Spec break again

[AtleK]

A minimum flow from a high pressure pump back to a low pressure vessel is controlled with a minimum flow valve. This pressure loss over the valve is 90 bar. Upstream the valve we have #1500 piping class and downstream it will be #150. We have a spec break over the valve.

Now, we need double block and bleed valves downstream the flow valve, to isolate the pumps in case of maintenance. The equipment is placed on a skid and the skid will be shut down when maintenace is needed. The isolation valves are there to block out the downstream low pressure system.

Do we need to place the spec break downstream these isolation valves (which are locked open when the skid is in operation) or is it ok to have the spec break between the flow valve and the isolation valves?

 

[pennpiper]

Down stream of both B&B valves

 

[SeanB]

I disagree. As long as the valves are locked open, the spec break is fine where it is.

 

[BigInch]

If your spec break is at the minimum flow valve, theoretically anything downstream should be 150#. The question is, is it protected from overpressure. You don't say how you're doing that. IMO you should have fail closed action on the minimum flow valve and a PSV located somewhere in the downstream system such that no point of the 150# piping, from the min flow valve to the tank and beyond, can never experience overpressure from the pump.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[SeanB]

You are absolutely correct biginch - I should have stated the same in my post.

 

[BigInch]

That goes for both of us.

I should have added,

I really don't like locks between overpressure protection devices and ANYTHING they are supposed to be protecting.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[AtleK]

The vessel, and therefore the piping downstream the flow valve, is today protected by a PSV. The block and bleed valves are retrofit work and I hoped to be able to use 150# valves provided they were kept in a locked open position by a key lock system. They would then never be closed without a very good reason and under strict surveillance when the high pressure device was shut down.

My initial reaction was to purchase 1500# block valves, and from the response from you guys, there seem to be no way out of that.

Btw, do you have any referance to any API, ASME or other code for where and when spec breaks may be set.

 

[BigInch]

To be safe in this case, If the piping before the DDB and the DDB itself are 1500#, and the PSV is somewhere downstream, on the tank for example, you would have to set that PSV for DDB outlet pressure = about 275 psi (including a 10% overpressure transient allowance - pressure drop in your lo-press piping (at maximum possible flow) between your DDB outlet and the PSV. The problem here is being sure that the outlet of the DDB never reaches above that 275 psig. Given that flow control valves always are assumed not positive shutoff, full pressure can reach to the DDB, which, if open, pretty much says that maximum pressure can also reach the 150# piping downstream. That is why I prefer not to rely on the pressure drop downstream to the PSV to protect low pressure piping upstream. If I was doing this, I would place a PSV set at 275 psig right after the min FCV. That would place the PSV actually on the high pressure piping between the FCV and the DDB. This arrangement ensures that low pressure piping is always protected regardless of what happens downstream of the spec break and no need to depend on the lock for the DDB.

As far as locating spec breaks, you can put them anywhere you like. The trick is, as above, only insuring the lower pressure side is protected from the high side.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[SJones]

Question from a non-process engineer: would there not be a HAZOP of this set up to identify all of the issues?

Steve Jones
Materials & Corrosion Engineer

http://www.pdo.co.om/pdoweb/

 

[BigInch]

One would hope so, but it pays to know them when you start.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[bosse1950]

Questions - What type of pressure letdown valve do you have? What size & schedule piping upstream & downstream? The reason for the questions that another issue exists - one of internal errosion due to extreme pressure drop & relative increase in velocity. Has this been considered in your case?

 

[BigInch]

bosse, as you note the pressure drop across that valve is a very legitimate concern. It will probably require special anticavitation and noise trim, which might not help a lot, as its a very high pressure cut to do with one valve. Any sand and the valve won't last 3 months. I have put a restriction orifice ahead of valves like this, so the valve wouldn't have to take the whole press cut by itself.

I didn't want to get into all of that, but as long as we're at it, why not?

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein

 

[AtleK]

We have had a Hazop on the system, but the isolation issue didn't come up until later as a client request.

BigInch: You say: "If I was doing this, I would place a PSV set at 275 psig right after the min FCV. That would place the PSV actually on the high pressure piping between the FCV and the DDB. This arrangement ensures that low pressure piping is always protected regardless of what happens downstream of the spec break and no need to depend on the lock for the DDB."

That would have made it possible to place the spec break at the FCV and keep all downstream piping and DBB 150#, because I would size the PSV to take the entire flow if one of the block valves were closed or if the FCV should fail. I would also need a PSV on the tank to cater for a fire case, which is the dimensioning case if the recycle is taken care of up front. No, I think it'll be better to keep the entire recycle system (FCV, DBB and piping) 1500# and keep the PSV on the tank.

The FCV is three stage valve, quite sophisticated, with a plug especially designed to take the high pressure drop. The medium is LNG, i.e a mix of light hydrocarbons which is more gentle than e.g. water wrt cavitation, since it has no sharp boiling point definition. The pipe and valve is 6" with valve inlet velocity below 5 m/s. The velocity increases internally in the valve, as it will be a two phase flow at some places in the valve.

I hope we have considered the cases correctly.....

 

[BigInch]

Even better. I had assumed previously that you wanted the 150# pipe for some unknown reason. The big costs are the valves, so a bit of strong pipe isn't too much more $. Excellent.

http://virtualpipeline.spaces.msn.com

"We can't solve problems by using the same kind of thinking we used when we created them." -Albert Einstein