Experimental Parabolic Microphone

I've always wondered how well a parabolic mic works. Here is my experimental setup for testing such DIY parabolic microphone. It is a great weekend project and will let me experiment with high-gain / low-noise audio amplifiers. The heavily wooded area off my back yard is plentiful of singing birds.


Photographic tripod is used as mount for the parabolic dish. I had to construct a simple mounting bracket and a "focusing" contraption that will let me use different size and shape microphone elements.

I got the actual parabolic reflector from "the place where you can find anything" - eBay. The parabolic reflector is made out of polyethylene plastic. The diameter is 21 inches with focus point 4 inches from the bottom. It came as one solid reflector - I had to drill the mounting hole. I had a few ideas for mounting the dish - I wanted to be compact and simple so I decided on a single hole in the center.

The microphone mounting frame is made from semi-rigid coax (RG-402) and small PCB board for attaching the mic element. I used a threaded cable feed-thru to both - attach the dish to the bracket and mount the RG-402 frame with the microphone. A "sandwich" of metal and rubber washers - including two large and thick rubber washers provides "shock-mount" for the dish. The mic frame is fed through the threaded feed-thru using silicone cemented inserts.

The "focusing" rig lets me move the microphone element to the exact focus point of the dish. The focusing range is about 2 inches. Normally, the mic can be fixed in the focus of the parabola, but I am planing to experiment with different mic elements and they all vary in size and shape so I wanted to be able to adjust the mic frame with no hassle. Two spring-tensioned wing nuts fine-tune the mic frame.

Small plastic container is holding the battery pack, microphone preamp and 900 MHz FM transmitter (a.k.a Baby Monitor - For the initial testing and to validate the concept I just modified a baby monitor and then used my IC-R20 scanner to listen and record). Currently, I am working on my low-noise/high-gain preamp. The radio-channel link sort of works but the noise levels are way too high. Proper mic pre-amp and better (broadcast grade) FM transmitter are planned for the next stage.

Hand-held mode. Large plastic handle salvaged from old angle-cutter provides comfortable grip.
A word of warning - the surface reflectivity of the polyethylene dish is high enough to produce smoke from the mic's wind guard while I was playing with the dish and decided to verify the focus by pointing it at the sun. It took less than a second! I was able to act quickly and saved the guard from catching on fire :) (stupid move but the damn thing looks transparent :-)

Awl modification for Leatherman Wave

I had my original Leatherman PST (Pocket Survival Tool) for many years and when I decided to upgrade to Leatherman Wave, I was a bit disappointed to find out that they have omitted a very useful tool from the set - the awl. If you want to puncture a hole in a coconut, leather belt or an old plastic bottle with Gorilla Glue, a good awl will do the job just fine. The PST has a somewhat small but sturdy and useful awl so I decided to add one to my Wave. I've seen people grinding a 1/4" hex extension driver to make an awl for the Wave's bit driver. The problem with this - it takes A LOT of grinding and it is difficult to precisely adjust the thickness and shape of the base part so it fits and locks in the bit driver - Leatherman is using a modified (flat) version of the 1/4" hex bit with most of the new multi-tools in order to save space in the handle.
This mod is for the NEW Wave tool and it will not work on the old Wave (pre-2004) as the old model is missing the bit driver.

This is the starting point for my awl mod - the bit kit (#930368 - only $5 at Leatherman.com) for Leatherman MUT (Miltary Multi-Tool). Looking at the MUT tool I realized that the bits are compatible with the Wave's bit driver. The set includes 3 combo bits - short Slotted/Phillips, long (2.5") Slotted/Phillips and a long (2.5") hex 7/64 / Torx T15 bit. These bits will fit in Leatherman's standard bit driver (equipped on Wave, Charge, Surge and Skeletool)

I used the long Hex/Torx bit to grind my awl from. This bit is less useful to me than the other two. Leatherman is using pretty good tool steel for these bits as it took me well over half hour of grinding on my improvised grinding wheel (Electric Drill with grinding wheel mounted in the chuck). First, I used a scriber to draw the desired shape in the black oxide coating and used my Dremel Tool to cut off the tip and roughly correct the shape in order to reduce the time spent in grinding. The bit can be shaped in many forms - I wanted a heavy duty awl so I kept more metal around the tip and reduced the size less gradually. This resulted in a "straight-back knife blade" tip that is very strong. For lighter duty it can be shaped with more aggressive reduction of the size from tip to base (sharper, "needle-like" form). The remaining two bits from the kit can be shaped into other useful tools - like a miniature V-blade line cutter/wire stripper (the Wave has one already on the bottle opener), a sharper awl, a punch-down tool or a small pry tool.

