Thursday, April 18, 2019

Antenna coupling


Figure 1, Figuring out where the damn antennas are going to go.
Radios require antennas, and unless you're blessed with a decent acreage, lots of money and an understanding spouse, there's limited room to put them up. As things currently stand, I have a 6m-2m-70cm beam already up. I want to put up one or more wire antennas for HF use. I also need a receive-only antenna for my scanner(s).

I also need to keep in mind that I also need for a spot to mount a WeatherFlow weather station.

Without a tower (expressly forbidden by Mrs. Freeholder) and given the wooded nature of our back yard, the house itself is the logical place to mount all of them save for the HF wire(s). This also serves to keep the cabling short.

For the scanners, I'm going with a wideband discone antenna. It's relatively small and this one is relatively visually unobtrusive. I like that with my antennas, given that there are a lot of them around here.

But where does that discone go?

I've considered just getting a 20' TV mast and mounting it on the back of the house, but I can't find those any more. The longest I've found so far are 9', but you can stack them. I've found 10' non-stackable masts. I've also found a lot of telescoping masts of various lengths, but those cost more than my budget wants to allow. I'd like to use things I have on hand. That means the front runners for this method are:
  • Use the two 9' lengths of 3/4 galvanized pipe I have laying around, connected with a coupling and mounting it on the back of the house. This would probably leave me a bit short of the needed height.
  • Get another gable-end mount and put it at the opposite end of the house from the existing beam. That gable is 2 stories in the air. I have a ladder that should reach it, but my days of hanging on a ladder that far up are probably over. I'll likely need to rent a lift. There goes the budget.
I've also considered mounting it on a 5' mast extension over the beam. It would be simple and I have the stuff to do it. Being a discone, it's omni-directional, so which way the beam points isn't an issue for it. But when looking into how far the two antennas needed to be separated, the subject of "antenna coupling" comes up.

Based on what I've found so far, this shit is dark magic it's a complicated subject. I'm still working on it, but so far here are the bullet points:
  • If you have more than one antenna, you need to consider the issue of antenna coupling. It becomes more important if you have multiple antennas connected to multiple transmitting radios at the same time, and even more important if the antennas are resonant on the same frequency or harmonics of the frequency in use. A lot of people don't take this issue into consideration, and some of them blow up equipment. It appears to be rare, but it happens.
  • Antenna coupling causes two main problems: Desensitization and receiver overload. Of the two, the protection of your receiver(s) is much more important. You can physically damage a receiver when a very large amount of signal comes in through an antenna.
  • Horizontal and vertical separation are important. Incredibly rough rule of thumb is that you need to be at least one wavelength away in all directions from the neighboring antennas but more is better. You may be able to get away with less than one wavelength as long as it isn't a harmonic of the wavelength. More separation is always better.
  • Higher frequencies cause more trouble than lower frequencies.
  • There are on-line calculators that will give you the necessary vertical and horizontal separation of your antennas. Unfortunately, these assume omnidirectional antennas.
  • Unless you are operating at very high power, VHF/UHF and HF antennas exhibit little coupling as long as you keep then reasonably separated.
  • There are no perfect solutions.
I'm working on the assumption (oh-oh) that the only two antennas I need to be concerned with are the 6m-2m-70cm beam and the discone. 2 meters is uncomfortably close to the VHF public service bands.

I've found the following useful in this research:
And believe it or not, I've found something on a ham forum that I think is useful.
I like this concept. Doing so may show that the difference between the Real World and Theory World are enough to solve my problem for me. Unless it gets worse once I do the modeling.

But after all that, what if there is still no clear answer to the question "Where do I mount the discone and the weather station?" Given that I've ruled out almost every possible location, I'm thinking it has to go on the side gable over the driveway, expense be hanged. That leaves the back gable as a place for the weather station, using a gable-mount mast mount and a TV antenna mast. Done correctly, this gives me the largest possible physical separation of the two antennas that are my main concern.

This solution will cause me some grief down the road if I ever want to mount another antenna, but I don't see any realistic options.

An idea I'm toying with is to keep every antenna that's not in use disconnected and grounded. Operate on one radio/frequency at a time. Scanner is disconnected when running 2m. That should pretty much cure the location problem since only a single antenna would be in active use at any given time, and it has the benefit of being budget friendly. It also addresses the issue of lighting, something that is always present. Even if it isn't a final solution, it gets me past this hump and gets the antennas up.

I should have gotten a simpler hobby. :-)


Monday, April 15, 2019

Our Lady is burning


Notre Dame is burning. 850 years of history and art are gone. Reports from the scene say that "It is all burning." Firefighters are trying to save any art they can. The Ile de la Cité is being evacuated. No word yet on the cause of the fire.

Current drop vs. voltage drop

In doing a bit of looking relating to B's correction to my earlier article, I ran into this video.



Interesting and yes, I'm buying a copy of his book.

"In the real world, of course, everything is an oversimplification."

I love that.

