70cmQuadiDesign

From NebarnixWiki
Revision as of 12:03, 18 April 2015 by NebarnixWikiSysop (talk | contribs) (Added V3)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigationJump to search
The printable version is no longer supported and may have rendering errors. Please update your browser bookmarks and please use the default browser print function instead.

Note: This page is a reconstruction to a now defuct Univeristy of Idaho nearspace balloon wiki article I posted in 2008

70cm Quagi

Project Status

V1 Flight Tested on F08L01, two built : status unknown
V2 One built, no longer in service
V3 Preconstruction

Problem Statement

Ginger is a bidirectional data routing system equipt with 70cm j-pole antennas. These antennas have a gain of around 2.5dBi and using quarter wave whips the link range is around 37 miles at full power (250kbps) at 10 Watts of output power. To extend the range of operations which might be due to a run away balloon, damaged transmitters, or geographical obstructions, we have built two directional quagi antennas.

The quagi antenna consists of a quad driver (a driven element and a reflector) which already has a higher gain than a dipole fed yagi and reflector. This means that by feeding the same number of directors, a quagi will outperform a yagi at the cost of not being a completely flat design (which is merely a transport concern anyway). The 8 element quagis we have built should have a gain between 12 and 13 dBi and should increase the working distance to 80 miles at 250kbps.

Also, this makes a really good satellite antenna. While I no longer am involved with UIVAST, I still want to be able to use these for sat usage. These are linearly polarized antennas, but they have a lot of gain and are easy to do.

Photos

V1

V2

V3

Dimensions

V1

The dimensions, in a 1977 article by K6YNB in QST April 1977

Element Lengths 144.5 MHz 147 MHz 222 MHz 432 MHz 446 MHz
Reflector (all No 12 TW wire, closed) 2200 mm loop 2159 mm 1432 mm 711 mm 689 mm
Driven Element (No 12 TW, fed at bottom) 2083 mm loop 2032 mm 1359 mm 676 mm 657 mm
Directors 913 mm to 889 mm in 5 mm steps 897 mm to 873 mm in 5 mm steps 594 mm to 568 mm in 3 mm steps 298 mm to 291 mm in 1.5 mm steps 289 mm to 280 mm in 1.5 mm steps
Spacing
R - DE 533 mm 521 mm 346 mm 178 mm 173 mm
DE - D1 400 mm 391 mm 260 mm 133 mm 130 mm
D1 - D2 838 mm 826 mm 546 mm 279 mm 272 mm
D2 - D3 445 mm 435 mm 289 mm 149 mm 144 mm
D3 - D4 663 mm 651 mm 432 mm 222 mm 215 mm
D4 - D5 663 mm 651 mm 432 mm 222 mm 215 mm
D5 - D6 663 mm 651 mm 432 mm 222 mm 215 mm

V2

The deisgn has been optimized for 437MHz using NEC2. The dimensions are as follows:

RWIRE = .00258826 wire for quads diameter
RROD = .003175 yagi director rod diameter
RE = 0.755962 Total Length of Reflector Element
DE = 0.718693 Total Length of Driven Element
D1 = 0.280124 Total length of Director element
D2 = 0.28434 Total length of Director element
D3 = 0.283594 Total length of Director element
D4 = 0.276571 Total length of Director element
D5 = 0.244994 Total length of Director element
D6 = 0.283213 Total length of Director element
RE2DE = 0.177867 Distance from RE to DE
DE2D1 = 0.138696 Distance from DE to Director
DE2D2 = 0.383653 Distance from DE to Director
DE2D3 = 0.547346 Distance from DE to Director
DE2D4 = 0.851997 Distance from DE to Director
DE2D5 = 0.984569 Distance from DE to Director
DE2D6 = 1.092358 Distance from DE to Director

V3

I wanted to model it with tubing of a different diameter, and maybe even model a tuning stub. I changed all diameters to 1/8" and added a tuning stub.