As one can see - the overall thickness is pretty good, resulting in a strong and solid tool. Now, there is no need to use (and possibly damage) the knife blade when puncturing holes.

The awl inserted in the Wave's bit driver. It locks nice and firm in the bit holder with absolutely no wobble. For finishing the surface I used a drop of Perma Blue solution (Liquid Gun Blue). The bit has a black oxide coating and the selenium based gun blue blends in perfectly.

Here is another mod. The original Leatherman sheath has a side pocket for a small flashlight and inside pocket to fit in the the two plastic holders with the extra bits kit (#931014). Unfortunately, there is no space for the bit driver extender (#931009) or any of the long MUT bits (and my awl). I used a cheap aluminum (2x AA batteries) flashlight ($1 form AutoZone) to make a container for the long bits/driver extension. I just cut a portion of the aluminum tube of the flashlight body and capped the cut end. The original end cap (normally used to load batteries) serves as the container's threaded cap. The container slips nicely in the side flashlight pocket of the Leatherman sheath. I can fit the driver extender, two long MUT bits, one short bit, large needle and some thin steel wire and rolled up Band-Aid :-)

Wire Antenna Tension Breaker

Wire Antenna's enemy #1 is the wind (Corrosion being #2, Lightning is a topic of entirely different discussion).
In high winds, tall trees sway a lot - the taller the tree is - the bigger is the amplitude. What makes the matter worse is the fact that different trees sway with different frequency and amplitude due to the specifics of each tree - height, canopy, etc. The wind could also blow from different directions for each tree if they are far apart.
Wire antennas are often stretched between tree-tops where the swing is in it's maximum.
In other words, during strong wind almost everything works against the antenna, putting it at a great mechanical stress.
One solution for long wire antennas is to let it sag - the sag could provide enough slack in high wind situation so the antenna is never tensioned to the maximum thus reducing the chance for break. Such approach works fine for long-wire, end-fed antennas.
When it comes to dipoles, one would want the antenna as high as possible. In addition, preserving the flat-top geometry of the antenna also helps the radiation pattern so people tend to tension them a lot.
In order to protect my G5RV from breaking due to tensile stress and to reduce the unnecessary sag in calm weather at the same time, I made a "tension breaker" (it is more of a "fuse" actually)
The idea is very simple - to create an artificial "weak point". If high wind occurs and the sway of the tree-tops puts the antenna under excessive stress, the "tension breaker" opens at a predetermined tension load, releasing more slack in the antenna rope and relieving the stress by letting the antenna to sag. When the weather calms down, the "breaker" could be easily "reset", stretching the antenna back to it's original state.


The "tension fuse" is located near one of the antenna rope's anchor points. The actual "fuse" is two lengths of "50 lb test" Spectra Line braided (yellow) filament between two Quick-Links (each rated for 220lb load). The filament should break at a load >100 lb (2 x 50lb). A test sample broke at ~120 lb. I am using AWG #12 wire for the antenna (tensile strength ~ 220 lb) and 3/16' double-braided polyester antenna rope with break strength of 770 lb). Even if I de-rate the breaking load for the whole antenna because of the antenna insulators, knots, soldering etc., the filament "fuse" is still going to be the weakest link. Once it breaks, it will release a slack of approx. 6 ft of antenna rope (bottom-left on the picture). To "reset" it, I have prepared a couple of extra "tensile fuses" which can be installed between the Quick-Links in 5 min.
The trick is to have such weak point to break only at a load dangerous for the antenna and withstand the load of moderate wind conditions while keeping the antenna tensioned for minimal sag.

Antenna Launcher Digital Scope Part 2

These are the final touches to the Digital Scope System and the Antenna Launcher.

I made a hood out of a sheet soft closed-cell foam and attached it to the aluminum frame of the scope with Velcro for easy and quick installation/removal - takes 5 seconds to install it. This picture also shows the trigger support bracket and the safety strap for the trigger.

This hood works very well - it reduces dramatically the glare and improves the display contrast on a bright sunny day.

What's better way to calibrate the accelerometer sensor than Mother Earth's gravitational force? I glued a small Spirit Level to the frame to aid the GeoCam software calibration. After the scope is installed and level (according to the Spirit Level), the GeoCam calibration routine is executed to establish 0 degree pitch reference point.

Update on performance: The antenna launcher and the scope system work fantastic and I couldn't be happier! At 40 psi (less than half of the maximum 100 psi pressure), shooting at a very steep angle (75 degrees) and towing a line, I was able to go over a 110 ft tree with huge reserve in the trajectory. The scope allows for repeatable and well controlled launches - I was able to produce a nice group - 3 consecutive launches where the ball falls within 2-3 yards area every time with peak height of the trajectory of over 150 ft (4 oz ball, no wind, no line). Spray of silicon lubricant on the ball, lubricates the inside of the barrel too and makes for easy loading and probably decreases the friction during launch - with Schedule 40 2.5" barrel the tennis ball is a tight fit - SDR-21 type pipe is recommended for better fit and weights less but it is also less sturdy.

