Don’t get zapped by spark testing

Understand the intricacies of checking for defects in a liner, advises Dirk Willard, contributing editor, in this month's Field Notes column.

By Dirk Willard, contributing editor

Share Print Related RSS

From the outside, most companies exude an air of confidence. They have standards and procedures; everything seems under control. Once you’ve worked at a company for a while, though, you find it just isn’t so. For instance, I was on my own when it came to validating the thermoplastic liner on a refurbished vessel. Without too much effort, I adapted a procedure from a maker of glass-lined vessels; the company didn’t have a validation procedure for glass, either.

My procedure called for a spark test followed by a vacuum test and pressure test. I ruled out ultrasound because I was told it wasn’t as rigorous as the spark test. I developed a grid pattern for the heads, shell and each special nozzle larger than 12 in. in diameter. For small nozzles, test in the center and leave it at that. Sparking has a few drawbacks, though.

Perform a spark test once — and only once; polymers age, chemicals attack chains and the material become less resilient. Use a grid with lines spaced from 9 in. to 15 in. apart. This avoids the risk of the brushes of the electrode going over the same territory too often, or missing a void or thin spot. Use a wax chalk that is easy to remove with a sponge, not a permanent marker that might damage the polymer.

Where you do the test is important. Work on a loading dock or a nearby area. There isn’t much point in moving the vessel through the plant — if the liner fails you’ll just have to move the vessel again. Plus, docks tend to be outside and ventilated. Spark testing can produce dangerous fumes. If your unloading area isn’t well-ventilated, bring in some fans or test outdoors.

Choosing the test voltage is the most critical decision you will make, even more important than the pattern. A good rule of thumb is 5,000 V DC for every mm of liner thickness for fluoropolymers and 2,000 V/mm for most rubbers. Verify the voltage with the polymer manufacturer and get buy-in from the vendor who lined the tank. If an outside contractor will conduct the test, make sure the vendor has used that firm before.

Liners are measured in mm, mills and gauge. A mill is 0.001 in.; 1 mill is 100 gauge. Most liners are from 2 mm to 5 mm thick; 1/8 in. (3.2 mm) is the thinnest practical liner in chemical processing.

It’s a good idea to prepare a few coupons of different known thicknesses for testing voltage. Properties, especially of amorphous materials like plastics, depend on manufacturing as much as on physical characteristics or chemistry. I remember years ago hearing of a comparison a valve manufacturer made between polytetrafluoroethylene (PTFE) and ethylene tetrafluoroethylene (ETFE). Although PTFE was superior to ETFE in permeability, it was harder to manufacture into a valve liner — it tore. The valve manufacturer found that it was easier to work with ETFE and valve liners made from it suffered fewer leaks. Because a material is stretched and compressed properties may change, so testing a few coupons makes sense.

Choose the lowest voltage practical for the polymer. Remember that too low a voltage renders the test useless while too high a voltage may damage or destroy the liner. Burn a few coupons, if you can, to determine the limit.

So, what should you look for while witnessing a liner test? First, make sure the brushes are constantly moving. The liner may be destroyed if the brushes stop in one place. You’ll want to record the voltage at each node on the grid and other physical data. Chances are you’ll see a spike in the current flow but you may see the actual arc if you watch carefully. A leak can be recognized from a white or blue arc that occurs because the voltage carried through to grounded steel on the other side. Although patching is possible, and necessary in some cases, I don’t recommend it. Mark the spot for the arc on the grid and pause. Look for patterns. If, for example, the liner failed at a particular point, see if there are similar ones elsewhere in the vessel. Check the grid for additional failures.

The particular liner produced an arc and so according to the contract failed its acceptance test. The vendor took the tank back. It passed the second test with a new liner.

Share Print Reprints Permissions

What are your comments?

Join the discussion today. Login Here.

Comments

No one has commented on this page yet.

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