How to Find a Bad Spot in a Feedline Length - A Simple Technique

[K1JJ]

It's antenna erection time at JJ Radio! This is the continuation of a multi-year project. Over the last two weeks I've hauled up a new, homebrew 40M Yagi, triple stacked homebrew 10M Yagis, and now two homebrew 15M Yagis. Much more to do yet. Everything worked so far except for these 50' boom, 6el 15M Yagis at 45' and 90' high.

After finishing them, I was disappointed to find the coax end of the array had an infinite SWR.  I had already checked these Yagis individually on the tower with an MFJ-259B antenna analyzer and they were both 50 ohms, so it was somewhere in the complex matching or coax/hardline feeds.

Here's why it could be a nightmare to troubleshoot: Each antenna starts out at 20 ohms, gets stepped up to to 200 ohms with a T match, then down to 50 ohms with a 4:1 coaxial balun, then the two antennas tie together making 25 ohms. Finally, there is a 1/12th wave coaxial balun that matches the 25 ohms to 75 ohms for the CATV hardline which runs to the shack. Obviously, there are many places there could be a connection problem.  Where do I start looking for the bad connection?

Many of us are familiar with an expensive feedline analyzer scope that will show the actual bad spot on the feedline. I don't have one. But, did you know the MFJ-259B antenna analyzer can be used to do the same thing, using some simple calculations?

To find the exact spot for an open connection or short in a feedline, hook the analyzer to the coax end. Then sweep the frequency until you see a sharp dip of near 1-2 ohms. Find the lowest frequency this occurs. Mark down this frequency.  Now use the formula:  492/freq X velocity factor  X 1/2.   This will give you the distance in feet down the line where the open is. If it is a short, then multiply X1, not X 1/2.   (A short can be identified first by using an ohmeter, of course.)

So here's how it went with my situation -  it was an open, so:  492/ 3.12 mhz * .875 (vel factor of my hardline) *1/2 = 68.99'.   I measured up the line 69' and low and behold there was a connector that I had reused and did not check for a few years. I pulled it apart and found moisture had contaminated the inner pin connection, causing an open. I replaced it and the swr was back down to 1.2:1 on the total system.

If I had looked for the problem by trial and error, it might have required ripping up at least ten connections that were already weather sealed. It all depended on whether or not I found the bad connector in the beginning or near the end of the session...
Because of this simple technique, I was able to pinpoint the bad connector as a first choice, even though other connectors were within 3 feet of this one. Not bad, eh?

Save this info for future reference in case you come across a similar problem. (With this technique, the classic trick, "pin in the coax" would be a piece of cake to find)

73,
Tom, K1JJ
 

[WA1GFZ]

Tom,
It is very easy to build a TDR. Square wave source and a resistor monitored by a good scope. Feed the pulse down the line and look at the return time and divide the time by two X VF X C. I bet a generator with series resistor monitored by a scope would also work.  ain't tricity grand!

[kc2ifr]

I measured up the line 69' and low and behold there was a connector that I had reused

Tom, At the risk of sounding like a smart ass..........if one invests all that time and money in such and nice antenna system, why use a used connector?

[K1JJ]

I measured up the line 69' and low and behold there was a connector that I had reused

Tom, At the risk of sounding like a smart ass..........if one invests all that time and money in such and nice antenna system, why use a used connector?

Yep, I wondered that myself, Bill. I think when I realized all the work involved making the harnesses, etc,  I got lazy.  That bad coax connector was "new" two years ago and well sealed in an interface to the hardline. It looked so good on the outside I woulda bet $100 it was good inside - but nooooooo! 

Frank - I never thought of using that technique. It would require hauling the scope outside, but otherwise sounds do-able!

T

[WA1GFZ]

The good thing is you found it without ripping things apart......

[W3SLK]

Tom said:

But, did you know the MFJ-259B antenna analyzer can be used to do the same thing, using some simple calculations?

