Governor Dick Hill APRS Station

for the Appalachian Golden Packet Event

Bob Bruninga, WB4APR@amsat.org, Annapolis, MD

2013 Planning: Bob, WB4APR and Rory Schaffer NJ3U are leading the effort this year. See the summary spread sheet.

2012 Planning: Bob and Rorie manned GD Hill for the combined GOlden Packet and Operation ON-Target experiment. But Bob lost most of his kitchen pass and had to do set-up early and then leave early.

2011 Success: We probably get the "weakest link" this year. WB4APR Bob and KC2UML Rorie along with the mule-pack AJ Bruninga, WA4APR assaulted GD Hill again in 2011. But as we got all thgear to the top of the hill, we were joined by Murphy, N0CALL. As an invisible uninvited guest, he insisted on climbing the tower with us. The rest is history. See all of his contributions for 2011.

Despite his assistance, we immediately heard all the way from SPRINGER mountain in Georgia and eventually heard from Mount Washington in NH! Problem is, we did not realize we were not digipeating until 1450! Fixed that, but then there were only a few beacons before ending time.

2010 SUCCESS! Bob and Rory did it again this year. See the 2010 GD Hill Summary Report and the 2009 Report.

See the Golden Packet plan. . This is one of the 15 hill-top sites from Georgia to Maine that we use on the third Sunday in July for the annual GOlden Packet Event which attempts to relay text messages using hand-held radios the 2000 mile length of the Appalachain trail and other long linear trails in the USA. This is a no impact Leave-No-Trace type of event of a few individuals at each site. . Other hikers equipped with APRS ham radios are welcome to participate during the time slot for hikers.

Governor Dick Hill is an isolated hill just to the east of Mt Gretna, PA that rises above surrounding terrain in the relatively flat area south east of Harrisburg. It is ideally located between the Maryland Mountains and Eastern PA. It can get over the ridges northeast to Camelback Mountain in Big Pocono State Park with a little help from knife-edge refraction. Although it has tall trees, it is a public park site with a public climb-up lookout tower! The Radio team has to hike up 1/2 mile on a trail to the tower and carry the equipment in back-packs.

ALTITUDE: . . . . . . . . . . . 1150 feet

POSITION:. . . . . . . . . . . . 4014.77N / 7627.33W

LINK NORTHEAST:. . . . Camelback (Big Pocono SP). has 146.865 T100. See RF data
. . . . . . . . . . . . . . . . . . . . . . . And Highpoint, NJ as a possible 1 hop bypass to MT Greylock!
LINK SOUTHWEST:. . . . MD Mountains (has WN3R-10 on 145.77 MHz). See RF data

VOICE REPEATER:. . . . . 147.315+ T082
ECHOLINK NODE: . . . . W3NOD-R 446.925 T082 #146925 (Laporte PA)

INTERFERRENCE: . . . . None. No RF on site

TEAM LEADER:. . . . . . . Bob Bruninga, WB4APR (2010,2009)
. . . . . . . . . . . . . . . . . . . . . . Rory Shaffer, KC2UML - schaffer250 * comcast.net (2010 & 2009)
. . . . . . . . . . . . . . . . . . . . . . Stephen M. Shearer wb3lgc * verizon.net (?)
. . . . . . . . . . . . . . . . . . . . . . AJ, WA4APR, and Bethanne, WE4APR (2009)

COMMENTS 2010: . We had solid comms south through MD Mountains to White Rock Cliffs using HOP7-7 paths. But WHTRCK was not digipeating onward, so we used their actual call WHTRCK-6 and were able to link down to AO mountian near Roanoke and back. We never heard anything to the north.

GD Hill is a lot of fun. Clint told us about this site and its great observation platform with its steel-bar cage at the top (dimensons) does not pose any problem for a vertical 2m antenna on a short 12 foot mast stuck up through the bars see photo). The operating position is convenient if you bring your own table and chair as shown to the right (2009). In 2010, we left the furniture at home and sat on a cushion!

Power Planning 2010: This year we decided to schlep sufficient batteries and backups up the hill. We brought two 20 AH gel-cells and three 4 AH Nicad packs. But with the rain, we never even hooked up the laptop and so ran the entire event on only one battery and on charging it aftert the event, we had only used a total of 2 AH or less! Next year we will carry less LEAD up the mountain! (and no laptop!)

Full Power in 2009! The potential for poor weather suggested a backup plan B for our original plan A solar power. In the google earth view above you can see that it is over a half mile from the nearest parking space. Our plan B was to string a 3200' run of #20 wire up from my car and use a single-wire-earth-return (SWER) technique to deliver about 100 Watts to the laptops and battery charger.

