Heat Exchanger Evaluation - Pressure Test 1

[sgwhip]

I wish to pressure test 2 heat exchangers (that have been in service for many years) to justify their suitability for continued service, and to avoid replacing them. These are simply tubes attached to a flange that inserts into the oil sump of a machine, and plant water passes through the tubes to an open drain. They certainly need cleaning, but I am not at all concerned with their heat exchange capacity, as that is well established. I only wish that they will not leak water into the reservoir. Does anyone know how much pressure constitutes an adequate pressure check of this type of heat exchanger?

 

[rmw]

Depends upon what your goal is.

Do you wish to pressure test the tubes to induce the weak ones to fail on a pre-emptive basis, or do you just want to see if they will be safe in service.

If it is the latter, then I wouldn't exceed test pressure by very much more than the maximum cooling water pressure that these tubes can expect to see.

If it is the former, then you can test the tubes up to 1.5 times the design pressure. One of two things will happen. You will either fail any tubes too weak to take this pressure, or you will assure yourself of the integrity of the tubing that survives the test.

Or, pick some point in between the two values above (maximum cooling water pressure and 1.5 X design pressure.)

You know your situation better than we do, so you will have to use your judgement.

Sometimes Hx tubing is tested to failure because it is better than having contamination of the process of having to bring the Hx off line during production.

rmw

 

[sgwhip]

RMW, Thanks for such a fast response.

This testing is a maintenance function, so we are seeking reliability. Given that, your suggestion that we test to 1.5 times the design pressure seems the way to go. However, this is a proprietary part and I do not know the design pressure. The manufacturer's representative has already suggested that we should just replace the units, since they are old (I'm getting old, too, so I take umbrage with that method of evaluation). However, factory reps come in all flavors, so I am looking for an objective test, which they are reluctant to provide. I have no problem with that since it is safest for the manufacturer, and more profitable, to simply recommend new parts. But I thought there must be a "rule of thumb" pressure that heat exchangers that operate at very little water pressure could pass and be considered reliable against leakage. The machine is a vetical roller mill. It is not going to the moon, but it would be terribly inconvenient if the heat exchanger leaked, but it is also inconvenient to waste $14000 on 2 new ones for no objective reason.

 

[weeeds]

Depending on your capabilities, you may want to subject the tube side to a full vacuum. Put a vacuum gage on the tube side and watch for a decay in the vacuum. This will tell you if you have leaks. In the event of a decay in the vacuum, subsequent inspection of the tube side tubesheet should pinpoint any tubes with damage as you should see traces of oil. Plug those tubes and you should be good to go. This type of test is great since you do not have to remove the bundle from service. There are some precautions in dealing with vacuum such as the possibility of contaminating and damaging the pump so you need to be carefull, but this is a great diagnostic tool. For higher sensitivity applications we use Helium Leak Testing to support the vacuum test.
And a Merry Christmas to all!

 

[rmw]

Take care in cleaning the tubing. If the build up is old and hard, removal of the scale might open up pits that would leak.

I don't know the value of replacing the bundles, but there are services out there that can eddy current test the tubes (after cleaning, of course.) The eddy current test will tell you if you have wall loss and how much.

rmw

 

[abeltio]

According to TEMA stds. for class C heat exchangers:

C-1.31 Standerd Test
the HE should be hydrostatically tested with water.
The test pressure should be held for at least 30 min...""... The hydrostatic test press at room temperature shall be 1.5 times the design pressure, corrected for temperature

C-1.32 Pneumatic Test
When liquid cannot be tolerated as test medium... (abeltio: e.g. if a test failure will create water leaks into the oil sump) the exchanger may be given a pneumatic teset in accordance with the Code (ASME Sec VIII Div I)... The pneumatic test pressure at room temperature shall be 1.25 times the design pressure, corrected for temperature...

C-1.33 Supplementary Air Test
... Leaks may be located by halide probe, or by other means...

Per your description... there is little, if any, pressure in the system: outside the tubes: the oil sump (~ atm press) and the tubes are connected to an "open drain" (~ atm press).

Without a significant pressure in the system:
The main concern would be corrosion caused by the water... if the oil side is corroded there is something very bad happening in that engine.

If there is no possibility of pressurizing the tubes to an unknown design pressure, because the OEM would not release the information...
What i would do is:
1. borescope the tubes to detect any defect, e.g. oil inside the tube will indicate a punctured tube, corrosion by water is obvious to visual inspection.
2. eddy current inspection or an ultrasonic testing of the tubes to map any wall thining...

Another possibility is to prepare a device to do a pressure test tube by tube, from your description there is not SHELL outside the tubes as we know it...
and use the design pressure of the flanges in the system at the operating temperature... e.g. if the design temperature is 300F a 150# carbon steel flange will be rated to 210 psig.

the problem with the pressure test tube-by-tube is that the experiment is part of the problem.
i.e. it is difficult to eliminate the leaks from the device used to do the test.

Unless the tube is definitely punctured, and u cannot raise any pressure at all... it is cumbersome and time consuming.

the device for the tube-by-tube is a tube with threaded ends that goes inside the tubes and has two headers with gaskets... very messy.

So, summarizing:
1. visual inspection by borescope
2. non destructive testing
3. pressure test if required

HTH



saludos.
a.

 

[sgwhip]

abeltio,
I did not think of using the pressure ratings of the flanges to determine the pressure rating of the heat exchanger...how incredibly practical! Though we are in the middle of this rebuild, it is not the critcal path of the larger project, and we await parts anyway. I'll get on this right after the holidays.

Thank you all for your input. I knew I'd get some good ideas here.