TheRadioBoard

I've got a 1m run of RG-58/U cable coming off the coupling loop of my magnetic loop antenna. The 1m RG-58/U is terminated in a male RCA connector, followed by a female-female adapter, connected to another male RCA connector that is connected to 10m length of RG-58/U, terminated in another male RCA connector that goes through another female-female adapter, that connects to a male RCA connector that is soldered to a 10cm cable of unknown impedance (probably 75 ohms, as it was intended for TV) that is terminated in a non-threaded F connector that I have bent out of shape so that it mates with the BNC connector on my QRP transmitting rig.

It works fine on receive. When I get around to transmitting, is this patchwork of connectors and cables, some with unknown impedances, going to cause problems? Or is it close enough (mostly 50 ohm with some weird connectors and a short 75 ohm section) for amateur work?

Good grief. Can you get a single piece of RG-58 and terminate it with a BNC male connector? The bent-F onto a BNC would curl my hair if I had any left. This kluge will probably work -- Rf gotta go somewhere! -- but I'd try for something more in line with good practice.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Sorry about that. Since recently moving the electronics stores are a little inconvenient to visit, so I visited the local dollar store and made do with the consumer electronics they had available. It appeared to me before the purchase that the TV antenna cable connector would fit the BNC connector, but only after buying and opening the package did I realize the mismatch. So I made it fit.

I've seen the use of RCA plugs proposed for QRP work and I kind of like them myself for cost, ease of soldering, and availability.

You probably can get away with using 75-ohm CATV coax. As your antenna is tunable, you might find a setting that will reflect 50 ohms back to the transmitter. Mismatch losses in a short run of coax won't be ruinous.

As for RCA phono plugs, they used to be available with silver plating. The common nickel-plated ones today melt if you hold the iron on for more than a few seconds. I also don't like them because they can be accidentally pulled out. Heathkit used them for rf sometimes, but most people replaced them with SO-239s, which were the standard ham HF rf connector. QRPers like BNCs because they're easy to attach-deattach, and the female connector needs less panel space. BNC male connectors take some practice to assemble, or the cable will pull out. Crimp-on connectors are designed for receive-only applications, connect-and-forget. I never would trust them. I've standardized on BNCs. All of my test equipment uses them, and it's easier to only have to deal with one type.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

I visited a proper electronics store today with the intent of buying a BNC plug to create a BNC-RCA adapter cable. It turns out, there were ready-made BNC-RCA adapters available - along with an M-to-RCA adapter. So apparently there are still enough other folks using RCA connectors for RF.

My old VCR uses RCA connectors for rf. I've been known to use RCA-RCA cables designed for audio at rf. But all finished designs at KR1S use BNCs up to 100 W. In the old days, solid-state PAs would self-destruct if operated into an open circuit, and cat-proof connectors were a necessity. Nowadays it's less of a problem until you're scratching your head wondering why the band seems so dead, when the spotting networks are reporting lots of activity.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Finally getting around to testing some stuff I've been wondering about, like the effects of RCA plugs and impedance mismatches.

Currently I have a ~10m run of RG58A/U coaxial cable terminated at both ends in RCA plugs. One end goes to the rig and one end goes to the small transmitting loop antenna.

On 30m, the antenna bandwidth is such that I get 1 SWR bar at the low end of the band, zero bars over the middle of the band, and 1 SWR bar at the top of the band. (One "bar" is the undefined unit of my rig's uncalibrated SWR indicator.)

Now, the coax isn't long enough to run to my computer room, which is where I've decided I want to do my operating. So I added a 5m run of coax. But, this coax had its shield broken in two places as part of its previous life as a shielded receiving loop antenna. So, knowing full well that this was a hack and would likely lead to impedance bumps, I spliced the two breaks in the coax by tightly wrapping copper foil over the gaps and soldering the foil to the shield at both ends of the gap. I soldered RCA plugs on each end of the new cable. Finally, to connect the 5m cable to the 10m cable, I used a cheap female-female RCA adapter.

So the entire transmission line looks something like:

rig --> RCA plug --> 2m RG58A/U --> splice over broken shield --> 2m RG58A/U --> splice over broken shield --> 1m RG58A/U --> RCA plug --> RCA female-female adapter --> RCA plug --> 10m RG58A/U --> RCA plug --> antenna

How did it work?

