Up one level
Back to Home
Some Thoughts on Off-The-Grid (Solar, Micro‑Hydro or
Wind Powered) Repeater Systems
By Mike Morris WA6ILQ
This page is just some random thoughts, not a really polished article. Comments and contributions are welcome.
The majority of the efficiency engineering on small power systems has been done by the "off-the-grid" homeowners... those that live far from the power lines and depend on solar panels, wind generators, micro‑hydro water systems, etc. The "bible" of their community is the 6-times-per-year Home Power Magazine (at http://www.homepower.com), published by Richard and Karen Perez (N7BCR and KA7ETV) in Oregon. If you are interested in solar power for a repeater site, or in cutting your home power bill, this is the magazine for you. Since 1987, they have published hundreds of articles on solar, wind, and microhydro electricity, energy efficiency, solar hot water systems, space heating and cooling, green building materials and home design, efficient transportation, and more. Their web site is worth a visit for subscriptions, CD and book orders. And the magazine staff walks the walk as well - the entire magazine production is done with off-the-grid computers. Their web site is a fount of information, as are their back issues which they sell in PDF form on CDs. Yes, every issue ends up a single searchable PDF file, and they sell multi-year collections on CDs. No, I've no interest in the magazine except as a long-time (over ten years) subscriber. The advertising pages in the magazine pay for the production and are useful as a source of vendors of panels, charge controllers, batteries, energy efficient appliances, and more.
No, they didn't pay me to write and publish the above paragraph. I'm an over-ten-years subscriber, feel it's a valuable resource, and wanted to share the information that they exist.
I don't know how they stay in business,as they sometimes give away the current issue as a free download, available at their web site... and that site also includes free downloads on batteries technology, battery care, solar power theory and basics, construction articles, solar space heating systens, and more. They were talking about dropping it, but if it's still there the "homebrews" page is particularly interesting, as there are construction articles there including a battery charge controller/shunt regulator and a low‑voltage disconnect circuit.
Back to repeaters - If you try some experimentation you will be amazed at just how much transmitter power you DON'T need if you have a good site, and a good antenna system. And a better antenna will give you more performance boost than a more powerful transmitter since it will benefit you on receive as well. As an example of reduced power, locally we have a UHF repeater sitting at the top of a 4,800 foot mountain. The system has two 4-pole antennas on the tower: a shared receive antenna at 120 feet with 7/8 inch heliax feeding an Anglelinear preamp stuck up the base of the antenna, then the feedline, then connected to a multicoupler that drives the receivers of several different repeaters that range from 406-512MHz. Between the multicoupler output and the 440 system receiver is a single 8 inch or 9 inch pass cavity. The transmitter feeds a circulator and a second 8 inch or 9 inch pass cavity then to 1 5/8 inch heliax to the transmit antenna at 80 feet (the 1 5/8 feedline was leftover on the tower from a previous tenant). Even with only two cavities plus lots of physical antenna separation there is less than 1 dB desense. One day I noticed that the system was not as audible inside the parking garage at work as it normally was... it was a little scratchy as I drove into the parking structure. Later that day I asked the owner about it and he commented that the 100w transmitter power amplifier had died the previous week and the system had been running on the exciter since. Hmmm.... and the output power was? About 150 milliwatts into the circulator / feedline!
As evident by the previous paragraph, antenna efficiency
is everything. Antenna pattern and antenna gain is a frequently
misunderstood term. Every dB you get in antenna "gain" is a dB
that isn't going somewhere else. If that "somewhere else" is some
place you will never be, then it's A Good Thing. The gain in
an antenna comes from limiting the pattern in some way - by
tilting it, flattening it, or by changing the directivity.
Look at this web page: http://www.marcspages.co.uk/tech/antgain.htm.
So before you go to all the effort of engineering a ultra low drain 50 watt repeater, do an FCC web search for geographically nearby systems and see what else is at the same site (even commercial or public safety), and then put an appropriate attenuator on a scanner receiver to give yourself the performance of an effective 10 watts (or whatever your situation calls for) ERP from that site and drive around as you do a lot of listening. You might find that you can get by with a 20-30w ERP system instead of the 100w you think you need. That means a lot less transmitter, which means a lot less battery (i.e. fewer dollars spent) and fewer solar panels... (i.e. even fewer dollars spent). Or a lot more run time for the battery you do buy.
Another weay of saying the above is that many times, cutting your transmitters power by 3 dB and using 3 dB more antenna gain can cut power consumption by 75%. And the 3 dB increase in antenna improves your receive performance as well.