CM  QUagi 437Mhz Sattelites fiddled with and optimized by KE7PHI from ARRL handbook model
CM  RP 0 181 1 1000 -90 0 1 1
CM  RP 0 1 361 1000 90 0 1 1
CE
SY RWIRE = 0.0015875	'wire for quads diameter
SY RROD = 0.0015875	'yagi director rod diameter
SY RE=0.74	'Total Length of Reflector Element
SY DE=0.702517	'Total Length of Driven Element
SY D1=0.290414	'Total length of Director element
SY D2=0.29013	'Total length of Director element
SY D3=0.285381	'Total length of Director element
SY D4=0.276766	'Total length of Director element
SY D5=0.245045	'Total length of Director element
SY D6=0.287901	'Total length of Director element
SY RE2DE=0.177909	'Distance from RE to DE
SY DE2D1=0.138761	'Distance from DE to Director
SY DE2D2=0.380606	'Distance from DE to Director
SY DE2D3=0.554942	'Distance from DE to Director
SY DE2D4=0.851939	'Distance from DE to Director
SY DE2D5=0.982862	'Distance from DE to Director
SY DE2D6=1.096131	'Distance from DE to Director
SY STUBLEN=.0635
SY STUBPOS=0.030165
SY GAP=.01
GW	1	6	-RE2DE	-RE/8	-RE/8	-RE2DE	RE/8	-RE/8	RWIRE
GW	2	6	-RE2DE	-RE/8	-RE/8	-RE2DE	-RE/8	RE/8	RWIRE
GW	3	6	-RE2DE	RE/8	RE/8	-RE2DE	RE/8	-RE/8	RWIRE
GW	4	6	-RE2DE	RE/8	RE/8	-RE2DE	-RE/8	RE/8	RWIRE
GW	20	7	0	-DE/8	-DE/8	0	-GAP	-DE/8	RWIRE
GW	21	7	0	GAP	-DE/8	0	DE/8	-DE/8	RWIRE
GW	6	6	0	-DE/8	-DE/8	0	-DE/8	DE/8	RWIRE
GW	7	6	0	DE/8	DE/8	0	DE/8	-DE/8	RWIRE
GW	8	6	0	DE/8	DE/8	0	-DE/8	DE/8	RWIRE
GW	9	6	DE2D1	-D1/2	0	DE2D1	D1/2	0	RROD
GW	10	6	DE2D2	-D2/2	0	DE2D2	D2/2	0	RROD
GW	11	6	DE2D3	-D3/2	0	DE2D3	D3/2	0	RROD
GW	12	6	DE2D4	-D4/2	0	DE2D4	D4/2	0	RROD
GW	13	6	DE2D5	-D5/2	0	DE2D5	D5/2	0	RROD
GW	14	6	DE2D6	-D6/2	0	DE2D6	D6/2	0	RROD
GW	22	3	0	GAP	-DE/8	0	GAP	-DE/8-STUBPOS	RROD
GW	23	3	0	-GAP	-DE/8	0	-GAP	-DE/8-STUBPOS	RROD
GW	22	3	0	GAP	-DE/8-STUBPOS	0	GAP	-DE/8-STUBLEN	RROD
GW	23	3	0	-GAP	-DE/8-STUBPOS	0	-GAP	-DE/8-STUBLEN	RROD
GW	24	1	0	-GAP	-DE/8-STUBPOS	0	GAP	-DE/8-STUBPOS	RROD
GE	0
LD	5	0	0	0	37700000	'Copper wire
GN	-1
EK
EX	0	24	1	0	100.00	0	0	0
FR	0	0	0	0	437	0
EN

Performance

V1

  • At 437MHz the SWR is measured at 1.0 for Quagi Orange and 1.2 for Quagi Green (due to coax losses from crappier cable)
  • Raw data showing rough measurements of directionality Excel XLS file

V2

  • At 437MHz the gain should be 13.43 DBi best theoretical. SWR S/B 1.07

Insights

  • Wind is a real problem and keeping both aimed at the same location is not a trivial task. Mount both on a single crossbar and have a lockable central pivot.
  • It would be pretty neat to read the GPS data from the CNDH board (already scheduled) and have a QT module to decode them and steer the antennas in the right direction.
  • I don't trust ARRL - always simulate those "best anntenna ever" designs first before building!