Check out before you start up

In his monthly Field Notes column, Senior Editor Dirk Willard reveals that regardless of the preparation, check-out tests always bring out some surprises.

By Dirk Willard, senior editor

Share Print Related RSS

In his autobiography, “Portrait of a Chemical Engineer” (CP, July, p.7), Ben Horwitz described a drop test with acrylamide briquettes. This test was performed by someone dropping briquettes from a ladder. Not all start-up tests are quite so inelegant. Here are a few tests you might find useful in check-out prior to starting up a chemical plant:

1. Communication check - Disconnect the instrument or switch to confirm that you have the correct instrument. Sometimes, an alarm will be activated. For an instrument, use a calibrator to confirm alarm points. This method is useful for any transmitter, flow switch, pressure switch, etc. The communication check is not useful in diagnosing problems with the element communicating with the transmitter such as a flow element. Maintain a good supply of fuses for this test: accidents happen.

2. Bump test - Verify the direction of rotation by bumping the switch, physical or logical, to “on” or “auto,” to see if the equipment, e.g., a pump, agitator, responds. Shaft rotation is generally clockwise from the front of the pump, or agitator. Make sure the oil bowl is half-full and seals are set up correctly. Bump the starter only for a few seconds; large motors cannot be bumped.

3. Flow test - Before a plant can be operated, the flow loops must be verified. This is especially true for environmental equipment such as scrubbers. If a meter is unavailable, the flow can be calculated by measuring the change in level in a tank. Use this test to validate a curve for a pump or blower. A flow test during a start-up in the Philippines revealed a capacity problem with the scrubber pump caused by swelled gaskets and glue-globbed fittings.

4. Stroke test - The position of automatic valves should be confirmed using at least two points: 0 and 100%; control valves need at least three points. Doing this test usually requires removing the actuator cover to see the shaft move. Where position indicators are installed, they can be a form of communication check.

5. Pressure test -  This is a familiar test with a choice of pneumatic (air or nitrogen) or hydrostatic (water or process liquid). I find that pneumatic is the most convenient, provided you have a good test rig. Hydrostatic is safest but most difficult to clean up. Air may be dirty and clean up is sometimes required. Only hydrostatic is usually available for pressures above 80 psig.

6. Temperature Loops - In my experience, thermocouples and RTDs either work, or they don’t. Overseas, however, it is best to bench-test the element or wire. If the element is wired directly to the PLC, a voltage check is probably unnecessary. If a temperature transmitter is used, the voltage should be checked at the thermowell against the datasheet values provided by the vendor. Only a high-quality calibrator should be used for this test. For ungrounded, two-wire, installation, confirm that the shielding is grounded at the PLC chassis exclusively. Grounding and power supply irregularities should be checked for all new instruments since these are common problems.

7. Load test - This test is conducted more to verify the electrical system supporting a pump, fan, or blower than certify equipment performance. It is best used to assess vibration and to check current loading.

These tests look at individual equipment. Inevitably, the process must be proved. Water, or air are useful test media but be aware of differences in density and potential contamination. On one project, I had to reset the ranges on a magnetic flow meter because of a pump. Where hazardous chemicals are used as the media, there must be a plan for disposal.

Some equipment requires testing with the process running. Fans, blowers and sequenced equipment like packaging lines and lock hoppers typically fall into this category. These were always the tensest moments with one question on my mind, “Will it work?” There are some surprises. I once spent two sleepless nights fixing an untried box-erector while wadding knee-deep in bakers yeast. Regardless of the preparation, check-out tests always bring out some surprises.

Dirk Willard, senior editor
dwillard@putman.net

Share Print Reprints Permissions

What are your comments?

You cannot post comments until you have logged in. Login Here.

Comments

No one has commented on this page yet.

RSS feed for comments on this page | RSS feed for all comments