You might get some ideas for your design if you peruse the manual for an old Motorola or GE fire tower (also known as a "lookout repeater") radio. This was a combination base station and repeater which could be used locally on a simplex channel, and by flipping a switch (labeled "base" and "repeat") could be used as a repeater. It was a handheld based design that had a PA deck that put out either 10 or 20 watts depending on the model but was very stingy with DC power. The lookout repeater was originally developed in the 1960s (with the HT200 handheld boards) and was redesigned as the radios went through various generations. I've seen the manuals on a couple of generations of the radios and seen a couple of the radios on eBay (the latest manual I've seen is on the Motorola MX page at this web site - look for the MTR-300 FireTower repeater PDF file). Aerotron, Repco and others also made firetower radios. Then there was the suitcase repeater Motorola made from some MX series handhelds originally for an unidentified intelligence agency but it became a commercial product (and got updated to Saber handhelds, then to Spectra mobiles)... look for "suitcase repeater" on eBay (but before you buy one make sure you can get it programmed - that particular software is NOT one of the ones that is common in ham circles).
One of the most useful features in the more advanced repeater controllers is analog voltage inputs (i.e. remote metering). You can measure battery voltage, battery temperature, charging current, discharge current, and more. You can trigger announcements based on certain measurements - i.e. when the battery discharges to a certain voltage threshold point the repeater can announce that fact and switch to a shorter carrier delay (hang-in timer). This change in repeater characteristics can clue the users to keep their transmissions short. Additional trigger points can make more pointed announcements until the repeater transmitter is disabled completely (until the battery is recharged).
The most common technology for powering off-the-grid repeaters is solar panels. Don't think for a moment that you will have a solar powered repeater - you won't. It's not solar powered, it's battery powered, and the batteries are solar recharged. If the sun doesn't shine for a week your batteries die, and the repeater dies. If your power source is a micro-hydro system, then it's water powered, and at the mercy of the battery and the water source (and there is no reason that you can't have a mixed source - micro-hydro and solar). The batteries are the critical part, and will probably cost you as much as the power source does, and require more maintenance. More on them later.
The power generating difference between a horizontal solar panel and a properly angled (tilted) panel can easily be over ten to one... as proven in my own personal experience: I had an 12v 16 amp array of four 4 amp panels in parallel lying on the ground in my back yard charging a parallel bank of three almost discharged 12v 80ah gell cells and saw a charge rate of 1.4 amps. Tilting the array (of 4 large panels) by hand while watching the ammeter I saw over 16 amps at the best angle - and to maintain the peak current I had to reset the panel angle about once per hour as the sun traversed across the sky.
There are tracking solar panel mounts that maximixe the power generation but they are not practical at rural, occasionally visited repeater sites (moving parts need maintenance, and most of my repeater sites get visited only when necesary, and there's one that I've not been to in over 4 years). Vandalism is another concern - I've seen photos of a 8-panel array that had 30 bullet holes in it, plus a shotgun slug through the tracker drive. The entire solar array was a total loss.
Before you buy a tracker mount or a DC battery charge controller do your research - there is also a controller technology called Maximum Power Point (MPP) or Maximum Power Point Tracking (MPPT) that adjusts the input impedance of the DC controller on the fly to match the output impedance of the solar panel (which varies drastically with the amount of sun falling on it). A good MPP controller can get you as much of a boost in power generation as a tracker mount. But be careful - some MPP contrllers are RF hash generators. If you can, try before you buy, or have an exchange arrangement before you hand over your money.
Back to panels...
The most effective fixed mount is to point your panels perpendicular to the sun at solar noon (which translates to true south at an angle of roughly the latitude plus 15 degrees). Hence at my location (which is near the 34th parallel) I'd use an angle of about 48-50 degrees... but that changes with the seasons.
However.... if you think about it the best power tilt angle is NOT the best winter snow shedding angle. So in snowy areas you either end up making a site visit twice a year to change the angle, or you end up mounting the solar panels at the optimum winter snow shedding angle and adding enough panels to keep the system working in the winter (short solar days), and taking the charge reduction in the summer knowing that the more intense sunlight (higher solar flux) and the longer solar day will more than make up for the non-optimum mounting angle. And even with the longer solar day you may have to use a charge regulator that has an output that switches on an auxiliary load to bleed off the summertime excess DC power (one popular method is to run a fan or two to blow some cooling air through the building, another is several resistive loads in a well-ventiallated rooftop housing).