The bandwidth of the antenna increased on 30m. I was able to tune over all of 30m with zero bars SWR. I believe this indicates increased losses in the system. Thus, yes, impedance mismatches with RCA plugs, RCA connectors, and splices do cause measurable differences in performance at my station. (That, and/or the extra 5m of transmission line is inherently introducing loss.)

Short of replacing the entire transmission line with a new 15m run of coax, I suppose the next best thing would be buy a new 5m length of coax, and connect it to the existing 10m length of coax with a standard 50-ohm connectors (BNC) to properly terminate the coax segments where they meet each other.

To really satisfy my curiosity, I could buy a 20m length of coax, use 5m as an unbroken extension, and compare the properly-impedance-matched 10m+5m run with an unbroken 15m run.

After all the effort you put into building that loop, does it make sense to shave a few yen on spliced coax, when you know it will likely introduce unnecessary loss?

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

Mostly, I'm curious about if a "properly" done coax extension will exhibit the same measurable loss as compared to my current improperly done coax extension.

But you're right. With all the careful work balancing the alternatives and finding a feasible, minimum-loss construction technique for the antenna, it makes more sense to buy a new unbroken coax run and to know that transmission line losses are minimized.

For my QRP situation avoiding loss is critical, but in general, are coax runs for HF ham work almost always continuous and unbroken? Or are there any rules of thumb for how to extend already-installed lengths of coax cable?

Actually if I'm buying new coax anyway, maybe it's time to think about the type of coax and the inherent loss in the coax. I think I now have RG58A/U, which if I recall correctly is a shade lossier than RG58/U. I'll have to look to see what types my dealers stock, but what would be a good low-loss HF/VHF coax type for a cable length of 15m?

Transmission line theory is a rather foggy subject in my mind, but I look forward to learning more.

You may be over-obsessing about "critical." At HF, the difference between 15-20 meters of those '58s isn't worth worrying about. You really really really need to get some actual on-air experience.

Over the course of a typical multi-hop HF contact, signal levels will vary 6-20 dB. A few tenths of dB in the coax is not going to make or break a contact. There's no universal law that says you have to run 5 W. To gain operating experience you're better off running more power so you can make more contacts. When making contacts becomes commonplace, you can handicap yourself.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

OK, good to know. And slowly I am inching towards getting on-air. Extending the coax run to the computer room is to ease my first operating attempts by allowing me to use the PC for all sorts of convenient stuff like automatic CW decoding or at least visual display of the dits and dahs on the waterfall, great audio filtering, recording audio of the QSO, easier logging, and maybe even CW-keying the rig.


As long as my antenna is on the balcony and in close vicinity to the neighbors, I really want to avoid TVI/RFI. Running low power is I think the best way to do this, as higher power might invite overload of nearby devices that no amount of filtering could fix. If I can't make contacts, then I'll probably haul the collapsible loop antenna to a nearby park or outdoor area and try operating there. (I'd need to get or build an iambic keyer first though, as straight-key operating from a park bench doesn't seem feasible.)

Skip the straight key.

Don't sweat power. I run 1 kW within 20 feet of my building with no RFI. Your 10 W won't be a problem.

If you go portable, take a dipole as an inverted vee.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/

I finally worked up the nerve to answer, on my straight key, a somewhat rapid contest CQ. He didn't hear me. Thus begins my foray into QRP.

He was plainly audible here but less than S1. Guess I should start by finding and calling the S9+ stations. This was on 40m, for which my current loop will have probably less than 10% efficiency (plus any losses through my spliced coax cable). I'll give it a go at the higher bands, but right now they were noisy.

On the positive side I have my operating desk and PC digital audio recording setup mostly worked out, so even if a contact wasn't made, the station was improved.

Congrats on trying! My won-loss record, when I started in ham radio, and when I started in QRP, was worse than the Chicago Cubs', so don't be discouraged. Yes, call the strong stations. The QRP gang hangs out around 7030 in your part of the world.

73,

_________________

http://kr1s.kearman.com/
http://qrp.kearman.com/