The KC7GR Techmobile

One thing about amateur ('ham') radio people has held true since Hiram Percy Maxim made his first QSO: We love to tinker around with electronic equipment and, in many cases, build our own.

This also applies to assembling any amateur radio service station. From the smallest and simplest backpack or handheld radios to the kilowatt-plus multimode home stations that can match the performance of some military setups, every station is an expression of the particular owner's interests and philosophies.

I'm no exception, and I've learned quite a bit about mobile radio and roving electronics installations from my tenures at Motorola and Boeing. When I set out to equip our Dodge Grand Caravan SE, I wanted something that could grow with me, something that I could use both for normal amateur work and for future public service work with RACES, ARES, or similar organizations.

Thus was born 'Project 400' (from the Quicken accounting code I set up to record the costs involved), or what my wife more affectionately calls the 'Techmobile.' This page will detail the whole project, and (I hope) provide some valuable tips for others who may be considering similar installations.


The project has been through four phases and, except for routine maintenance issues, is complete.

Phase One took place during the summer of 2000. It consisted of initial planning, equipment procurement (specialized accessories, such as the mobile keyboard, display, and housing for the onboard computer), building infrastructure (the equipment shelf, cabling, and front console arrangement), and doing the first wave of installation (all equipment initially placed and wired).

Phase Two took place during the summer of 2001 and part of 2002. It was a cleanup of all the mistakes made in Phase One, plus implementing corrections and modifications based on 'Lessons Learned.' This means, in essence, that I found better ways to do some things that I hadn't thought of before, and decided to implement them as revisions to the initial install.

Phase Three came in the summer of 2003 (can we see a pattern here? Guess I don't like to work outside when it's cold...) and brought the addition of a lightbar and associated wiring.

Phase Four, which was completed July of 2004, added a PA amp, a matching external speaker, a jump-start connector and isolator assembly, a map light, a wireless network card in the onboard computer and external antenna for same.

In the details below, each descriptor block will be marked with a number indicating which phase it was a part of. In cases where modifications were made after the initial phase, multiple numbers will appear.

The Antennas (1, 2, 4)

The antenna system, to my mind, is easily the single most important part of any amateur station. You can have the slickest, feature-loaded, DSP mil-spec radio on the planet, and it won't do you a bit of good if said radio is on, say, 440MHz while your antenna wants to resonate somewhere in the middle of the FM broadcast band.

When we (my wife and I) first got the van, I was edgy about punching holes directly in the roof for the antennas. The initial setup, therefore, consisted of some creativity involving 19" rackmount panels, held in place between the cargo rack bars with Adel clamps.

In the summer of 2005, I decided to remove the rack panels and switch to through-roof mounts all around. This was done for a variety of electrical and mechanical reasons, not the least of which was that rain started leaking in through the rear hatch where the coaxial cable bundles were coming in.

This is the result of the rebuild.


The new layout is as follows: Left side, main VHF quarter-wave (PCTEL BMAXMFTS series) for the main Spectra VHF radio. On the right front, another identical whip for the APRS transceiver. Left center is a dual-band (800 & 1950) elevated-feed antenna for the cellphone. Right center: Larsen dual-band for the 2m/440 section of the Icom 706.

Left rear: Antenex 'Phantom Elite' series, model ETRAB4303, for the Spectra 440 radio. Believe it or not, that little blade has 3dB gain to it, according to Antenex's specs. Finally, right rear: A dual-band (2.4/5GHz) whip for the onboard PC's wireless network card.

The most notable change you'll see is that the 'Pod' GPS antenna has been replaced with a much newer and smaller unit. Specifically, a Micropulse (now PCTel) model 1213FW, supplied by Navtech GPS. Although expensive, at around $225 shipped, it features 40dB gain, and is rugged enough to stand up to just about any kind of weather. The higher gain helps greatly in 'urban canyon' and mountain terrain as well. Here are bottom and top views of the antenna itself.


Installation went pretty smoothly, being that the antenna is a single-hole mount. The first step was to temporarily pull out the van's overhead console as shown here.

The next step proved to be a bit tricky. Cutting the initial mounting hole went smoothly enough, thanks to an Irwin 'Unibit,' (sometimes known as a 'step-drill'), but I quickly discovered that the antenna had an anti-rotation key built into its mounting stud. This required the use of a nibbling tool to notch out the forward section of the hole, as shown here.


Once this was done, the antenna mounted snugly into the hole, and it was then a simple matter of fabricating and running the appropriate cable (in this case, a section of RG142) back to the receiver. The overhead went back together smoothly enough, but it turned out I needed to slice a bit more plastic out of it to accommodate the new connector and cable.

I've had two questions frequently asked. First, why (with two exceptions) did I choose unity over gain antennas? Second, do the radios interfere with each other, or have degraded radiation patterns, with the antennas in the configuration they're in vs. a straight line down the middle of the roof?

The answers, in sequence: I chose unity gain for two reasons. First, their small size. As you probably noticed in the opening shot, our garage does not have a lot of vertical clearance. I needed to strike a compromise between performance and being able to pull the van in at night (this is not one that I want to leave exposed on the street for any length of time!)

The second advantage that unity-gain antlers have is their broad bandwidth. I can cover the entire 2m band, and beyond, with a single 19.2" whip, and that's important considering that I operate on both amateur and commercial frequencies. The UHF side is not a concern, as the Phantom series UHF antennas have a rated bandwidth of at least 20MHz for 1.5:1 SWR.

As for interference: I've only ever noticed two issues. One was where the APRS transceiver was desensing the VHF Spectra radio. This was cured (mostly) by swapping antenna positions around, and by replacing the preamp front-end module in the Spectra with a standard version. The other is that the Icom gets into the cellphone handsfree adapter when I'm transmitting on 6 meters. I was able to make some headway against that by installing ferrite beads, but it's still there, and I still blame Motorola's cheap design for the overall problem.

On the radiation patterns: Again, I've not noticed any problems with any of the local repeaters, or with simplex QSO's. I don't really have the equipment or expertise to develop rad-pattern charts, so I've simply not bothered. If any of the readers of this page are still curious, and want to 'recruit' me to do some pattern tests, drop me a note and I'll see if I can work it into my schedule. ;-)

Not shown in the roof picture above is the HF ballmount. It punches through the body on the left rear quarter-panel, inside the jack compartment, as shown here.


The antenna itself is an eight-foot (108 inch) CB whip that, in combination with the AH-4 (Icom) automatic antenna tuner, serves the Icom 706 radio from 6 to 40 meters. While 80 meters is within the radio's capability, covering it antenna-wise would have required a 23-foot whip, minimum, and I have enough issues with an eight-footer. I've talked over halfway across the country (from west of Seattle to Philadelphia), and clear up to Saskatoon, on around 60 watts on 40 meters, so it obviously doesn't do badly.