When you build up the mounting frame for your solar panels for the repeater site you may want to plan on tower mounting them. Not only will it get you above the snow on the ground, but it will get you above some of the vandals (or thieves). And when you mount the panels you may want to survey the tower for a second panel (i.e. plan for system growth). And don't forget that the panel will add wind loading, and the weight of the snow will add torque loading to the tower.
And as was mentioned above, if you use solar panels on your system, it's not solar powered, it's battery powered, and the batteries are solar recharged. Just as in your car, the battery acts as a sponge for electricity... the load squeezes it out, and the charging source pours it back in, but you lose a little in each direction. If your inbound isn't up to refilling it while also keeping up with the outbound, it's going to run dry.
Your solar panels get to both run the repeater and charge the batteries in the daytime and the batteries are all you have at night or on cloudy days (and what if you get ten cloudy days in a row? Or 15 days? That's not unheard of in many parts of the country, especially in those where the seasons are measured in almost winter, winter, late winter, and road repair). In those areas you can bet that you will need panels equivalent to at least 5, sometimes 7 or as much as 8 times the transmitter drain to run it 7x24x365 (i.e. you may need 100 watts of panels in the winter to run a repeater that draws 15 watts of DC)...
One fact that bites the behinds of some folks that are just starting out with solar panels and batteries is that running some batteries down to under 40% to 50% charge will damage them permanently (the exact percentage depends on the battery chemistry and the ambient temperature). Yes, your 200AH battery bank may only have 100AH of usable capacity, and even less when it gets cold. And cold temperatures can do a real number on battery charge, as anyone who lives in Alaska, Michigan, the northeastern states or Canada, parks their car outside, and tries to start it at 3 AM can attest. You can easily lose 1/2 to 2/3 of your battery capacity in the swing from 75 degrees F (about 25 degrees C) to 35 degrees F (about 2 degrees C), and if your batteries freeze they can be destroyed.
So if your repeater site is in an area where temperatures get down to freezing you will have to buy or build a insulated battery box. And some extreme areas may require a heater inside the batery box (which is just that much more energy that you will have to generate). And don't scrimp on the insulation - you will discover that styrofoam is MUCH cheaper than batteries. Don't forget the insulation under and over as well as around the batteries.
If you have a large enough panel array to generate a decent charge in the winter you will have serious overkill in the summer despite the increased repeater usage caused by the longer days.... Depending on your latitude the lower angle of the sun in the sky means that the sunlight hours in December provide as little as 1/3 of the charge available for the same hours in July. And in Canada, or the northern states like Washington, Montana, or Maine it's a lot worse.... the higher latitude means fewer sun hours, plus there are lots of overcast days, lots of rain, plus snow.... You may need to include a low voltage disconnect sensor that protects the battery bank by either shedding the load (shutting off portions of, or even the entire system) or fires up an automatic "helper" generator - be it wind, gasoline or diesel... There are several little single cylinder auto-start and auto-shutdown 1800rpm diesel generators available (they resemble an oversize horizontal driveshaft lawn mower engine mated to an oversize automotive alternator). In general, diesel is preferred over gasoline as the fuel lasts longer in storage, has a lower evaporation rate, has more energy per gallon, and the engines are simpler and last a lot longer. Another point that was made to me over lunch one time is that you can backpack a five-gallon can of diesel into a site and pour it into the site tank a lot easier than you can backpack an equivalent amount of propane and tranfer it over. And you can pack in multiple 5-gallon cans in one day. But if you do chose propane, there are both "gas" engines (i.e. spark-ignited) and diesel engines (i.e. compression-ignited) that run on propane. When you go searching the catalogs, look for an 1800rpm industrial or marine unit instead of a 3600rpm "consumer" version as it is built much more rugged and will last a lot longer. Visit your local library - the various farm magazines are a good source of ads and information articles on small diesel generators, and the various Yachting and RV magazines have ads on both gas and diesel remote start (i.e. electric start) and remote shutdown generators. Some of the magazines have annual buyers guide issues that you might find in your local library magazine collections (or as a purchaseable back issue).