So, about those expiration dates on your meds

Dr. Joe Alton on using meds after their expiration dates.


Obviously there's more to it than what can be covered in 6+ minutes, so some other sources for you to take a look at:
Unfortunately, the actual website of the Shelf Life Extension Program requires a login. And their security cert is bolloxed up. :-)

Sunday, April 14, 2019

It's all about the voltage drop

Most people, if they have half a brain, will check what wire size is needed to carry a given ampacity before they start hooking up stuff. It's important if you're wiring a building and important if you're wiring electronic equipment. You don't want things you value to burn down or up.

Unfortunately, an ampacity table, while easy to find with a quick Intertubz search, doesn't tell the entire story. While researching why a small 2 meter amplifier I recently purchased called for 10 gauge wire "for a short run" or 8 gauge wire "for longer runs" for a maximum current draw of 345 watts, I kept finding various notes on solid vs. stranded wire and "current drop".

So I did my own research, and I feel pretty confident with what I'm going to present. Think of these as rules of thumb, and remember you should always check the numbers yourselves.

In 3 bullet points:
  • Resistance in ohms per 1000' of wire generally decreases as the wire gauge goes up and the wire goes from a single solid conductor to multiple strands. Not always, but generally.
  • When sizing wire for a given voltage draw, normal ampacity tables have simplified things down to the point where their accuracy needs to be in question. Compare this table with the one from the previous bullet point. This table uses a large amount of fudge factor, so based on something in the next bullet point, you may be buying a lot more expensive copper than you need. Or not. You have to do the math. Every time.
  • The correct way to size your wire is to look at the voltage loss for the particular type of wire and its application. The "acceptable range" for current drop is generally said to be 3-4%. So once you know the volts and amps that need to be delivered to a device, it's relatively simple to use this formula to determine if a given wire will work over a specified run. That formula is:

    ((Rw * 2l * .001) + 2k) * A = Vd


    where:
    Rw = the 1,000 foot resistive value
    l = Overall length of the cable assembly (include your connectors)
    k = resistive value for one fuse and its holder, conservatively 0.002 ohms
    A = Peak current draw in amps
    Vd = Cable assembly voltage drop


    (This formula, along with a master class on all varieties of wiring for amateur radio, can be found at the web site of K0BG. The entire site is a great resource, not just for the mobile operators it's aimed at, but all hams.)
Let's put this into practice using my situation as the example.

As noted before, the product manual says to use 8 or 10 gauge wire. The problem, as I originally saw it, was with the connector on the amplifier. It's known as a Clinch-Jones connector, but a picture is worth a thousand words.:
Click to embigginate
For reference, that wire is 10 gauge. The lugs measure 0.143"/3.64 mm, and there are 4 of them. You're seeing the two in the vertical orientation; there are two more that are horizontal. One set positive, one negative, and you need to connect your wiring to both lugs in a pair.

Can this be done? Sure, but I think you're running a sizeable risk of damaging the connector with the necessary heat. I considered using quick disconnects, but I can't find any that fit 10 gauge wire and lugs that small.

So I started looking at it from the standpoint of "Do I really need that much wire for 345 watts?" Look at the power cord for a 1500 watt electric heater - that wire is nowhere near 10 gauge - it's more like 16 or 18 gauge.

So let's put all down. I need to provide 13.8 volts DC @ 22 amps, per the amplifier manual. A 3% voltage drop is 0.414 volts; a 4% drop is 0.552 volts. Our acceptable range for Vd:  0.414 - 0.552. I want to use 200oC silicone insulated all-copper wire. 

An immediate problem popped up. The wire vendor I'm looking at doesn't provide resistance values. I'm using values from here, using the closest wire they have to the 14 gauge I'm considering using. (I'd use their wire if I could, if only because they provide full information, but they appear to be an industrial supplier.) The required length is 6' and I'm using 22 amps as called for in the manual.

For 14 gauge wire, it's

((2.99 * 2(6) * .001) + 2(.002) * 22 = Vd, or 0.88 - not nearly good enough.
For 12 gauge wire, it's

((1.6 * 2(6) * .001) + 2(.002) * 22 = Vd, or 0.51 - We have a winner!

Of course, I could give the 10 gauge a try, just to see

((1.1 * 2(6) * .001) + 2(.002) * 22 = Vd, or 0.38 - at the other end of the acceptable range.

So 12 gauge stranded copper it will be. Now for how I'm going to make the connections - back to research!

Edit, 4/15/2019: Once again, something generated a lot of messed up HTML and the math for the wire gauges I looked at was blanked out. Fixed it.

In the comments, B points out that this is actually voltage drop, and he's correct. I used the term "current drop" because that is what the majority of links I found called it and I wanted to stay with the common term, even if inaccurate. But he's right, and  after thinking about it I'm going to change it to the correct terminology in the title and the bullet points. Accuracy is more important than keeping Google's search results happy. I'm leaving it in the body where I refer to how this information was originally found.

Edit, 4/18/2019:You might also wish to read "Current drop vs. voltage drop".