Digital Scope for the antenna launcher

Sometimes great ideas shine, just because of their simplicity and I think I had one the other day. (You can tell that I am modest person :-). I am claiming to have the most sophisticated aiming system of any pneumatic antenna deployment apparatus out there.

Here the Head Up Display of my "Augmented Reality Digital Scope". aiming at 56 degree up at a tree-top in my front yard. The scale for pitch (gun barrel angle) is displayed on the right.

The Scope mounted on my "steam-punk" antenna launcher. I know! The mount of the scope should have been made out of wood.

Front view. The Scope mount is attached with Velcro tape to the barrel for easy installation. Calibration of the scope is possible while mounted but it can be performed before installation on a perfectly level surfaces for greater precision.

This is the Scope's quick mount. I fabricated the mount out of aluminum L-stock. A piece of scrap 4" PVC pipe was used for the bottom part. I softened the plastic using heat-gun and formed it to fit tight around the 2.5" barrel. Small Velcro straps are used to secure the instrument in place. It takes about 30 seconds to install or remove the actual scope .

Cross-hair view down the barrel. Scope is secured in the mount and installed on the antenna launcher. Note the blue closed-cell foam padding between the barrel and the mount. It gives a very firm, non-slip grip for the mount

Ta-Daaah! Here is the secret :-))) The Scope is actually my Android OS smart-phone and it cost me absolutely nothing :) (I already had the phone). That's why it was important for me to be able to easily install and remove the phone from the mount. I fabricated the mount to fit my specific model phone - T-Mobile G2 (HTC Desire Z) but it can be modified for any phone.

Another view of the mount. Note the little cut-out for the phone's digital camera. The construction is very light and sturdy. The phone has a gel-skin to protect and further enhance the mounting.

The heart of the aiming system is a free app called GeoCam v.2.07, available on the Android Market or it can be DL from here. I came across this app accidentally and my first thought was - this could be useful for something one day (it used to be called Theodolite). It displays a wealth of information and it can triangulate objects in order to measure distance and height. The phone's accelerometer sensor is used to measure pitch (essential!!) and roll (the roll is kind of useless for the aiming system). It is using the internal magnetic sensor for the heading (compass) and GPS receiver for coordinates.
All this information changes dynamically and it is super-imposed over live video image from phone's digital camera. One can adjust colors of the HUD data and control the iris (exposure compensation) - very useful for use in bright, sunny day. (I am thinking of making some sort of rigid hood for the display to enhance further the contrast in bright ambient light and reduce glare). There is also a nice calibration procedure - very useful to account for differences between each installation.

The Digital Scope system works fantastic and brings the spud guns to the 21st century!! I think I'll apply for a patent on this one:)

Update: I am currently working on a BBTS or Basic Ballistic Trajectory Solver for the Android OS. The idea is to be able to input the distance to the tree, assuming the tree will be located under the peak of the trajectory, Tree Height + some padding , Tennis ball weight (for drag force calculation, I have data for the average drag coefficient of a tennis ball), Air density/Altitude (again for drag) and additional drag value (caused by line, the ball is pulling - there is no easy way to model this so it needs to be determined experimentally and entered as correction to the drag force)

The output will be a firing solution - Angle and required Muzzle Velocity. The tricky part will be to establish the correlation between the air tank pressure and the initial velocity of the ball leaving the barrel, but I'll work on a way to figure it out (police radar?).

"Say Hello to my little friend!"

Update: If you are looking to buy an Antenna Launcher for Field Day, please check this post.