If you are engineering a custom repeater, do a power budget spreadsheet. Don't just read the spec sheets, measure how much current each item draws. Look at the run time of each item and calcualte the amp-hours for each item on a best-case and a worst case basis. Then total up the amp-hours you will need to run the system for a day, then determine the solar hours for your location, and remember that peak power is only for a few hours around solar noon each day. Then figure the amp-hour of battery you will need - but take into account the voltage sag curve of the battery and the cutoff voltage of the repeater and controller. You must consider whether the end voltage upon which the battery capacity is based is the same or higher than the voltage at which the repeater shuts down. If you overlook that and the repeater were transmitting continuously the current draw times the hours would suggest a longer backup capability than you actually get. And because repeaters are normally not continuously transmitting the batteries will last longer than the calculations.
When you get into the actual repeater and controller electronics look at what can be powered off, even temporarily. Even a commercial repeater can have some massaging done on it. For one, a crystal based radio uses a lot less power than the same synthesized radio, and since a repeater never has to change frequency (and the duplexer would need a full retuning if it did) so a synthesized receiver and transmitter really isn't needed. Crystal radios are a lot cleaner (spectrally) than synthesized, and generally don't need as much duplexer (or get better performance from the same duplexer).
One "battery saver" trick that was popular in handhelds in the 1970s and 1980s was to shut off the entire receiver for 90% of the time, then to power it up for 10% of the time. The same trick can be applied to the repeater receiver. The timing was based on how long it took for the receiver to recognize that there was a signal on the channel after it was powered up. Once it was on, if the COR saw a signal then the receiver stayed on until the signal went away. Yes, in a worst-case scenario where someone keyed up and started talking the same instant as the receiver powered off you might miss the first half-word but that was the price you paid for the extended battery life. Something similar can be built up for a repeater receiver.
There are radio specific mods that can be done to save power - for example, a MICOR station receiver can save a lot of idle current (over 1/4 amp) by disabling the receiver audio amplifier. Since repeat audio is taken off before the volume control, the audio amp is only used to run the local monitor speaker, which can be switched on only when you are there to hear it. The disabling can be done by cutting the trace to pin 16 of the Audio & Squelch board, then jumpering the cut with a switch (one trick is to swap the the volume control for one that has a switch on it, and use that switch to enable the audio amplifier). When you aren't there, just rotate the volume control to the off position. This will work with any MICOR, be it a converted mobile or a real station, on any band as pin 16 is always Audio A+.
Or if you want to have a full-time speaker and low drain, just remove the audio speaker amplifier transistors and replace them with an LM386 chip - and even that can be turned off with a switch. Or an electronic switch - the Wilson 1402 hand-held switched the LM386 with a series PNP transistor that was driven by the squelch circuit. Or both a physical switch and an electronic one in series.
Likewise, some of the receive audio circuits (like the CTCSS decoder) in the controller can be power-switched by the channel busy (COR) signal. If you are running a repeater that will require CTCSS the touchtone decoder doesn't need to be powered up until the CTCSS decoder is active. The controllers transmit audio circuits, in fact the entire transmitter exciter can be switched by PTT (some already are) so that there is zero power drain when the repeater transmitter is off (the stock MICOR runs the lowest level transmitter circuits full time 24x7).
After you have modified the radio gear you can turn your attention to the repeater repeater controller hardware and software design. For example, the entire DVR system can be switched on or off by the DVR access signal. When you sit down and examine the schematics closely it's amazing how many power consuming circuits can be switched off by judicious use of the proper control signals. I once saw a homebrew controller that had the CPU spend 99% of the time stopped. Ten times per second a hardware clock pulse started the CPU, it swept through the code and then halted the CPU (which used almost no power when halted). The code was table driven, and each pass through the code resulted in different paths being taken, different subroutines being called, different things happening, all depending on if a COR was active, if a touchtone digit had been received, if the IDer was active (and the CPU switched the tone on and off for each dit or dah), which audio crosspoints were off, in monitor, or in transcieve, etc. But the key was that the controller (one version was four radio ports, another was six ports) used less than 60ma when idle with everything that wasn't needed powered off, and less than 440ma at worst case - when every transmitter was IDing and each receiver had a touchtone digit in the buffer). Sixty milliamps of idle current seems high today, but this was the late 1970s / early 80s when the circuitry was mostly TTL as CMOS didn't have the wide chip variety or availability). The point I am trying to make is that even today a power reduction of as high as 7 to 1 or 8 to 1 is possible by careful controller design (or redesign).
