Tech Forum Questions

Hi, Gary! I am a ham (amateur radio operator) and here is what I know about Yagi antennas (actually Yagi-Udi antennas, but that's a story for a different time).

I would advise against WD40, as it will almost certainly mess up the impedance where it matters (at the driven element). But first, a bit about Yagis.

A yagi antenna is constructed such that the center point of all the elements (other than the driven element) is grounded to the mast. So shorting the elements to the mast is done at the factory for all but the driven element.

The driven element is the one that connects (usually via some sort of matching network) to the coax, and my bet is that THAT is where you need to focus your attention.

Also, there are affects on the antenna from 'nearby' objects, including 'ground' (or anything horizontal below (or even above) the line of sight of the antenna). If the 'nearby' objects are enough wavelengths away from the antenna and not in the 'line of sight' then their impact should be minimal.

However, 'ground effects' can be helpful or hurtful, even out to multiple wavelengths away from the antenna. This is where 'higher is better' comes from. By the way, the higher the antenna is above 'ground' the better the antenna works.  However, if you mount an antenna higher above the earth, but your rooftop is in the line of sight to your other antenna, your 'ground level' is the roof, not the ground you walk on! (Especially at the frequencies we are talking about here).  So, all other things being equal, it is better to put the antenna on the side of the house closest to the antenna you are 'talking to'.

For more information, I recommend the ARRL Antenna Book.

Anyway, back to the question:
There are a few places where you are likely to lose signal strength:

1 - At the end of the coax, or inside the end of the coax. If water gets into the coax, it will change the characteristics of the coax such that you WILL lose (lots of) signal strength. If the coax terminates in a connector of some kind, then you want to protect the connector from water. If the coax does not terminate in a connector, then does it end such that the wire angles DOWN to the end of the coax or up? If up, then you have a great place for water to get in to the coax. If possible, orient the coax such that water on the coax will not travel toward the open end of the coax. In either case, protect the end of the coax, and all connections, with something non-'conductive' at RF (at microwave, to be more precise). Either some sort of silicone sealer, or use a plastic margarine tub or similar (assuming the antenna is far enough from the chimney to avoid melting the plastic!). Before using a sealant, make sure it is not conductive at microwaves (PVC is notorious for being conductive there). If in doubt, one partial test is to place some of the material, along with a cup of water, in a microwave and cook for a while (start with a few seconds, if the material in question does not get hot, go for longer). If the material will 'bother' your antenna then it will likely get hot (or at least warm) in the microwave. Do this same experiment with your margarine tub also!

2 - At the matching circuit. What sort of matching circuit is being used will determine how easily affected it will be by water and  so forth, but generally the matching circuit wants to be dry. Some stay drier by design than others. In any case, the margarine tub from 1 could probably be placed over the entire driven element and coax end. Another possibility would be to build a 'roof' for the antenna, but if you use metal or anything affected by microwaves (see experiment in 1 above), then you need to keep it a reasonable number of wavelengths away from the antenna.  Since we are talking microwaves, or at least 900MHz, then that isn't so bad as trying to build a roof for an HF antenna! If I remember correctly, 5 to 10 wavelengths away should be more than adequate. At this point, I should recommend the ARRL antenna book, as it has MUCH more info, including the frequency to wavelength formula and how far you should put the 'roof.' You might get away with 1 to 2 wavelengths away, IF the 'roof' is not resonant at the frequencies of your internet provider. This is where the ARRL Antenna Book is probably your best bet.

3 - At any point between the antenna and the receiver where water can enter the coax.

Since you say that the fireplace seems to exacerbate the issue, then you probably have a combination of soot/whatever plus water in the matching circuit, coax, or both.

Random ideas and thoughts:
Can you move the antenna further from the chimney?

Perhaps you need to shield the antenna from the smoke as well as from the rain.

Rather than moving the antenna higher, is there anywhere else where you still have line of sight to the cell tower (or whatever is the other end of your radio circuit)? In other words, if you could protect the water-sensitive areas of the antenna, AND move it to where it does not pick up 'stuff' from the chimney that would be a win-win. One concern is  water dissolving stuff from the fire that then attacks and corrodes the antenna!

Given that we are working with at least 900MHz, as long as you have line of sight to the other antenna, anywhere on your roof should be almost as good as anywhere else. In fact, if you have a clearer shot 6 feet lower, then 6 feet lower is better, as long as you don't move your roof, (or neighbors roof), into the 'ground effect' area) So, if you can mount the antenna away from the fireplace and still get a good signal, that's probably better all the way around. In fact, as I say, start out with the side of the house closest to the other antenna.

Bottom line: weatherproof the coax, especially the ends; protect the matching circuit from excessive water; keep the exhaust from the chimney from mixing with water and depositing on the antenna (possibly by moving the antenna elsewhere).

Water combined chemicals in smoke from a fireplace could get into connectors and cause signal loss in Gary Rathburn's antenna system. He notes, "If it's rainy and we're not using our fireplace for a few days, the signal also improves."  Perhaps the rainwater by itself flushes some soluble chemicals out of the connectors. It's unlikely water on the antenna or rain droplets cause a problem.

During a dry period, wrap exposed connectors with Coax-Seal or Scotch-brand "2228 Moisture Sealing Electrical Tape." Better yet, replace the connectors and the cable and then seal the new connectors. If moisture and chemicals from smoke got into a connector, they probably got into the cable too. Also, always create a drip loop to keep water from running down cables and into connectors.

If you try to "seal" the antenna, you can de-tune it and cause signal loss. Antennas are manufactured to work outdoors in bad weather.

Yes, fire makes carbon smoke a conductive. I'd try a plastic bag to see if it corrects it. Or, place a shield one foot under the antenna.