DC Power Sourcing (1-4)

Next most important factor in a mobile installation: Clean DC power, and lots of it! To reduce the risk of draining the main battery, I chose to install a second deep-cycle battery and a charging isolator to handle most of the mobile hardware.

This is the charging isolator. Instead of using simple diodes, as most isolators do, it uses a series of high-current power MOSFET transistors as electronic relays. MOSFETs have much lower forward voltage drop than diodes, which makes them ideal for high-power environments.

The unit is made by a company out of Montana called Hellroaring Technologies, and is their model BIC75300A. It turned out, in retrospect, that this particular unit was overkill for my application. If you're doing something similar, the BIC75150A will probably work just fine. You really only need the 300 if you're going to be running a configuration where the auxiliary battery is expected to be able to crank the engine on its own, something that I chose not to worry about.

The mounting location shown is on the inside surface of the forward cross-beam in the engine compartment, just above the twin electric cooling fans for the radiator core. This was about the only place I could find under the Caravan's hood that was available (it's TIGHT in there!) Another thing I liked about this particular unit, besides the use of MOSFETs, is that you can bypass its isolation simply by applying +12V to one control wire. This allows the secondary battery to, if need be, charge up a dead starting battery. Let it do the bypass for about 15-20 minutes, and you'll probably be able to start up again.

CAUTION! Regarding drilling through the firewall: I can't stress enough the need for extreme caution when running any sort of extra wiring through a vehicle's firewall. Based on my own bad experience (despite all my precautions, I drilled through my heater hoses, causing a $276 repair bill), I would like to offer these guidelines to others.

Triple-check where you want to drill, on BOTH sides, for obstacles like hoses, control cables, other wiring, etc. Ideally, put a thick piece of cardboard or similar substance on the engine compartment side between where you expect the drill to come through and whatever might be in the way. Another thing that can help is slipping an aluminum spacer over the drill bit, leaving only about a half-inch exposed. This limits the penetration depth once you break through. If you don't have a spacer, you can achieve much the same effect by wrapping the bit with a short piece of electrical tape.

Start small and work up! It may be a lot of extra work to drill a hole, change bits, and enlarge it step-by-step, but the alternative of causing damage to the car, your drill, or yourself is much worse. I started with a 1/8th inch bit, and worked up to just over 3/4 inches, and I ended up going through just about every bit in a 29-piece kit. Even though I did hit the heater hoses, I missed some nearby underbody wiring, something that would have been a lot more serious had I hit it.

This is an older shot showing the original configuration of the van's starting battery and part of its new wiring. It's all type THHN #8 wire, rated as oil, moisture, and gasoline-resistant, and therefore an ideal choice for the entire job. Although most of the auxiliary equipment is on the secondary battery, I did want to put the cellphone's handsfree kit on the starting battery, as well as leaving room for other stuff in the future.

This is a more current shot of the new battery, and the newly-installed jump-start connector. Space is so limited up front that my best choice was simply to tie-wrap the connector to the battery's carrying handle.

The rectangular black object, directly below and in front of the battery, is a home-built diode isolator bank that serves to let jump-start power flow only one way -- from us to the other car! The last thing I wanted was transients and other junk from a foreign electrical system finding their way into the van's systems, and this isolator was how I did it.

The unit consists of four Motorola Schottky diodes, each one rated at 35V and 300 amps, for a total forward current capacity of 1,200 amps. The diodes are 'flat-pack' mount, and are sandwiched between two heavy copper busbars that are 1" across by 1/4" thick. The busbars are then mounted into a pair of Delrin plastic strips that I had machined specifically for the job. I chose Delrin because of its high mechanical strength, high temperature tolerance, and ease of machining.

The wiring is all #4 copper to match the jumper cables. The diodes are mounted very close together, and their spacing is within 1/16th inch of each other at the base, so I didn't think I'd need equalizing resistors. I did my first jump-start for someone not too long ago, and the diodes didn't show any ill effects afterward, so it seems to work.

This is the heavy-duty contactor for the jump-start connector. It's rated for intermittent duty of about 4 minutes on and 10 off. I chose it over a continuous-duty unit because the intermittent-duty versions were the only ones capable of passing enough starting current (300-400 amps) to handle most situations.

This switch controls the jump-start contactor. Note the switch guard, which forces it into the 'Off' position when closed. The bracket is home-made from sheet aluminum stock, primed and painted. The LED, of course, lights when jump power is being provided.

If you look again at the old battery shot, you'll see a 50-amp aircraft-type circuit breaker in the upper right of the picture. This sources the auxiliary lead that I ran to power "Bus A" which runs the cellphone and any other minor accessories that don't need to be on "Bus B" (which is the auxiliary battery side).

This picture shows the front of the master DC distribution box. While the first stop for the high-current DC cables is in a different section (more on that later), I want to show actual distribution first. The box holds aircraft-type circuit breakers for just about everything I installed, and the DC/DC converters to run the audio panel and onboard computer display (both devices required more than 13.6V nominal to function reliably).

This is the 'door' section of the DC distribution box. If you ever happen to get a look at the breaker panel on a Boeing jetliner, don't be surprised if you see some similarities. The grid that the breakers are mounted on follows the same spacing, and I'm using the same style of busbars that Boeing does to source the 'line' side of the breakers.

The breaker enclosure itself is a standard NEMA 4X rated box that I got from Platt Electric Supply. Like all the NEMA boxes I used, it's made of a Fiberglass-filled plastic resin that is incredibly easy to drill or cut.

HOWEVER -- WEAR GOOD GLOVES AND A DUST MASK WHEN YOU DO SO! I neglected to take such precautions while I was drilling the front lid for the breakers, and I ended up with the worst case of Fiberglass 'itchies' I've had since the last time I pulled communications cable through an insulation-laden ceiling! Be sure that you vacuum up the dust afterwards, too, and use a shop vac or one similarly able to pick up things like fiberglass. Household vacs won't cut it.

This is the 'box' section of the DC distribution box. Although pretty much empty, to allow space for the door to close, I still had room to mount those two DC/DC converters I mentioned earlier. Since they were both originally designed for PC board mounting, I secured them each to their own piece of perfboard, readily available from Radio Shack or most other electronic supply places, and made connections with the aid of some surplus terminal strips I had lying around.

The top one drives the onboard computer display, which (so I discovered) gets unhappy if its input voltage drops below about 12.8VDC. Jameco Electronics supplied the unit, made by MeanWell, their model SDM30-12S15. It takes 9-18V in and produces a steady 15V at 1.75A out, which guarantees the display's stability unless my electrical system drops below 9V (and if it ever does, you can bet I'll have other worries than keeping the computer up!)