Tony Montana's famous words as the answer to the question: How do you get an antenna rope over a 100+ ft tree-top? Forget about slingshots or bow and arrows. Even the crossbow with optical sight is so last century :). Enter: the pneumatic antenna launcher or shall I say "pneumatic blunderbuss";-)Here is the result of a couple of evenings spent in the garage - cutting, gluing and painting PVC pipes.
This antenna launcher is based on (WB6ZQZ) Alan Biocca's CSV19 with some modifications / improvements on my part. His web site (http://www.antennalaunchers.com/) is an excellent source of information on these launchers and it has very detailed build instructions. I had to do something about the white PVC look which I REALLY hate! The paint job was inspired by K4ICY and his "Steampunk" antenna launcher.
The main changes from Alan's CSV19 design are:
- Slightly larger compressed air tank - my launcher is using 10 inch length of the 4" diameter pipe for the tank vs. Alan's 8 inches. The reducing coupler I am using as part of the tank gives a little extra volume too.
- Longer barrel - 18.5 inches vs. Alan's 16 inch barrel - I had lengthen the barrel a bit in order to account for the larger pressure vessel and have enough clearance for the Zip Reel.
-More reliable and safer pressure vessel - instead of drilling a hole for 1" pipe and epoxy gluing the pipe for the high pressure outlet in a 4" end cap, I am using a 4" to 2" reducing coupler and then 2" to 1" reducing bushing as part of my pressure vessel. Another advantage is that I don't have to drill precision large diameter hole - unfortunately I don't have a lathe.
-More reliable and safe coupling between the barrel and the high-pressure pipe - I am using 2.5" to 2" reducing coupler and 2" to 1.25" inch reducing bushing. It is much easier to assemble the launcher that way! Alan's design yields for drilling a 2.5" end cap and epoxying the 1.25" inlet (actually, a 90 degree elbow) in the hole
-I made the spacer between the pressure tank and the barrel out of two pieces PVC, sliced from 4" pipe scrap. I adjusted the curvature of each piece to follow the outside diameter of the corresponding pipe and glued the pieces back-to-back.
-In a moment of sheer brilliance, I came up with the Augmented Reality Digital Scope. The HUD (Head-Up Display) on the scope shows the firing angle of the barrel (pitch), heading, roll, and geographical coordinates. It is also capable of measuring distance and most importantly height of an object (tree). (Scope is not shown on the picture above)This is probably the most significant improvement to the launching system I am willing to take credit for as it allows to correct your shots in a precise manner by adjusting the exact angle of launching. About the only thing I am missing is on-screen display of the air pressure in the tank. More on this in a later post...

The main source of PVC hardware for this project was http://flexpvc.com/. Trigger, pressure gauge and Schrader valve are available from McMaster (the trigger is part 6852K11). Rainbird 100DV-SS sprinkler valve is from eBay. The bow-fishing zip reel is from an online archery store. Brass fittings, brass street elbow and aluminum stock (for the Zip reel mount and support strut) - all from Home Depot. Tennis balls and Spectra line (150 yards spool / 50 lb test) from Sports Authority. For all threaded connections (sprinkler valve to trigger valve, pressure gauge, Schrader valve) one should use the yellow type Teflon tape - it is made specifically for gas/high-pressure applications and seals much better than the standard white plumber's tape. I hated the rattling sound of the coins (used to bring the weight to 4 oz) inside the tennis balls so I injected the balls with polyurethane foam (used to fill gaps). For the tennis ball tie, I used a loop of string with a knot, drilled a penny right in the center and inserted the loop thru the hole. The knot should be large enough so it cant go through the hole. Then I inserted the penny vertically in the tennis ball thru the narrow slit I previously made. When I pull on the loop, the penny wedges flat across the slit - this solution works just fine and after I filled the ball with foam there was no need to stitch the slit - the foam glued the slit and the pennies inside.


This picture shows the installation of the Saunders Bowfishing Zip Reel. Two aluminum bars are attached to the zip reel and the 2.5" coupler is mounted in the center with countersunk screws. (the bottom side of the coupler was filed flat to form two "saddles" for the mounting bars). It is loaded with 150 yards of high-visibility Spectra-Line (50lb test). I even installed a little cutting blade (the yellow thing on the bottom) for added convenience. This line cutter was part of the Spectra Line packaging - i just had to cut it out from the plastic spool-holder.

Update: In the original design the trigger valve could loosen or over-tighten if one is not careful - the valve is not fixed - it relies entirely on the thread and because it must not go all the way in (the street elbow is just partially threaded, 2-3 turns max), accidentally rotating the valve in either direction could cause a variety of unwanted effects.

Alan, WB6ZQZ suggested to use a strut to support the trigger so this is what I came up with. Small piece of curved PVC (scrap 4" pipe, heat gun, 2.5" pipe used as a form for bending and sanding) is drilled for a countersunk screw then glued to the barrel with the screw in place to create an anchor point. An aluminum bar is used as a strut between the anchor point and a brass trigger outlet extension. The red cable-tie is the "safety" (currently in ON position) - it prevents accidental operation of the trigger.
Another solution for the strut anchor point is to drill, countersink and install the screw from inside of the 2.5" to 2" coupler BEFORE the 2.5" barrel pipe is glued. The hole for the screw should be drilled in the middle (or closer to the edge) of the 2.5" portion of the coupler and the countersinking should be deep enough to allow for smooth installation of the barrel after the screw is inserted. There is not much clearance for right-angle drill -the countersink can be done with RA Dremel attachment or manually by hand.