Test pressure limits for flanges 3

[nhliew]

Good day all,

Earlier today we were instructed to conduct hydrotest on a pipework package for with ASME B16.5 Class 300 rating. Specified test pressure was 7320 kPag (1061 psig). Design code is B31.3 and flange material was A105.

Referencing ASME B16.5 Group 2.1.1, the 100 deg F pressure rating was 740 psig. Section 2.5 of the same standard states that the maximum test pressure is 1.5 times the 100 deg F pressure rating (740 psig x 1.5 = 1110 psig). To quote the standard, "Testing at any higher pressure is the responsibility of the user, subject to the requirements of the applicable code or regulation".

My question is, although there is a slight margin (49 psig) for safety in the specified test pressure, do you normally design pipework to that close of the maximum? I normally take 90% of the pressure rating as the absolute maximum design pressure. If the system is required to operate at higher than that, I will specify a higher pressure rating (eg, uprate the flange from Class 300 to Class 600). That way, I will have a 10% safety margin for any temperature/pressure gain that occur during hydrotest.

I believe that it is the responsibility of the designer to take into account any temperature/pressure gain during hydrotest. An outdoor system tested in situ (project in Malaysia, average daytime ambient 33 deg C) is bound to experience some temperature and pressure gain.

Would welcome any advise for future reference.

 

[BigInch]

You can test to 100%. Over 100% gets a bit risky.

Monitor the pressure-temperature effects to make sure they remain above the required minimum test pressure and below your maximum. Record the temperature and the volumes of water added, or released if necessary, and the corresponding increase or decrease in pressure and you can tell if there is cause to suspect a leak or if the pressure was simply due to a temperature change.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[ColonelSanders83]

If you have the time you could run a flange design analysis on the flange that you are performing the test on. The BPVC Section 1, Appendix 2, Rules for bolted flange connections with ring type gaskets apply.

This way you will know where you stand from a code standpoint. Since this is a hydro test, remember to replace the BPVC allowable stress with the B31.3 allowable stress under hydrostatic test.

Also remember that B16.5 flanges have been known to fail a stringent BPVC design analysis, so this may not help you at all.

Just my two cents worth.


Always remember, free advice is worth exactly what you pay for it!

 

[rneill]

It is a routine and standard industry practice to field hydrotest flanges to 1.5X their cold rated pressure. I've been doing this for many years and never had a problem.

I suspect (but can't say for certain) that the governing criteria for many flange designs probably has more to do with sealing against leakage rather than approaching allowable stress limits of materials. If this is the case, then you would not be risking yielding the materials but rather leaking the joint.

 

[nhliew]

Many thanks everybody.

BigInch, for accountability we always record the pressure & temperature readings using a digital recorder set to 5 second intervals. At the same time I attach a tiny relief valve at the low point, set to 1.5X pressure rating to bleed off the test medium in case of pressure gains. I also attach another needle valve for manual bleed.

Colonelsanders83, if the specified design pressure exceeds 90% of the flange pressure rating, we normally run a simple stress check to ensure nothing exceeding 90% of SMYS is applied to the flanges during hydrotest. However, in this project we were only in the position of erector, not designer.

rneill, to draw from your experience, I am interested to know if you carry out any of the following for hydrotest:
(1) Use a relief valve and another needle valve in parallel to bleed off some fluid if there is an overpressure;
(2) Check bolt torque or tension (JIT);
or are there any other ways of ensuring a safer hydrotest?

Appreciate your comments.

NH Liew

http://www.yuentat.com.my

 

[rneill]

nhliew

1) we would normally have two independent devices to monitor pressure (e.g., gauge and chart recorder) and a bleed valve to let of pressure and we would have someone monitor the system for the duration of the test. Don't normally have a relief valve except for pneumatic tests where it is required by B31.3.
2) we don't routinely check bolt torque or tension but may depending on the type of system and the risk of leakage (e.g., hot oil systems, sour gas systems). This raises the next question of what preload you want to achieve in the bolts, and how do you ensure it (torque is notoriously unreliable)?