The bottom converter runs the audio panel, which (coming out of an aircraft environment) requires 24V. I used a surplus converter, 9-18V in and 24V out, custom-made by Astec Power and obtained through Marlin P. Jones, an electronic parts mail-order place in Florida. They don't seem to stock that exact converter now (probably ran out), but something very close to it may be found at this link. If you end up having to use that particular converter, you could simply run its output through a three-terminal regulator for dropping its 36V output down to 24. Be sure to use a good heatsink on the external regulator, though.

A Place for Everything...(1-4)

I decided to construct a custom equipment rack, shown above, out of 3/4" plywood and aluminum angle stock (to help support the shelves). I built it to have the same interior width as a standard 19" equipment rack, for reasons which should be obvious to any of you familiar with EIA standards.

In case it's not obvious, I'll explain further. Take a look at the bottom shelf, below and behind that silvery rectangular object. You should be able to (just barely) make out the backside of a DEC (Digital Equipment, now Compaq, now HP, now a real mess, but that's another story) AC power controller. It happens to be designed to secure to a 19" equipment rack, and I needed it around to provide remote power on/off switching for the onboard computer, back when I had it running off the inverter. Thus, I decided to make the entire shelf to the 19" standard.

The primary fasteners are 1/4-20 bolts, nuts, and washers, all stainless steel, run through holes in the floor. Additional stability is provided by lateral braces (shown later), made of 1/8th x 1" aluminum bar stock, secured between the top edges of the shelf (front and rear), and the van's body.

CAUTION!!! In the Grand Caravans equipped with rear air conditioning, as ours is, the Freon lines for the rear A/C run under the floor, very close to where I drilled the primary holes for the equipment rack's floor bolts. BE EXTREMELY CAREFUL ABOUT WHERE YOU DRILL! Our regular mechanic says I missed the lines by a hair (scratched them, in fact), and that the repairs would not have been cheap had I drilled through them.

I also made liberal use of a mounting product called ThreadSerts, made by the Marson Corporation. Unfortunately, Marson seems to have been bought out by Alcoa, so their web site is no longer organized as well as it was. If ThreadSerts are of interest to you, check with Grainger -- they carry the entire product line.

ThreadSerts are, I think, one of the greatest inventions of the 20th century, and they proved indispensable throughout the entire project. Their sole purpose in life is to provide a stable, metal-sheathed threaded hole in materials that would normally be too thin to support it, such as sheet metal or plastic.

One example would be where the antenna tuner got mounted to the rear wheelwell interior trim panel. It's just 1/8th inch plastic, normally way too weak to support screws. Once I installed ThreadSerts, however, mounting the tuner (and other things like the front console) became a snap.

This shows the side of the equipment rack. From top to bottom, you'll see the master DC distribution box, covered in detail above, followed by the power & control junction box, and finally the auxiliary battery in its (black) case at the bottom. The battery was a special case for two reasons: First, it had to be a deep-cycle type, resistant to damage from deep discharge. Secondly, since it is mounted in the passenger compartment, outgassing of hydrogen needed to be at an absolute minimum.

With this in mind, and a little shopping, I discovered that a new lead-acid technology called 'AGM' (Advanced fiberGlass Mat) would be ideal. I used to use Optima 'Yellow-Top' deep-cycle batteries, but have since switched to Exide's 'Orbital' marine series (blue top) as it seems to have better long-term life.

The auxiliary battery is joined to the system through some heavy-duty connectors made by Anderson Power Products. More specifically, from Anderson's Multipole line. They go together easily enough, but you will need a special crimping tool to deal with the large-gauge contacts.

Don't skimp on this requirement!

There's nothing worse in a high-current environment than having your contacts suddenly become resistors, so invest in a high-quality hex crimper, along the lines of the Greenlee K09-2GL. It can handle anything from #8 through #4/0 terminals and splices. I got mine from Grainger Industries, their catalog #4NV94, at a cost of around $220 with tax.

This is an interior shot of the main power, control, and audio junction box. Note that the high-current power leads are all white, with colored heat-shrink tubing to denote polarity. This is because I got a great deal on some #4 aircraft-grade Teflon-insulated power cable. At 13.6 volts nominal, #4 is good for 100A, which made it the perfect choice for my primary leads.

The black cylindrical object, visible in the lower right, is an alternator whine filter installed between the charging output lead from the isolator and the auxiliary battery's positive lug.

The black rectangle, labeled 'No Step,' is an aircraft-type high-current terminal block used in (are you ready for this?) Boeing 737 and 757 jets. Its normal purpose in life is to serve as a junction point for the three-phase 400Hz AC power from the engine-driven alternator. It also, as you can tell, makes a wonderful junction for DC power systems.

The smaller relays surrounding the 'NO STEP' block on the upper edge are the various low-current control relays for power to accessories and other functions. Visible on the door are the barrier terminal strips for the audio and control cables running between the equipment shelf and the front console.

Radios, Computers, and Lights, Oh My! (1-4)

Safety is something I kept in mind throughout this entire project. One of the firefighters on the Batlabs radio discussion board pointed out (rightly so!) the possibility of, in a collision, the entire equipment rack assembly breaking loose and shooting forward at the original speed of the vehicle.

To deal with this possibility, and to provide a second electrical ground path, I constructed two lateral braces out of heavy aluminum bar stock, and installed them in such a way that the entire shelf assembly is solidly anchored to an upper body member as well as through the floor. More on these later.

This is an older close-up of the upper section of the rack. The 'Warning' panel about lethal voltages is there just for the 'Wow!' factor, and is not really true. The only DC voltage higher than 13.6 comes out of the audio panel's power converter, the only higher AC voltage ouf of the inverter, and that only when I'm testing something I got from, say, a swap meet where there's no commercial power readily available.

This is a later shot, taken at the completion of Phase 4 (18-Jul-04 to be exact), of an additional feature that I found necessary to add to support the PA amplifier module and the audio panel both needing to be connected to the DB25 on the front of the Spectra. More on that below.

Early or late, the equipment positions have not changed. Top center, hidden behind the 'Warning' panel, is the RF section of the Icom IC-706MKIIG radio. It serves as my primary HF-through-6 link, as well as playing backup for the VHF and 440 Motorola radios.

Top left, just below the 'Warning' panel; You can just see the 'Power' LED on my TNC, a Kantronics KPC-3+. The TNC itself is attached (with industrial-grade Velcro) to the top of a Motorola MaxTrac VHF mobile radio which has been modified to serve as a dedicated APRS transceiver. In this case, the only real 'modification' needed was to blank the radio's logic board, reinitialize it to be in the correct part of the VHF sub-band, reprogram, and realign. I have it set up to cover all three APRS frequencies known to be in use in the U.S., switchable by selector buttons on the front panel.