The bars are about 7.5 feet tall and have a ring about 4 feet above the standing floor. The GD Hill web page provides lots of detail including reference to an old radar site (approximately where the white "north" arrow is on the image above. It evens has an interactive 360 view video!

The first site visit was on 26 April 2009 by Paul Shultz, K2PCS, the son of Roger Shultz who was at the Camelback end of the link. The ladders are vertical, 7 feet between landings with a "back board" for easy climbing but preventing a backpack. But just handing bags up each flight is easy. See ladder photos vertical view, and one landing to another.

RF TEST REPORTS: Paul reported that Roger on Camelback running 50 watts and a beam had noticable "static" but he could hear him on a 2m HT quite clearly. On a 1-5 scale of 5 being clearest, he was probably 3.5-4. This tells me that we will definately need a good 3 dB at least vertical antenna to get above the iron cage and between the bars. Bob followed up with another test on 20 June and made detailed link measurements. It showed rock solid links to Maryland but that a beam antenna was needed at Camelback. See WB4APR's test report. UHF takes full power to omni's.

See Trail Entrances radar road and old rail bed. The hike is about 1/2 mile or 3/4 mile depending on the trail. Parking is available on the side of the roads near these trail heads.

Since both of the RF paths to the SW and NE are very long shots with grazing ground clearance, a good antenna at least a wavelength above the iron cage is a must. Fortunately, the Hill is a nice curvature at the top giving ample height above average terrain in most directions. (see view). The image below shows the path from the Maryland Mountains to GD Hill, followed by the RF path analysis.


DETAIL LINK ANALYSIS BY KX4O: Be sure to follow the two RF links at the top of this page to see the excellent and very detailed RF link analysis performed for every link in this project by KX4O John Huggins. Each RF link shows the terrain map elevations, the AT trail in yellow, the RF path loss itself including fresnel zones and finally a statistical plot of the probabilities of link availability. We will very much be looking at the results of this project compared to the pre-event analysis and predictions. All of KX4O's links are shown on his AT Links Page.


See also N1NCI's RF analysisMD-to-GDhill and GDhill-to-Camelback

There is clearly an excellent RF path from the Maryland Mountains. The RF analysis used Salamander Rock which is not the highest, but is in the vicinity of easily accessible sites. The analysis towards Camelback shows the challenge of crossing the Pennsylvania hills onward to the northeast to Camelback Mountain in Big Poconos State Park.

This path has two significant ridges to overcome but on a 4/3rds RF Earth projection, the clearance is good. The path loss means we should be able to do the link on LOW POWER. The sensitivity of a D700 or D7 is around -117 dBm for good packets. SO this and a 10W medium power was used for analyzing all links.

EQUIPMENT AND POWER: A typical AT Golden Packet Digipeater consists of a D700, a mast as required to clear the trees, and a power source. The mimimum event is 3 hours long, but the optional phases can last up to a total of 5 hours, so ample AMP-Hr capacity should be considered. The D700 draws about .6 amps on receive (D710 draws .7), and about say 4.5 amps on TX (medium power 10W). Under normal operation, the duty cycle of an APRS digipeater is typically 20% so the average power demand of the AT digipeater will be about .2 * 4.5A +.6A or about 1.5 Amps per hour. You can either carry a 7.5 Amp Hour battery to the mountain top, or if the day is sunny, a 1 Amp-Hr battery and a 1.5 amp solar panel as shown. Or use the single wire earth return system shown below if you can carry enough wire to get back to the car.

SWER REPORT: Our plan was to carry the wire spool to the top of the tower and then pull wire off of it down hill 1/2 mile to the car. But the drag through the woods was too great. Even with one person half way down the hill, the pullling force was too great and broke the wire. Finally took 3 people and nearly 1.5 hours to string the wire. (Next year, we will simply unspool from the bottom up as we go). Once connected, the laptop power supply simply cycled on and off.

This cycling was due to the nature of Laptops that when connected to external power, first do a peak current load test to measure the current capabilty of the power supply. It then uses that information to chose its operating and charging profile. Problem is, with the very high impedance of the SWER system, this quick peak current causes the voltage to drop and the laptop to dismiss the power supply as a viable current source. So it cycles and tries again, and ends up in an infinite loop. We solved this by momemtarily swamping the oscillations by paralleling a 12v battery and/or solar panels to let the laptop at least complete its boot.

From then on, we had reliable power and the laptop ran for the remainder of the event. Retrieval of the 3200' of wire in the rain went quickly since we cut it in the middle and were able to then easily re-spool each piece from the down-hill end.

See ya next year!

Bob, WB4APR

See my other GENERAL page on APRS applications and Ideas on the AT


Return to the APRS HOMEPAGE or SiteMap.