I suggest you try wrapping the elements of your Yagi antenna with PVC electrical tape. It might protect the antenna enough to provide the necessary improvement. As a young "shop gopher" in a two-way radio shop, I earned many hours of minimum wage wrapping two-way base station antennas with electrical tape. The owner of the shop insisted it reduced received noise and extended the life of the antenna.

It sounds like you have a faulty cable and/or cable connector at the antenna. I would replace the cable with Wilson 400 low-loss cable, and put a boot on the connection at the antenna to keep out moisture.

Additionally, the Wilson 301111 Yagi is quite directional; try aiming.

If it has been several years since you installed the antenna, maybe more cell towers have been built in your area.

I think you should look at the connectors. They may not be water tight!

Electrical duct butter, electrical duct seal, and coax seal can all be used to encapsulate the connector and socket area to seal it from water inclusion.

I’ve used Beam antennas up in the air for decades with good results.

First, you won't find (as far as I can tell) any negative regulators that will take more than -40V input. Indeed, you will have trouble finding ANY regulator (positive or negative) that will take 36 to 75 VDC input.

So, since it appears that you have only +35 to +75 VDC available, I recommend something like the MIC28500 switching regulator (Digi-Key has them; http://search.digikey.com/us/en/products/MIC28500YJL%20TR/576-3969-2-ND/2764089 — note that these are surface-mount deals ... sorry!) to provide something in a voltage range that is easier for other regulators to handle (or you could just use two of those for the five and 6.5 volt outputs, and one for supplying the other regulators — note that one of these things can supply up to four amps!). So, I'd say:
• One MIC28500 for the +5V
• One MIC28500 for the +6.5V
• One MIC28500 supplying +12V (its max output) to the following:
Some sort of switching inverting regulators (like the LT3483 or similar) to convert the +12V to -6.5V, and -15V outputs.

Another switching regulator to boost the 12V to 15V, for example, something like the LM2574N.

If the MIC28500 chips are too expensive, you could just use one to supply +12V and choose other regulators from there (for example, an old run of the mill 7805 or 78L05 for the 5V).

I found all the above regulators by going to digikey.com and searching for 'voltage regulator IC.'

On the other hand, you could drop the voltage down to something easier for 'normal' voltage regulator ICs to handle using a resistor and a zener diode. So, you'd have a resistor of proper wattage and resistance in series with a 20V (for example) zener. Allowing for 2A of current (assuming we end up using inefficient linear regulators here, since the load is only 640 mA max) with an input voltage of 80V and an output of 20V, we need to drop 60V in that resistor at 2A, or 120 watts. Ouch! That's nuts! So, let's go back to the MIC28500 or similar.

Is there some way of getting a 'nicer' voltage input range beyond regulating it inside this power supply? That is, if you can manage to get the input voltage to a max of 30V, you greatly open up the possibilities for off-the-shelf voltage regulator ICs.

On the other hand, if you don't want to build it, I found that Acopian makes DC-DC supplies that could work — even if you just use one to convert that 36-75 range to something more friendly. The 48C5FT1000 supplies 5V and the 48C15FT400 supplies 15V, but they are pricy; about $188 each. TRC Electronics also makes such things, with various output voltages (trcelectronics.com).
 

Summary:
The options seem to me to be:
1 - Buy an off-the-shelf regulator(s), e.g., Acopian (expensive).
2 - Use some sort of down-regulator (Acopian, TRC, MIC28500, zener(!), etc.) to supply a lower voltage to the various end-user regulators, then have some number
(probably five; one for each voltage output) of regulators supplying each output voltage from the step-down intermediate supply. Use inverting switches for the negative; use whatever you feel comfortable with (LM7805, etc) for the + voltages (still somewhat expensive, and now you have to make it work).

Although it doesn't use a microcontroller, Intersil described a circuit for making an auto-ranging DVM in their Application Note AN046. The URL for the PDF file of this article is www.intersil.com/data/an/an046.pdf.

Their basic circuit (Figure 1 in the app note) is for a 7106 DVM chip which is meant to drive an LCD display and run off batteries. A modification of that circuit is required to use a 7107 chip which uses a dual supply instead of a single supply, and has a higher power drain due to the LED display needs. That modification is given at the end of the Intersil article, in the text and in Figure 8. This circuit design utilizes the over-range/under-range states of the DVM chip to trigger the auto-ranging function, so depending on your application it might be of benefit to use an improved DVM chip that incorporates a zero-integrate phase feature to improve the recovery from over-range. Some manufacturers have added zero-integrate phase to their modern versions of the 7107 (e.g., Maxim), but DVM chip versions that explicitly include this improvement include the 7136 chip (sub for the 7106) and the 7137 chip (sub for the 7107). I noticed a couple of minor typos in the Figure 1 circuit of this article. The CD4028 logic chip is mislabeled as CD402T, and both output lines are labeled as over-range ("OR") instead of one line being labeled over-range ("OR") and one being labeled under-range ("UR").

After a bunch of looking, I found this on Amazon. www.amazon.com/DROK-Synchronous-Voltage-Converter-Waterproof/dp/B00J229O9A/ref=sr_1_36?ie=UTF8&qid=1418062090&sr=8-36&keywords=dc+dc+converter+Buck

You didn't say how much power you needed but, this looks pretty robust. Good Luck

There is a digital oscilloscope from Yokogawa that might fill your needs. The DL850 ScopeCorder will sample at 20 MHz with a 12 bit resolution. Since this is a modular instrument, you will need to select the components to fit your needs. See www.scopecorder.net.

I think a buried antenna will not radiate very much.