To the right of the MaxTrac is the 'Black Box' containing the GPS receiver, and it has gone through some changes between phases as well. The original unit purchased for Phase 1 (summer of 2000) was a Garmin model 25 OEM module that I ordered from TAPR. In early 2003, as part of Phase 2, I replaced the G-25 with a later model, the Garmin GPS-15L, obtained from GPS Central, a Canadian-based online store specializing in all forms of GPS hardware.

The 'Black Box' enclosure that I've got the receiver built into has some history behind it as well. The thing originally had nothing to do with Garmin, and housed a Trimble unit that was built to military specs and obtained from Boeing Surplus. Since the original receiver only had six channels, and only spoke Trimble serial protocol, I thought it best to simply gut the box's innards where the receiver was concerned.

I kept its nice DC/DC converter (8-32V in, +5V out, fully regulated and filtered), the enclosure itself (a very well-sealed two-piece hunk of aircraft-grade machined aluminum), and the mil-spec connectors, and simply built the Garmin units into the box (one at a time, obviously). Both the 25 and the 15 are 12-channel capable, and put out standard NMEA-0183 sentences, so they were ideal for my purposes. Better yet, the newer 15 is WAAS-enabled. This gets the accuracy down to the sub-3 meter range without the need for a differential beacon receiver.

I'm pleased to say that, with the combination of the Garmin receivers and the new antenna, I've yet to pick up fewer than seven sats, even in 'urban canyon' conditions. Nine is typical, and I've maxed out the receiver a couple of times by picking up 12 (usually on the I-5 corridor through California's northern interior valley).

One other thing I should mention about the GPS setup. You may note, at the extreme left center of the picture, just to the left of the MaxTrac radio, a tiny square black box. This is a DC injector, also called a 'bias tee,' obtained from AR-Squared Communications. It was installed to support the old 'Pod' antenna's requirement for 12VDC bias. This is why you see a breaker labeled 'GPS ANT' on the distribution panel. While the new antenna is perfectly capable of operating on the more common 5V bias, it can also handle up to 16V, so I saw no reason to get rid of the injector or its accompanying breaker.

This picture shows the next shelf down.One the left side, a Motorola Spectra VHF transceiver, 110-watt and 128-channel capable. Next to it, sporting a blue stripe on its backside, is a Motorola UHF Spectra, 40-watt and (also) 128-channel capable. These are my primary 2-meter and 70cm (440) radios, respectively. The extra D-sub 25 connector, dangling down from the front of the Spectra VHF, is there because the cable from the audio panel is plugged into the connector which would normally be used for the programming interface. I simply added an additional programming-only connector to that same cable so I wouldn't have to play 'Swap the Plugs' every time I decided to change something.

Some folks have asked why I spent more for commercial radios instead of just getting a Yaesu, Icom, or Kenwood. The answer is twofold. First, given my current line of work and possible future ARES or RACES volunteer work, I needed radios that could legally transmit on both amateur and commercial frequencies. Dedicated amateur transceivers are (usually) not type-accepted under FCC Part 90 for commercial service, so it is not legal to use them on, say, the local EMS or fire frequencies, even if you have permission from the appropriate agency to transmit there.

Second was ease of operation in terms of safety. In many cases, with amateur transceivers, you need to watch the radio as you dial up the desired frequency and offset. The 'memory channel' feature present in most modern rigs can help with this, but the displays of such radios usually present too much information to be taken in at a glance. The Spectra radios, once programmed, present a simple 11-character display, and a grand total of two LEDs ('busy' and 'transmit'). All it takes to change frequency is one simple motion of a rocker switch, and you can tell at one glance what repeater or simplex frequency you're on. This lets me keep more of my attention on the road, where it should be.

In any case, they weren't that expensive, considering what I got at the time I bought them (back in 2000). The UHF unit was around $400 used, and included a new control cable along with its other accessories. The VHF unit was a bit more, at $550, but it was complete with all its accessories and had the optional receiver preamp. I found this out when I tested it, and discovered that I was getting around .15uV for 12dB SINAD throughout the entire 2m band.

Spectra prices have come down since 2000. You can now get comparable VHF or UHF units for about $125-$300, depending on condition, model, and features.

On the right side of the frame is the junction box for the lightbar and 'SmartArrow' messenger (more on that later). Not visible in this picture, but shown in the overall shot earlier, is a silvery box hanging from the bottom surface of the middle shelf. This is a 12-disc CD changer, containing the mandatory Firesign Theater albums and some other goodies. After all, one can't be chatting on the 2-ways all the time, and hearing about the Giant Rat of Sumatra, or how one can be in two places at once (when they're not anywhere at all), can really help eat the miles away.

This is a shot showing the upper backside of the shelves, just behind the right rear passenger seat. On top, you'll see a blue box with a big white AC plug stuck in it. This would be the DC/AC inverter. It was originally installed to run the onboard computer, but I have since converted said PC to DC operation. I keep the inverter around to provide power for (as mentioned above) testing swap meet purchases in situations where commercial power may be unavailable.

You also have a great view of those lateral braces I mentioned earlier to the inverter's immediate left. I pray that I never have to find out how strong they really are, but their presence does wonders for my peace of mind.

Directly below the inverter, you can see the backsides of the Icom 706 at the top center, the front panel of the APRS radio to the right and the backside of the Kantronics TNC on top of it. You also have a better view of the GPS receiver to the lower left, just below the terminal strip


This is the next shelf down, also from the rear (of the rack). You can see the backsides of the Spectra radios, as well as the onboard computer in the dark blue case. Said computer consists of an industrial PC chassis with a passive backplane. Plugged into said backplane is a single-board PICMG-format industrial PC, using an Intel Pentium III-850 CPU and 512MB of ECC RAM. The box has a SCSI I/O subsystem (I do NOT like or use IDE/ATA!), an IBM 9GB UW/SCSI 'Deskstar' series hard drive, a Toshiba CD-ROM drive, and a DigiBoard multiport serial I/O card. Barely visible in the far left of the shot, just to the left of the computer, is the serial port distribution box.

When the computer was originally installed (2000), DC/DC supplies for PC's were not common, and they were fairly expensive. By 2006, however, prices and availability had fallen far enough to allow me to do an affordable conversion. This is the supply I chose, made by Opus Solutions, their model DCX3.120x.


After removing the original 110VAC power supply, and mounting the Opus board in its place, my next step was to provide a different power connector on the backside of the computer itself. I chose a six-contact mil-spec plug, simply because I happened to have a mating pair in one of my stash-boxes. This is the result.

I've added a multimode WLAN card as well. It's a D-Link DWL-AG530 that will allow me to link up with WiFi access points running any of the standard formats (802.11a, b, or g). This comes in very handy for doing some quick field research on items one might find at a swap meet (assuming there's an access point within range). It's also good for hunting around to see what wireless networks are Out There.