 Yep, and I used it at work to find a broken (of all things) transmitter cable, (instrumentation type). Most of our process transmitters are fed from a trio box, (the trio modules are built by GE/Fanuc). I had to prove to the electrician types that there was a bad spot in the cable they just ran. I'm not a big fan of anything MFJ makes but for the price, my -259B has saved me alot of pain and aggravation for the price.

[K1JJ]

I'm not a big fan of anything MFJ makes but for the price, my -259B has saved me alot of pain and aggravation for the price.

Glad yours works well there Mike.

Mine just blew out. What a pisser. I was testing a few antennas and switched the coax switch. At the same time it started raining and the slight rain static popped the darn 259B.  Now the freq counter reads 0000.

This is the third time in three years it's had to go back because of static breakdowns. The other two times were on clear days - the front end is just so sensitive.  Usually it's the front end diodes, but this time it appears to be the clock circuitry.

I'll just have to do bull work until it comes back.

T

[flintstone mop]

I think for $80.00 MFJ will replace/repair the defective part in your instrument.

fred

[Jim, W5JO]

Mine just blew out. What a pisser. I was testing a few antennas and switched the coax switch. At the same time it started raining and the slight rain static popped the darn 259B.  Now the freq counter reads 0000.

This is the third time in three years it's had to go back because of static breakdowns. The other two times were on clear days - the front end is just so sensitive.  Usually it's the front end diodes, but this time it appears to be the clock circuitry.

I'll just have to do bull work until it comes back.

T

That is the only thing about the MFJ that I don't like.  I am hesitant to even hook mine to an antenna when the wind is blowing much because of the static electricity.  Rainfall or a storm within 50 miles can do it.  Schottky diodes in the front end with other unidentified types are ithere for conpensation purposes.

I always ground the antenna before hooking it to the analyzer and so far after over 4 years, it still works.  Pretty good for a 250 dollar analyzer.  It is worth having repaired.

 

[Mike/W8BAC]

Some really good reading JJ. I love my 259 and use it daily when in the shack. I noticed shortly after I purchased it that connecting an antenna, wiggling the connector, or the weight of a coax hanging from the fitting caused an intermittent in readings. The first 259 went back to AES right away and the second acted up shortly after it arrived.

I found the N fitting flange is mechanically connected to a PCB with screws, nuts and washers. Tightening them helped for a short time. The copper surface of the PCB had tarnished and wasn't making good electrical contact with the chromed flange of the N fitting. I cleaned the copper on the PCB and the flange, brushed on some flux and soldered the N fitting to the board. NO more problems. Hope this helps somebody else. 73

Mike

[W1RKW]

I purchased my 259 used from a ham for a bargain as it was broke.  I fixed it.  I've blown it up at least 2 or 3 times in different situations.  If anyone needs theirs fixed let me know.  It's  a pretty simple device and easy to fix.

[WA1GFZ]

Bob
you and Tom Vu are almost neighbors.

[K1JJ]

Bob
you and Tom Vu are almost neighbors.

I already shipped it to MFJap. I think the problem is more serious this time, like a digital clock chip or something. I'll never use it during a rain storm again.

T

[Steve - WB3HUZ]

The next time you are at a fest, look for one of those Hewlett-Packard RF limiters. They had N connectors on each end and were an inch or so long. The least expensive versions were good up to 1 GHz (all you'll need). This should provide protection to the front end of the MFJ. I've used them on the front-end of S/As and RXs for year and never took out a front end due to overload or static.

[WA1GFZ]

Steve,
Do you know what is inside? Pin diode or pin in series with low c diode??
I would think it easy to duplicate. Heck a 1n4007 is usable as a pin.

[Steve - WB3HUZ]

Don't know for sure. Never took one apart. Probably not much more than that, based on the size. The hot lick was the flat passband. Let me see if I can get some info.

[WA1GFZ]

Heck even a transorb in series with low C diodes would work and not mess with the line Z.