Note: if you look at the latest B31.3 para 345.4.2, it requires you to test to 1.5X the component pressure rating so you are fully expected by Code to test a flange to 1.5X it's normal rating.

 

[BigInch]

And volume. Do you measure volume added or released?

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[rneill]

Normally, you use the bleed to adjust the pressure to the test pressure at the start of the test and then you often don't need to use it for the 1 hr test that we do (B31.3 plant piping). I wouldn't normally worry about the volume bleed off if the pressure did rise, just record the pressure before bleed off and the pressure after bleed off to ensure that there is no confusion in the records as to why the pressure dropped.

For a long pipeline system, there may be value in recording the volume bleed off in order to try and correlate it back to the pressure change but I'm not sure that this is common and I don't believe it is a requirement of the local code or regulatory authority.

There is always the question that sits at the back of my mind, if the pressure is rising due to temperature, how do you know for sure that this is not masking some underlying small leak in the system. However, for plant piping, if you are doing the required visual check of joints during testing and not solely relying on maintenance of test pressure, then this should not be an issue.

 

[BigInch]

You calculate the pressure rise expected by the temperature increase. If the pressure rise is not as much as should be expected for a given temperature rise, you've got a probability of a leak, or a bust in your calcs.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/

 

[nhliew]

Dear all,

(1) We always have two gauges, located at high and low points and a recorder in a hydrotest package. It's required by both the client and governing authority.
(2) As for bolt preload, for line class up to 600# we rely on torque values published by the client. I do agree that using torque as an indication of bolt loading is inaccurate. For this reason I've put into next year's operating budget to purchase an ultrasonic tension gauge. For line class 900# and above the client require calculation and hydraulic tensioning so that's not a problem.
(3) As for your note, I'll check for the newest edition of B31.3. I'm using 2004. I believe it's 2 versions behind.
(4) For aboveground pipework we do carry out visual inspection during hydrotest. Generally, a pressure/temperature calculation for the test medium forms a part of the test procedure. Using this, we pre-qualify the point in time when we need to bleed off or add some fluid. However, the typical procedure does not take into account thermal expansion/contraction of the pipe (nobody ever asked) and we are in the process of incorporating this aspect in our procedure for next year's qualification (got this idea when designing a liquid metering calibration tank in June).
In response to your comments, I don't record the volume being added or released, just a comment on the test records whenever that is done and why, i.e. "pressure increased by pumping (temperature drop from 30 deg C to 27 deg C)". Of course any addition and release is approved in situ by the client rep.
As a point of interest, how do you measure the volume of fluid added? Measuring the level drop on the holding tank is not very accurate. Is it sufficient to count the number of strokes by the pump and multiply that by the pump displacement? On the other hand, bled volume can be collected and measured in a jar with a certain degree of accuracy.
For buried pipelines, I believe it is even more tricky although the temperature remains fairly constant. I have not had the experience of working on such a project.

NH Liew

http://www.yuentat.com.my

 

[BigInch]

Yes count the strokes and multiply by piston volume.

Sometimes a tank level differential reading can be accurate enough, if you've strapped the tank and know the tank volume to level marks are good.

Not really more tricky, as more of the pipe is anchored by the soil and prevented from pressure and temperature expansion over much of its length, but the longer lengths and possible greater associated elevation changes make it more convenient to use up to the 100% limit to extend the test sections as far as possible. More still time is needed to be sure the pressure is stabilized and air is completely vented before final pressurization. Sometimes just short exposed test headers and 50ft of pipe is enough to change the average temperature, so more PT readings are sometimes necessary to assure stabilization along the length has been reached.

**********************
"Pumping accounts for 20% of the world’s energy used by electric motors and 25-50% of the total electrical energy usage in certain industrial facilities."-DOE statistic (Note: Make that 99% for pipeline companies)

http://virtualpipeline.spaces.live.com/