Controlling it all...(1-4)

This is an overall shot of the front. The entire installation was done with safety for the driver and passengers firmly in mind. On the floor, between the driver and front passenger seats, is a mounting frame made by Signal Measurements Corporation of Magnolia, Texas. SMC makes a huge variety of mounting devices for mobile radios, cellphones, and even laptop computers.

Starting from the back and moving towards the windshield: In the extreme bottom of the pic, you'll see the master hand mic and, barely visible to its right (sorry about the lighting), the main speaker.

Next item forward is the Motorola Systems-9000 control head for the 440 radio. This is followed by, in sequence, a Pipo 16-key DTMF pad, the handsfree/charging mount for my cellphone, the audio panel, the Icom 706 control head, and the scanner.

Important Note Regarding Motorola T720/T730 phones and their matching hard-mount hands-free kit (aka the 'Pro Install' car kit).

When I first installed the upgraded stereo receiver and cellphone in the van, I was also thinking about a neat feature that many modern phones have. Specifically, the ability to output a switched ground that can, in conjunction with a compatible mobile stereo receiver, mute the receiver's audio for the duration of the call. Some receivers, such as the Pioneer DEH-P560MP that I chose, will even pause the playing of a CD while the phone is off-hook if you install an additional adapter device (I chose not to).

However, after the installation was complete, I was annoyed to discover that the mute feature was not working. Some research on the web turned up the fact that this is a known problem with the newer Motorola SYN9760A electronics box when used with Verizon phones.

If you opt for a Motorola T730 phone, and decide to install the standard S9950 hard-mount car kit as well, be aware that you may have radio muting problems if you are a Verizon user.

You may have better luck, in that event, locating one of the earlier S9610 hands-free kits on Ebay. This kit contains the earlier SYN9271A electronics module, which seems to implement the muting feature correctly no matter what carrier's phone is used. Even if you do end up with a SYN9760A box, you can replace it at any time with a SYN9271A. They are 100% plug-compatible with the kit's wiring.

Going further up the console chain, built into a homemade inverted-U shaped bracket, is the control head for the 2m radio. On the right side of the bracket is a Federal Signal 'Littlite' series LED map light, model LF18ES-LED. These can be had at the wholesale level from for about $60, and they provide either white or red lighting at the flip of a switch. Best of all, with the LED model, there's no bulb or hot surface to worry about, and the current drain is a fraction of that required for an incandescent device.

Completing the stack are a pair of Motorola 'wildcard' switchboxes. These are model HLN1196A or HLN1241A, depending on when they were manufactured. In terms of functionality, they're identical: Eight pushbutton switches, six of which are on/off latching and two of which are momentary (at least in the stock models), provide control of any accessory you want to wire to them up to about 1.5 amps. The switch contacts are brought out to D-sub 25 connectors on the back of the enclosures.

The difference between the two models is in the backlighting. The earlier 1196 series uses incandescent lamps, where the later 1241 series has green LEDs. I've seen them turn up on Ebay for between $50-$75 each depending on their condition. One of mine came from Ebay, in fact, and the other from a local ham swap meet.

The only other thing I want to note about these boxes is that parts for them are still (as of June 2003) available from Motorola. This, in combination with the right tools, makes it easy to replace the momentary switches with latching ones. I did this in one of the boxes, and left the other one stock.

Back to the console itself. Poking up at the far-forward point, above the inverted-U bracket, is the mobile PC's keyboard on a 9-inch SMC gooseneck pedestal. The keyboard itself was explicitly designed and built for mobile operations by Texas Industrial Peripherals. It's completely sealed, environmentally speaking. In fact, it meets NEMA-4X specifications for keeping out liquids and dust. It also has a built-in PS/2-compatible mouse. Another nice feature is its backlighting, which makes it very easy to deal with at night.

Above the keyboard, mounted with another SMC 9-incher and some custom-made vertical braces, is the mobile computer's display. This is an industrial-grade (read: extended temperature range) unit, made by Earth Computer Technologies, their model number MTR-EVUE-12. What they did was take an OEM color LCD panel, made by NEC, and build it into their own enclosure with their own driver circuitry. It is exceptionally clear, sharp, and readable day or night, which is especially important when navigating by GPS and electronic mapping. Among the computer's software packages is DeLorme Street Atlas, a mapping package that can easily interface with a GPS receiver for real-time tracking.

This is a close-up of the console front. You have a much better view of the speaker, hand mic, and the UHF radio's control head. Barely visible, to the left of the speaker and the green/orange wire bundle, is the strain relief end of the mil-spec cylindrical connector that replaced the original one on the hand mic. If you've ever tried to find an easy-to-use panel-mounted mate for a Spectra's microphone connector (which is, essentially, a small cut-out section of a D-sub 50-contact assembly), you'll understand why I changed it.

This is a close-up of the lower console area. You have yet another view of the UHF radio's control head, the tone pad, cellphone mount, audio panel, and the Icom 706 head.

This is a tight-frame to show the audio panel. I chose it because it was inexpensive (about $60 from Ebay), available, and it had lots of channels.

The panel itself was made by ARC, Cessna's former Avionics division, before said division was bought out by Sperry. It's their model F-1010B, and it has proven to be a real life-saver in terms of preventing mics and speakers from breeding. It accommodates up to three communications radios, with full transmit/receive audio switching, and several more receive-only inputs.

It also allows the use of up to two hand mics and two headsets. I used one headset input for -- you guessed it -- my own headset, one of the mic inputs for the hand mic, and the second mic input for the DTMF encoder shown above the cellphone mount. The receive-only inputs currently serve the scanner, and the sound card output for the PC. The backlighting, designed to run on a 24V circuit like the rest of the panel, turned out to be exactly the right intensity for safe night-driving when run on 12V.

This is my headset of choice. It's a Plantronics model SHS-1890-15 PTT-equipped amplifier, along with a standard Supra H51 series single-ear (covering both ears while driving is illegal in most states) headset assembly. The Supra series is used by telephone operators and radio dispatchers the world over. Both amplifier and headset should be readily available through any branch of Graybar Electric. It interfaces to the audio panel through an old Motorola 'Centracom 2' dual headset jack assembly I found on the surplus market.

NOTE! Interfacing to an audio panel like this one has some pitfalls! If you plan to do something similar, there are a number of things you should be wary of.

First and foremost, these panels are often designed to run on a 24/28VDC electrical system. This can be overcome in one of two ways: Modify the panel, or install a DC/DC converter. I chose the latter course, as detailed earlier.

This next 'gotcha!' is a big one. Many of the late model radios from Motorola, Spectras included, cannot deal with either speaker lead being grounded, as happens with any aircraft audio panel's unbalanced receiver inputs. Doing so will DESTROY the radio's audio PA section!

Fortunately, it's Radio Shack to the rescue once again. I discovered that they carry a nice little 1:1 audio isolation transformer, catalog #273-1374. What you do is take the Spectra's speaker leads from the accessory connector on the front or rear panel, run them into one side of this transformer, and then run the other side of said transformer into your audio panel's unbalanced input.

