[BC] AM transmitter lightning sensitivity issue

[Gregory Muir]

Within my station group I have a Harris Gates Five AM transmitter which exhibits an unusual sensitivity to local lightning storm events.  Upon the appearance of a storm, the transmitter will begin to automatically lower it's output power through the power adjustment range with each lightning strike until it finally goes off the air after reaching the lowest power setting.  It is obvious that the unit sees VSWR events which causes it to react.  But what is interesting is that it will react to the same events even when the storms are 5-10 miles distant from the site with similar results.

This transmitter is installed in a three tower directional array with a recently installed Kintronics phasor.  The transmitter behaved the same way prior to installation of the new phasor.  Prior to my assignment to this cluster, a past engineer had designed and installed a home brew restart circuit to cycle the transmitter back to air after it had shut down from a VSWR event (I know what you are thinking - try to fry the whole unit if a "real" VSWR event occurs).  The circuit was disabled.   But the net result is that our chief operator now has to babysit the transmitter each time a storm appears (he had to reset the transmitter 30 times last night within a 3-hour period as a storm system moved through the area).

Our tower array is situated in a low spot next to a river with excellent ground characteristics.  The radial system actually consists of a copper mesh that covers the entire site. Direct lightning strikes are conveyed to ground with no damage to equipment.  4-inch copper strap is used profusely throughout the system and each tower utilizes four straps to the ground system for lightning dissipation from the Johnny balls.  I have checked the static bypass chokes in the TU huts and all have good continuity to ground.  This indicates to me that the transmitter is reacting to fast induced transients that see a high impedance through the chokes and instead probably find a path across the Johnny balls at the base of each tower.  Since I have not been gutsy enough to stand at the base of a tower during a passing storm, I also have an additional thought that if arcing is not actually occurring with distant storms, the EMP coupled into the tower may be enough to momentarily alter the system impedance which still perturbs the transmitter VSWR detect circuitry.

Harris does have a tech note (AM-409-TLH) that addresses this issue through modification of the VSWR detect circuitry to increase the time constant of detection and reduce the VSWR sensitivity threshold.  I have been hesitant to implement it mainly because we presently experience occasional blown PDM amplifier fuses and/or FETs when the transmitter does experience a storm.  To bulletproof the VSWR detection circuitry may increase this failure rate.

>From what I am gathering from other sources, this "storm sensitivity" appears to be a characteristic of this particular transmitter.  But I wanted to throw this out there to see if others have experienced a similar problem, what they may have done to help minimize or solve it or offer suggestions as to what else I should consider.

Greg