Power consumption engineering in radios used to be a lot better... I always liked the innovative design of the Motorola HT-200 handheld receiver (late 1960s)... the receiver and transmitter boards were a real pain to work on (each one was built in four "layers" of parts) but if I remember correctly the entire radio drew less than 10ma when squelched, and it didn't have a pulsed power save circuit - it actually was designed to have the nicad battery run the receiver for five days or more when squelched. The radio was a two channels max (but I've worked on one that was six channels). The audio stages and the PL decoder were powered off when squelched, each IF stage was designed for 3v operation and the receiver had four separate stages in DC series across +12v, and the same 2 ma ran all four stages. I've not seen anything like it since. Too bad it was positive ground... it was a really nice low power design that didn't sacrifice performance. One of the early 160 MHz USFS fire lookout repeater designs was based on the HT200 boards, with a 2w in /20w out amplifier deck behind the transmitter - they got around the positive ground situation by transformer coupling the audio in and out of it, and capacitor coupling the RF.
You can't really say that your modifications you have cut the power drain in half until you actually measure the before and after power consumption. You can measure AC power consumption with a meter called a "Kill-a-Watt", made by a company called "P3 International". The model number of the one I used is a P4400. DC consumption in 12v circuits could be measured with a West Mountain Radio "Whatt Meter" but this product has been discontinued. It has been replaced by the "PWRcheck".
Several mountaintop radio sites I visited in central and northern Oregon and southern Washington used thermoelectric generators (no moving parts at all) powered by propane just because the snow would have built up on solar panels if they had used them. The propane tanks at each radio site were sized to last the entire winter (I saw 1,500 gallons at one site, and there may have been more tanks that I didn't see). Even so the local propane supplier had an all-wheel-drive tank truck to refuel his more remote remote / rural customers and the forestry radio and firetower sites (dual rear axles, 4 tires per rear axle plus the front axle - a 10 wheel drive propane truck is not a common sight, but obviously was needed in that area).
If all you are looking for is a battery-backup system for a regular AC Mains powered repeater you can use a regular Solar Charge Controller with a Low Voltage Disconnect circuit as part of it (or a separate Charge Controller and a Low Voltage Disconnect). Get one that will handle the repeater system load current (plus some headroom for expansion later), then connect it as per the instructions except that connect the system power supply to the Solar Panel connections, and the entire repeater hardware package (including the repeater controller), is on the load connection. The solar charge controller will provide the low voltage disconnect function in addition to its main job of managing the battery charging and increasing the life of your your battery bank.
Once the total repeater site load is optimized then the budget determines the solar panel array size and the battery bank size. The larger the panel array the more it costs, and there is a tradeoff between array size, battery capacity, square (or cubic) footage occupied in the building, and more. If the battery voltage drops to a minimum too often then it is obvious that the amount of stored power is insufficient. The weather can be "normal" for years, and then mother nature can throw you a curve ball and deliver a solid month of cloudy weather. Installing an array sized for a one-in-ten years event is overkill and a waste of money (not to mention the snow load and wind load if the array is mounted on the tower). It will be more cost effective to plan to take a 5-gallon can of fuel and a small portable rope-start generator (500-600 watt) to the site and cable it in for a few hours to do a battery recharge, then take it home when you are done. Such a generator need not be expensive, QST had an article called "The 12 Volt Pup: A DC Generator You Can Build" that was essentially a lawn mower engine and a 50 amp car alternator. It was in the June 1997 QST.
If an auxiliary generator becomes a regular necessity you can swap it for a diesel you can hook it to the controller and remote start it and kill it via DTMF. I saw just such a system at one of the Oregon sites - an electric-start 2KW generator that used a Yamaha single cylinder diesel engine. I was told that it was purchased out of a magazine geared towards small sailboats (i.e. mail order). It was mounted in the bottom of a home-made welded angle iron frame with a 55-gallon drum of diesel in the upper part of the frame. This gave gravity feed - and no need for a fuel pump (one less thing to break and / or require maintenance).
I really suggest that you do some serious research before you spend your first dollar. Visit the Home Power Magazine web site, and the web sites for some of their advertisers. Spend a day at a large public library, look in the farm, boating and RV magazines. If you can find a firetower or lookout repeater manual you can look at the overall design and see how the total battery drain is minimized (i.e. the power control circuitry). There is a lot of information available for free...
There were at least two Solar Charge Controller articles in QST. One was by N1BBH in June 1987 issue, the second was by Mike Bryce WB8VGE in the October 2001 issue.
The author can be contacted at: his-callsign // at // repeater-builder // dot // com.
Back to the top of the page
Up one level
Back to Home
This page originally posted on 10-Dec-2006
Text, artistic layout and hand-coded HTML © Copyright 2006 and date of last update by Mike Morris WA6ILQ.
This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.