The only other thing you need to do, radio-wise, is connect a dummy-load resistor to the speaker output leads at the control head connector. I found that a 5-ohm, 5-watt resistor does just fine, and I was able to make up little 'termination connectors' by using the generic AMP-made plugs (the Spectra uses the two-pin 'Mate-n-Lock' series) and contacts, and then enclosing the whole thing in a piece of heat-shrink tubing.

This not only keeps the radio's output circuit from going up in smoke, it also allows you to have independent control over each radio's receiver level. Beware, though! Despite the isolation transformer, it is still very possible to overload the audio panel's receive amps. I've found, so far, that a volume setting of 8 (Motorola's default on the System 9000 series radios) to 9 is perfect.

You will also need an isolation transformer for the audio output from a computer's sound card. While grounding one side of said output may not harm the sound card itself, it will usually cause an objectionably-loud 60-cycle AC hum in your audio feed unless you're using a DC/DC supply to power your computer. The Radio Shack part is small enough to be enclosed in heat-shrink right on the cable itself, where it plugs into the sound card's output jack.

Another hiccup I ran into was one involving an esoteric feature of the audio panel itself. It uses several diodes in its push-to-talk leads to isolate pilot and co-pilot PTT buttons. It still pulls the PTT lead to ground to key the radio, but that ground can be sourced through up to two silicon diodes before it ever gets to the radio.

While this is generally not a problem for a relay, or a radio that has a sufficiently sensitive electronic PTT circuit, I discovered the hard way that the Icom 706 didn't like it. Initial tests, using just the straight PTT lead from the audio panel, failed to push it into transmit mode.

I solved the problem by adding a keying relay in the junction box. The relay is keyed by the audio panel, and the relay contacts then provide a very solid switched ground to the Icom.

Interestingly enough, I ran into a similar problem in Phase 4, during testing of the PA unit. The Motorola Systems 9000 PA modules do not have a 'hard' PTT lead, as the radios themselves do (for the benefit of external options). Instead, the PA module requires an actual PTT command, and other commands on which mode to be in, via the same RS485 serial bus that the radio gets programmed with. In this case, the PTT command is generated by the microprocessor in the radio control head in response to the grounding of the PTT lead that normally comes from the (hard-connected) hand mic.

When I did the initial wiring for the audio panel, I didn't even bother with the control head PTT lead. I went straight to the 'Option PTT' lead present in the radio's accessory connector. This, unfortunately, has the side effect of not allowing the PA to work with my audio panel setup.

To get around this, I first purchased a spare Spectra microphone cable from Motorola. I then ran an extra lead from the main junction out to the control head PTT, through said cable. Instead of keying the radio directly, through the option PTT lead, the PTT signal coming from the audio panel would now hit the control head.

However, I discovered that the PTT sensing circuit in said head is no more sensitive to a diode-inline ground than the Icom turned out to be. Another keying relay corrected this.

On the DTMF encoder: I've had some questions about where I got it. It's a 16-key unit from Pipo Communications, their model PK-2K ANI 4.0. They were nice enough to sell me one of the ANI models for the same price as a non-ANI unit, which comes in really handy in terms of being able to speed-dial repeater control codes for, say, the Evergreen Intertie network. The unit also includes a dry-contact keying relay, which made interfacing it that much easier. It's tied to one of the hand mic inputs provided by the audio panel, so it can send its output to any of the transceivers.

Moving a little further forward, we find ourselves in IC-706 and scanner territory. Nothing too unusual here, just the normal remote mount kit for the 706, with a ferrite bead added at each end of the cable to keep RF out. This, BTW, is a necessary step for any IC-706 remote-mount installation. The beads are the snap-around variety, normally used for computer cables, and can be found at Radio Shack and other electronics supply places. This is a shot of the bead at the radio end.

If you happen to be using the Icom's remote audio/PTT jack, as I did to feed the thing into the audio panel, you should also put a bead around whatever cable you select to plug into the remote jack, as shown in this shot (it's the bright orange one).

It's nothing more than a CAT-5 network cable that's been chopped at one end and tied to the screw terminal strips in the junction box. I chose it because it had a nice 8-conductor modular connector (what the Icom requires) preinstalled, and because it was cheap.

Moving on to the scanner: You may note that I had no separate antenna for it on the roof panels. This is because it shares the AM/FM radio's antenna through a multicoupler (hidden inside the dashboard) made by Antenna Specialists, their model MON-63. The only modification I needed to make to the coupler was to remove the Motorola plug on its output cable and replace it with a standard BNC connector. It works very well, and I've not noticed any problems receiving broadcast stations, or anything on the scanner.

The display you see on the scanner is its reception of part of King County's trunked system, run by Valley Communications, a county-wide organization that takes care of most of the police, fire, and public-safety dispatching in my area. Programming any scanner for trunked reception would be a tedious task if done through its keypad alone, so I can (with the flip of a switch) download everything it needs from the onboard computer.

If you're curious about the trunked or conventional radio systems in the Puget Sound area, you would do well to check out the Puget Sound Scanning web site. They have frequency and talkgroup listings for most of the local agencies.

As I mentioned earlier, the IC-706 needs a tuner to talk effectively through a simple 8-foot whip. This is the Icom AH-4.

If you're thinking that the coax going in and out of the tuner looks a bit unusual, you're not far wrong. I got a good deal on some Teflon-jacketed double-shield RG-8 equivalent stuff through Boeing Surplus. It's the same cable they use in the jets for the antennas on the radar transponder and their own HF transceivers. Although it has a little more loss per foot than regular RG-8, thanks to the Teflon dielectric, it is extremely rugged in terms of flexibility and moisture resistance.

This is a close-up of the computer display, complete with its warning about bludgers (and if you don't get the reference, you should watch 'Harry Potter and the Sorcerer's Stone'), shown with StreetAtlas up and running. In order to best adapt this display to my application, I found it necessary to make four simple modifications; One to the van's dashboard and three to the display itself.

The first mod was structural. The dashboard top is nothing more than molded 1/8th inch thick plastic that was certainly never designed to accommodate the weight of any such display (around 7 pounds) via the leverage exerted by that weight pressing down on the mount's arm. To get around this, I installed a wide piece of thin (1/16th inch thick) aluminum sheet metal under the dash's top panel as a reinforcing plate.

While this did well enough for supporting the monitor's mount, the thing would start bouncing around whenever I was moving, especially when I hit a bump. This not only kept flexing the dash top, it made the display bloody hard to read! To cure this issue, I made two vertical braces out of aluminum bar stock, and bolted the thing solidly to the dash infrastructure. You can see the braces on either side of the display in the overall picture farther up the page.

