/h

Ideally, it is the max, the 4900. But, imagine the capex and opex of this. So, can I still use with confidence 2558 Nm3/h?

3. We consider the nitrogen blanket primarily for safety purposes. What is the good reliable control for the nitrogen blanket? Is it the methanol vapor concentration inside tank, setting the instrument to less than the methanol LEL of 6%, say 3%? Or the oxygen content of the vapor space inside tank? If not, what parameter shall I use best?

4. Local regulation allows venting to atmosphere. Installing Vapor Recovery System using chilled water Heat Exchanger necessitates further study. So, how do I calculate the methanol evaporation loss with and without VRS using aforementioned conditions? The difference/recovery can be equated versus capex and opex of the VRS. If this would be feasible, I would be collecting the condensate in a separate tank (easy validation of calculations) which I will use in other products where water in methanol is not a problem (leak in HX is a potential source of water contamination). I wish somebody having methanol VRS share his data of actual recovery versus calculated.

5. The tank roof is not yet installed, hence the IFR can still be an option. Does anybody have an idea how much will it cost for this size of tank? Versus the capex and opex of the nitrogen blanket and the difference in vapor loss?

Again, thank you very much and have a nice day!!!

Dante

[ankur2061]

Dante,

The "Chemical & Process Technology" web page mentioned by you has an excel workbook for normal venting based on API STD 2000-2009, programmed by me and avaiable as a free download.

Since you have tried to make a comparison between the old and new API 2000, I suggest you carefully read all the threads of the post provided in the link below where there has been a great discussion on the normal venting rate calculation method adopted by the old as well as the new API STD 2000:

http://www.cheresour...normal venting

For estimating evaporation losses from your storage tank refer the article in the link below:

http://www.epa.gov/t...inal/c07s01.pdf

Hope this helps.

Regards,
Ankur.

[DB Shah]

1. Does it make sense (or more beneficial) having no more daytank, instead, the 80 psig methanol requirement of the Plant would be supplied direct from the storage tank that is 1.1 Km away?

The tank in the plant battery limit is preferred. We supply methanol from our storage area to plant located tank (we term it as shift tank) and then re pump to reactor. The idea is to give total control of process to plant battery limit persons, reasons being -
a. In case of pumping problem at storage location, main plant may trip. Whereas an auto start /stop can be given in plant DCS, this eliminates PLC installation at storage area control room.

b. In case of emergency if plant operator wishes to trip the feed pump the instrument cable will run 1.1 km from plant DCS to storage area, too costly. In case tank is in plant area during extreme emergency plant operator from control room can rush to field and stop the pump.

c. If you revamp plant capacity you do not need to expand day tank capacity. It is just a buffer tank whose residence time will reduce, and it does not matter. Hence I will suggest to use existing tank for plant feed with out any capacity addition. You may just have to check your pump/lines/CV for revamp flow.


2. What is the venting rate to use for sizing the PVRV, nitrogen blanket, etc.?
Use API 2000

3. We consider the nitrogen blanket primarily for safety purposes. What is the good reliable control for the nitrogen blanket? Is it the methanol vapor concentration inside tank, setting the instrument to less than the methanol LEL of 6%, say 3%? Or the oxygen content of the vapor space inside tank? If not, what parameter shall I use best?
It is the pressure of the tank you control with N2. Study a wonderful post by Art in this section where tank blanketing principles are explained at length with sketch.

Typical installation will be-
Tank design pressure = say 300 mmWCg
ERV (fire case,hatch) opens above 200 mmWCg
Breath out valve opens above 150 mmWCg
N2 Blanketing valve closes above 100 mmWCg
N2 Blanketing valve opens below 100 mmWCg
Breath in opens at & below -25 mmWCg
Tank design pressure (negative side) -50 mmWCg


4. Local regulation allows venting to atmosphere. Installing Vapor Recovery System using chilled water Heat Exchanger necessitates further study. So, how do I calculate the methanol evaporation loss with and without VRS using aforementioned conditions? The difference/recovery can be equated versus capex and opex of the VRS. If this would be feasible, I would be collecting the condensate in a separate tank (easy validation of calculations) which I will use in other products where water in methanol is not a problem (leak in HX is a potential source of water contamination). I wish somebody having methanol VRS share his data of actual recovery versus calculated.
Annual loss of vapour can be estimated by AP 42. Else following simple steps-
Loss-1 = Breath out of methanol saturated N2 at ambient temp due to pump in. Cool to say 20°C and find the methanol saturated at 20°C. Calculated condensed and hence recovered methanol. You may use Antoine equation.
Loss-2 = Actual thermal variation loss. Assume temp variation of X°C in Y hrs in a day hence X/Y =rate of breath out due to ambient effect.
Add Loss-1+Loss-2 and size your exchanger for this case.

5. The tank roof is not yet installed, hence the IFR can still be an option. Does anybody have an idea how much will it cost for this size of tank? Versus the capex and opex of the nitrogen blanket and the difference in vapor loss?

Check with your mechanical engg or fabricator if you can switch to IFR, I doubt. Capex Opex can be verified by you, we cannot assume N2 cost/referigeration cost at your end.