We are talking about the large cylinder that surrounds a tube. It is actually one aluminum cylinder within another, with insulating (dielectric) material in between. The purpose is to transfer RF energy from the tube anode to the output-tuning network without shorting high-voltage DC to ground. That direct current often is connected directly to the tube via a clip on the anode.

So there you are with a shorted blocking capacitor and no replacement on the shelf. (Why would the station have a spare when they fail only once in 20 years or so?)

My story goes back to a cold day in January, when it took a farm snowplow to clear a road to a transmitter site (Fig. 1). Such is life in Minnesota during the winter.

Fig. 2 shows the charred remains of a PA tube, a driver tube and a high-voltage blocking capacitor from a CCA FM2500B FM Transmitter. The normal tube appearance is of nickel-plating, not brown-red. The blower motor quit, but the transmitter kept running when the airflow switch stayed in the on position. The switch worked normally when tested a year before, so I can only assume it stuck from being on for a long time.

Fig. 2: These items have overheated.	Fig. 3: Hacksaw the outer cylinder.

NO SPARES

This transmitter normally runs 2,200 watts using a 3CX400A driver tube and 3CX3000A7 PA tube. The client keeps spare tubes on the shelf so that was not a problem, but who keeps an extra blocking capacitor? Since there was no spare, the only option to avoid two days of off-air time was to rebuild the original capacitor. This particular type of capacitor was not designed for rebuilding, but what else was there to do?

It�s not rocket science!

I took the capacitor to the shop and used a hacksaw to cut through the outer cylinder and just into the dielectric between the outer and inner cylinders.

Fig. 4: Note the arc-over spot.	Fig. 5: Clean to smooth rough edges.

After unwrapping and exposing the original dielectric, I found an arc-over spot. Now the real work began.

Plenty of cleaning was needed to get sharp edges and other blemishes down to prevent future arc-overs. A file and 3M Scotch-Brite scouring pads worked well here. I remember hearing of someone who replaced dielectric material without removing the sharp edges, only to have the same arc problem immediately again!

I found new Kapton dielectric material on the shelf at a nearby station and wound it onto the inside cylinder of the blocking capacitor; then the outer cylinder was put back on. This thin Kapton sheet is 4-3/4 inches wide and is used by Continental Electronics in its transmitter blocking capacitors. At less than $90 from Continental, it is worth keeping some on the shelf. A complete blocking capacitor, which includes the two metal cylinders, costs $450 to $1,500 depending on which transmitter and manufacturer you order one from. Kapton has a reddish-brown color, as opposed to the original clear-white, which was probably Teflon, in this case.

Fig. 6: Use some of the Kapton material you keep on the shelf.	Fig. 7: Measure capacitance to confirm it is right.

The original capacitor had six turns of dielectric sheet, but capacitance testing, using a Sencore Z Meter, showed that was one turn too many with this new material.

TESTING

The proof is in the capacitance of the assembly. Here the rebuilt blocking capacitor measures very close to the original 550 pf. Yes, I was able to determine the capacitance of the original shorted capacitor because testing was with just a few volts, rather than the normal 4,000 volts in the transmitter.

Two stainless steel hose clamps were used to hold the outer cylinder tight on the dielectric. Stainless is not ferromagnetic. I pointed this out in a December 2013 article in Radio World (Fix a Transmitter Tube Socket, radioworld.com, keyword Socket). This is important as anything that is attracted by a magnet in a high RF field will tend to vibrate at the RF frequency and sometimes melt. Use an ordinary small magnet to verify that nothing in the RF area is attracted magnetically.

Fig. 8: Here's the reassembled blocking capacitor.	and the blocker is back in the transmitter.

Fig. 9 shows the final assembly, back in the transmitter, just before high voltage was turned on. Yes, the outer parts of the blocking capacitor have an oxidized appearance. They are silver-plated and get to look that way from years of exposure to air. The oxidation does not hurt transmitter performance.

Success � no arcs and the operating parameters came back to normal.

This is a permanent fix for the problem and the blocking capacitor is more repairable if this should happen again in the future. It makes perfect sense.

Mark Persons, W0MH, CPBE, has over 30 years experience. He has written numerous articles for industry publications over the years. His website is www.mwpersons.com.