Second: I was not at all pleased with the default power connector that Earth Tech had designed into the display's housing. It was a standard friction-fit 'coaxial' style jack of a type found on many consumer electronic gadgets. I removed this jack, and replaced it with a 3-contact mil-spec cylindrical connector, which is much more rugged and secure thanks to its bayonet-style locking feature and its resistance to dust and moisture.

Third: The monitor's adjustment switches (for vertical size, position, contrast, etc.) were all deeply recessed, accessible only by poking an alignment tool (preferably a non-conductive one!) through holes drilled in the back of the case. Fortunately, there was also a six-pin header on the main circuit board, unused in the stock monitor, which was in parallel with the PC-mounted pushbutton switches. I simply drilled three 1/4" holes in the back of the enclosure, attached three bounceless pushbutton switches in said holes, and assembled a little wiring harness to connect them to the header. It all worked great, and the monitor can now be tweaked, adjustment-wise, without having to remove it from the dash.

Finally, there was the issue of the display's backlight. It was far too bright at night for driver safety, so I bugged Earth Tech about it. One of their engineers sent me back a picture of a two-pin header on the driver board, labeled (appropriately enough) 'DIM,' and said that putting a 50k-Ohm pot across that point would get me my dimmer control. I did so, and have had no problems since.

On Spectra receiver muting...(2)

Most commercial radios (and even some ham sets) have a feature where, if the receiver is operating in tone-squelch or 'PL' mode, it will not unmute until it hears a signal with the right PL code transmitted on it. This prevents having to listen to idle chatter on a shared channel.

However, before one transmits, one needs to make sure that the channel is clear. Thus, when a radio mic is removed from its (usually grounded) hang-up clip, or a handset pulled out of its hang-up cup, the PL muting is disabled, and the receiver will pass any signal on the channel.

Since the audio panel meant a single mic for multiple radios, some unique adaptations were required in terms of muting control. I set up one of the relays in the rear junction box to be pulled closed whenever the hand mic is hung in its clip. However, I also needed a manual method to unmute the Spectras, so I ran the sensing lead through normally-closed contacts on one of the wildcard switches.

The relay contacts apply a ground to both mic HUB (Hang-Up Box) leads, one in each Spectra head mic connector, when the relay is energized. If the hand mic is removed from its clip, or the MUTE OVD button is pushed in, the relay is disabled and the Spectras hear anything on the frequency. Neat, heh?

Light 'em up! (3)

The very first thing I got asked, when my neighbors saw me mounting the lightbar, was "Is that legal?"

My response was a firm "Yes! At least in Washington state." RCW 46.37.215 covers it nicely with this paragraph:

"(1) Any vehicle may be equipped with lamps for the purpose of warning other operators of other vehicles of the presence of a vehicular traffic hazard requiring the exercise of unusual care in approaching, overtaking, or passing..."

The trick, in terms of not getting cops mad at you, is to have common sense enough to know when to use your strobes, and (perhaps more importantly) when NOT to! All-amber lenses are perfectly OK, as are clear ones. Red lenses are only OK if they're on the rear, and I opted for a combo of red on the rear and amber everywhere else. Blue is also a no-no, as is a combination of red & blue, or all red, for obvious reasons.

This is the innards of the smaller NEMA box mentioned earlier. You can see the master A+ relays, one for the strobe circuit and one for the halogen circuit, in the lower left and upper right respectively. You can also see the edges of the barrier terminal strips on the door side.

This board is a homebrew construct designed to address a problem inherent to xenon-strobe based light bars: They're too bright at night! Whelen anticipated this, and provided a control wire on their strobe power supplies that, when pulled high (to +12V), cuts back on the strobe power by about 30 percent.

I didn't want to have to switch this function manually. Too easy to forget, and it takes up an additional switch. What I wanted was a photocell-controlled circuit that would engage a relay when the ambient light level dropped far enough, and I found one at this link.

I found that, based on component availability (or lack thereof), and to make the circuit work as a 'dark' detector, some modifications were needed. R2 and R3 were replaced by 510 ohm resistors, and the NTE128 transistor became a 2N2222A. The relay I used is a simple Potter & Brumfield 12V DIP unit, and it is wired to apply +12V to the lightbar's low-strobe-power lead when it engages.

Finally, I reversed the position of R1 and LDR1. The photocell is a common CdS device that I found at Radio Shack as part of their catalog #276-1657 assortment. It's enclosed in a metal TO-92 half-height transistor can, and is glued into a hole in the upper right side of the enclosure where it faces the right rear window.

Don't Shoot The Messenger!

You may have noticed, in the overall picture, a pigtail lead poking out from the lower section of the lightbar's junction box. This is the RS485-based programming input for the Whelen Smart Arrow Messenger. This ingenious device consists of an LED-based alphanumeric message board that mounts into the backside of the lightbar, between the two rear strobes. My unit is a 16-character panel (models are available that have from 12-20 characters). You can have it display pretty much anything that can be spelled out, within the limits of its memory.

This is what the 'enhanced' model control head looks like. The messages are pre-stored (see below), so you don't have to worry about encoding one on the fly. The message number can be selected from the keypad, or scrolled to with the arrows. The control head itself is mounted on the driver's side of the overhead console's plastic frame with ThreadSerts and #10 screws.

How do you encode the messages? Simple! You use special PC software, sold by Whelen as part of the programming kit, to create your message library. You then load said library into a memory box called the 'Transporter.' This box is then hooked up to that programming plug, and a +12V source, and a press of the appropriate switches dumps the newly-loaded library into the Messenger's NVRAM. You can also reverse the process, and vacuum an existing library out of a Messenger for editing.

The Transporter can also serve as a passive RS232-to-RS485 converter, to allow a connected PC to program the Messenger panel directly. However, I've found it to be more comfortable simply to work up the library on my regular system, and then transfer it via the box's memory.

This is the end result (one of them, anyway). The wording 'Please Go Around' comes up first, followed by scrolling arrows that move outward in both directions from the middle, signifying that anyone behind me needs to go around to either side. The odd blank line in the middle of both pictures is an aberration caused by interaction between the scan rate of my digital camera and the SAM's multiplexing action on the LED display panel. Trust me when I say that every row and column on that beastie is in great working order.

The content and composition of the SAM's messages are limited only by one's imagination. One of my favorites (and the one I use most often) is 'You Are Too Close. Please back off! Thank you.' I have that one programmed to display with the push of a single button, and it has given more than one other driver a most interesting (to watch) wake-up call when they get too close to my rear.

As for the bar itself, it's a Whelen 'Edge' model 9622. It came from a vendor on Ebay. The guy was part of a fire department in Georgia that had changed over to LED-based bars, and were getting rid of their strobe-based units.

The entire thing, with shipping, was not cheap (just under $900), but it was a darn sight cheaper than it would have been if I'd gotten the system new from Whelen. In fact, the lightbar alone, without the Messenger panel, would likely have cost me as much as the entire thing did from Ebay!

