Design – Klamath Cogeneration Plant

W501F igniter cable failure

Klamath Cogeneration Plant
Iberdrola Renewables

500-MW, gas-fired, 2 × 1 combined-cycle cogeneration facility located in Klamath Falls, OR

Plant manager: Ray Martens
Key project participants: Greg Dolezal, maintenance manager Bruce Willard, operations and engineering manager Sig Gonczeruk, senior maintenance technician


For years we have fought the battle of our igniter cables melting at the igniter connection. It became so routine that we developed a plan to replace these at every maintenance outage to ensure reliable operation until the next outage. The root cause was the excessive heat surrounding the igniter. The insulation can sag, cause an oven effect, and bake the igniter. The igniter cable has a Teflon® end on it that fits into the igniter itself. This Teflon® and silicon rubber seal both melted in the igniter and made it impossible to separate the two, essentially destroying them both.


The goal was to try to keep the igniters cool. The igniters are located at the bottom of the machine so keeping the insulation tightly secured can be a challenge. Pins attached to the combustor case are used to hold the insulation in place but these pins can fail or can be broken off by workers during maintenance intervals.We ran a test on one of our units. Two steel pads per igniter with tapped and threaded holes were welded to the case. A 6 × 7 in. steel box was fabricated from two pieces of channel iron and mounted to the pads around the igniters by using some all-thread. The insulation was then packed around the outside of the empty box. The test worked and the igniters within these boxes did not fail while others did. However, we felt it undesirable to weld on the case so we tried other ideas.

An extension was added to the igniter in order to move the cable connection outside of the insulation. This showed some promise but did not completely solve the problem on its own. Our box test was conducted using the two-layer system provided from the OEM (Fig 3). This insulation was old, had been removed and replaced many times, and was falling apart. We felt that a new single layer system would provide better protection and solve the issue.

Even after going to a new, single-layer insulation system we still had a dead space between the insulation and the case that would heat the cable. An attempt was made to stuff the gap with insulation in order to protect the cable connection, but it proved ineffective because it simply held the heat in (Fig 4). The single-layer manufacturer was contacted in order to build a box, similar to ours, out of insulation material. This proved ineffective as well because it attached to the outside of the single layer and could not provide a tight seal to the case. Heat would then bleed through and cook the igniter.

The final solution was to modify the single-layer pads to accept the steel boxes. The boxes were mounted around all four of the igniters. Insulation was stuffed around the boxes but nothing inside. We also modified the test boxes to include a 3-in. lip on the exterior to capture the single- layer insulation better. Since the igniter extensions showed promise, we simply purchased longer igniters directly from the manufacturer. The ambient air of the turbine enclosure has proven to keep these igniter cables cool enough to provide longer life.


The first test boxes were installed in spring 2008. The remaining boxes were added last spring on the second unit. We have not experienced any igniter failures in any of these boxed areas. We now spark check the igniters at regular maintenance intervals but do not wholesale change them. Additional benefits include the ability to remove the igniter without removing the insulation. The mounting bolts are readily available with a long extension. The operators like them as well. They are required to check freedom of movement prior to any start. This new system allows them to reach up and conduct the check without sticking their arm into the insulation or pulling on the cable.