LNG pipeline calulation of coolling down effects at start up

Posted by: kk15

LNG pipeline calulation of coolling down effects at start up - 09/24/15 05:34 AM

Dear Members.
We are conducting stress calculation for LNG pipeline. It Caries LNG from Jetty Unloading arm to LNG storage Tank total run of pipe is 1.5KM with expansion loops.
Now we have requested to study the case below.
During startup before LNG placed in to the pipeline to be cooled down to -160degC by using Nitrozen gas. This gas will be enter in to the pipeline at the jetty. during that time at the other end of the pipe is at ambient temperature 21degC.
Pipe size DN400, 4mm thick and 316L and insulation thickness is 170mm.

I am using thermal bowing delta T as 139deg (160-21) to calculate the support loads and thermal stresses in the pipeline.
Since I am i am doing first time may i Know is it the correct way to consider this case.

Can any one please advise me.

Best regards
KK
Posted by: MoverZ

Re: LNG pipeline calulation of coolling down effects at start up - 09/24/15 07:04 AM

For the nitrogen filled case ....

Why would you expect thermal bowing to occur ?

The nitrogen gas should cool the periphery of the pipe fairly uniformly, so there should be no great difference in wall temperature across the pipe and thus, no bowing.

The temperature range you quote looks ok for general thermal stresses in the pipe, except, assuming it's in Norway, the ambient could be on the high side.

When LNG filling .....

If you ran your LNG liquid into a 'warm' pipe without completely filling it, you would expect some serious thermal bowing.
Posted by: RK

Re: LNG pipeline calulation of coolling down effects at start up - 09/24/15 07:10 AM

Is the thermal bowing temp value is mentioned by process team? or you are assuming them?

Regards,
R.K.
Posted by: kk15

Re: LNG pipeline calulation of coolling down effects at start up - 09/24/15 07:21 AM

thanks for reply.
process team requested to check dela T as140, 100, and 80.

br
kk
Posted by: MoverZ

Re: LNG pipeline calulation of coolling down effects at start up - 09/24/15 08:57 AM

What exactly is causing thermal bowing according to your Process engineer ?

Is it the nitrogen gas ?
Posted by: kk15

Re: LNG pipeline calulation of coolling down effects at start up - 09/24/15 09:15 AM

yes it is Nitrozen Gas,
Posted by: MoverZ

Re: LNG pipeline calulation of coolling down effects at start up - 09/24/15 09:35 AM

So I repeat my question above .... how can a flowing gas create a temperature differential across a pipe diameter, in order to cause thermal bowing ?
Posted by: kk15

Re: LNG pipeline calulation of coolling down effects at start up - 09/25/15 02:33 AM

now I come to know from process team they also had plan to use liquid nitrozen insted gas.
Posted by: MoverZ

Re: LNG pipeline calulation of coolling down effects at start up - 09/25/15 03:28 AM

I guess you will just have to deal with bowing during testing if liquid nitrogen may be used.

Since it's a jetty pipe, it must be likely that the introduction of very cold liquid will be repeated every time the line is used, so fatigue may be an issue ?
Posted by: kk15

Re: LNG pipeline calulation of coolling down effects at start up - 09/25/15 08:06 AM

OK thanks for info,
Yes we are considering 480 cycles for 40 years for fatigue calculation.
Posted by: RK

Re: LNG pipeline calulation of coolling down effects at start up - 09/28/15 12:00 AM

Personally, I don't feel there will be a thermal bowing for this line. Check with process team once again.

Yes, Fatigue is main thing to be checked for these lines.

Regards,
R.K.
Posted by: MoverZ

Re: LNG pipeline calulation of coolling down effects at start up - 09/28/15 07:17 AM

RK ... as KK15's problem has evolved and been described, it seems like a classic situation to expect thermal bowing to occur.

Rather than just offering your personal feelings that bowing would not occur, I think some more tehcnically based justification would be helpful.
Posted by: RK

Re: LNG pipeline calulation of coolling down effects at start up - 09/29/15 01:13 AM

MoverZ,

For Liquid nitrogen as testing media and nitrogen gas a an process fluid, there will be bowing if you are not allowing enough time to adjust to the wall temperature.
If we allow the time frame for temperature adjustment, then there will be no thermal bowing effect..This is what I understood from our previous projects.

But again, I said this is my personal opinion because we followed this in one of our privious project. This is the process engineer's job who has to clarify (which I said in my earlier reply as well)

Regards,
R.K.
Posted by: kumar73

Re: LNG pipeline calulation of coolling down effects at start up - 09/29/15 09:34 AM

Hi Kk,

During periods when horizontal pipelines are partially filled with LNG during filling or draining operations, temperature differentials between top and bottom of the pipe up to 60°C may occur. This will then lead to “Cryogen-ic bending” with consequently extra stresses, support loads and angular deflections and rotations, particularly at the elbows of the pipeline. Stresses due to this cryogenic bending or bowing are to be classified as displacement stresses.

In the calculation, the temperature differential will be assumed to be linearly distributed from top to bottom. Temperature differentials for the various parts of the system shall be considered. However, the design shall al-low for a temperature differential of 60°C. Weight has a hold down effect on the cryogenic bending. In evaluating the cryogenic bending, the line shall be considered to be filled only half full with LNG

Supports uplift during cryogenic bending shall be limited to 50mmn and will be controlled with 100mm high guide thrusts.

Thermal Bowing Delta Temp +60.000 C

1. W+T1+P1 OPE Operating case at operating temperature + Design pressure - Operating case
2. W+P1 SUS Sustained case + Design pressure - Sustained Stress
3. T1 EXP Operating temperature including 60°C of Cryogenic delta T - Thermal Stress Check.

Important points.

For supporting use the cold shoes and hold down frames.
check the surge loads.

provided the limit stop at the loops.