The only change I made was to order new lenses and a mounting kit. In researching this, I found a terrific dealer to work with in Oregon called SirenNet. They're a big enough distributor that I was able to get a significant discount off Whelen list prices for the parts.

In case you end up with a similar bar, and you want to take advantage of your car's existing luggage rack rails to mount it, as I did, you would need to order Whelen mounting kit #MK-EZ43. It'll run you about $30 or so, and consists of a couple of T-bars with attached bolts that slide right into the cargo rack channels. The bar drops onto these bolts, and secures with 1/4-20 NyLock nuts and matching washers.

Oh, one other note. The stock lenses from Whelen come in standard lengths of around 17 and some-odd inches. They need to be cut to length to meet your specific bar's requirements. The ONLY good way to do this is with a table saw, overhead-rail saw, or compound-mitre saw, and a matching plastic-cutting blade. DO NOT try to use a hacksaw, as I did at first! You'll never get the straight, clean edge you need to seal properly against the divider or end-cap gaskets.

Can you hear me now? Good!

I made it a point, when I ordered the speaker for the PA module, to order the smallest and most compact one that I could get for a reasonable price (in this case, $85, shipping included). I went this route because, as you've already seen in previous photos, there is just not a lot of spare space in Dodge Caravans, under the hood or at the bumper line, to mount external goodies.

This is the speaker itself. It's a Galls unit, designed to mount in tight spaces. The most important requirement is that it have some sort of front-facing opening in the vehicle for the sound to escape. It's made of a heavy cast aluminum-alloy, is thoroughly weatherproof, and is rated at 100 watts.

Having a weatherproof speaker does you no good at all if your electrical connections to it are not equally weatherproof. With that in mind, I attached a mil-spec 3-contact plug to the leads as shown here. The backside of this plug is a potting cup that has been filled in with electronics-grade silicone sealant. Electronics-grade sealant is so named because it does not, unlike regular silicone sealant, release acetic acid during its curing process. Acetic acid, better known as vinegar, is corrosive to most electronic equipment.

The speaker cable's durability is just as important as that of the speaker's, as both will be exposed to exterior conditions ranging from freezing cold and wet to merciless summer heat. I made this cable out of three-conductor 18-gauge outdoor-rated AC cord from Belden. The stuff is jacketed in some sort of synthetic neoprene substitute that, according to its specs, is resistant to all kinds of temperature extremes, vibration, and abrasion.

The mate to the speaker's connector is shown here. Once again, mil-spec stuff all the way (it was not cheap -- about $30!) to effectively resist being exposed to potentially hostile environmental conditions.

Now to mount everything! It took me nearly ten solid minutes of hunting before I found a neat little space, tucked in behind the left front section of the "fascia" assembly (fancy term for the bumper and plastic parts that give the front end its nice appearance). The van's regular horn is mounted in there, and it turns out that there was more than enough space to accommodate the speaker as well. This is what it looks like, kneeling down in front of the van and looking up into the gap. The 'glare' effect was caused by the camera flash -- sorry about that.

This is a view of that same space from the backside. My fingers are visible in the upper right, pulling the plastic wheelwell splashguard back far enough to take the picture. As you can see, there's lots of room in there.

Here's a clearer shot showing the speaker bracket after installation. The backside of the van's regular horn assembly is visible to the left. The bracket itself is mounted with 1/4-20 stainless-steel bolts, secured into Marson Co. 'ThreadSerts.' Use of a normal nut/bolt configuration was made impossible by the simple fact that there's no way I could have gotten a wrench into the space behind the inner wall.

Here's the end result, at least what you can see of it from looking up into the bumper gap again. Plenty of room for the sound to escape and, unless you know exactly where to look, there's no way to tell that the van even has the thing installed.

This is a tight shot on the right side of the equipment shelf (about the only place I had left with space to mount anything more -- it's getting FULL in there!) showing the Motorola Systems 9000 PA unit. These will work with either the Syntor X9000 or Spectra 9000 radio systems, including the Astro Spectra, and can be located on Ebay (or from other suppliers) as the HLN1185A or HLN1185B.

You will, of course, also need the appropriate cable for your specific radio. For the Syntor X9000, a couple of conditions determine which cable you'll need.

If you have an HLN1185A MODULE:

You will need an HKN4246A cable.

If you have an HLN1185B MODULE:

You will need an HKN4304A cable.


For the Spectra radios, things are a little different. The siren module type doesn't seem to be that important (though I've never seen anything other than a 'B' revision unit used).

If you have an ASTRO Spectra:

You will need an HLN1439A cable.

If you have a STANDARD Spectra:

You will need an HKN4363A cable.

Pinouts of these cables, and an enormous wealth of other info for Syntor X9000 and Spectra radios are available from Mike Blenderman's site, and from the Batlabs site.

The only other issue I faced is that I was already using the VHF Spectra's front accessory connector for the audio panel hookup, and the PA module requires the use of that connector as well. After studying the schematics involved, I discovered that the accessory connector contacts were, for the most part, designed to simply daisy-chain to various accessories, and that no breaking of individual lines was required.

With that in mind, I took an old serial port distribution unit from Black Box Corp. (which I found at RE-PC), gutted it, and made up a nice little 'daisy chain' of my own from 25-conductor flat cable (or 26 with one conductor peeled off, if you like) and four D-sub 25-contact female plugs. A short, shielded all-25-conductors jumper goes from the Spectra to the box, and then both the PA module and the audio panel can connect to it simultaneously. There's even one extra connector left open for later use (like I have room or power capacity left for anything else!)

One amusing note. The original PC board I pulled out of the box frame was labeled with 'Copyright (c) 1983, Widget World, Inc.' No, I'm not kidding. I just wish I'd remembered to take a picture of the board before I pitched it.


There were quite a few companies who, through discounts, good customer service, or both, made this entire project a lot easier. In many cases, the project would not have been possible without their products or assistance. Speaking as the owner and sole proprietor of Blue Feather Technologies, and as one happy hamateur, I would like to publicly acknowledge the following companies for their most welcome assistance.

The Boeing Company - Surplus Sales Division (Wire, Cable, and subassemblies)

Earth Computer Technologies (Mobile Display)

Bill Stewart, ETA Circuit Breakers (High-current aircraft-style DC breakers)

Grainger Industries (Tools and shop supplies)

Graybar Electric Supply (Wiring supplies and tools)

Platt Electric Supply (NEMA Enclosures and tools)

SirenNet (Replacement lenses and lightbar parts)

SMC (Console and pedestal mounts and accessories)

Texas Industrial Peripherals (Mobile Keyboard)

Entire project completed, Phases 1-4. Last update 03-Jun-07.

Click here to return to the main page.