Antenna Testing, Home Built, simple J-Poles

[g7pnu]

Antenna Testing, Home Built, simple J-Poles
Aim:
To save you reading this test: – build a simple ¼ wave spider as I found that 95 times out of a 100 it will outperform a simple J-Pole.

The first set of results which were for ¼ Wave antenni and be found here. “Antenna Testing, Home Built, ¼ waves”; Antenna Testing, Home Built , ¼ waves.

As before, before installing the Flight Aware Rx on the outside of my home I wanted to check which type of antenna gave the best reception. I would like more gain than a ¼ wave (no gain) but I have a small 80m-110m wooded hill one mile (2020m) to the east of me. So the two antenna types I selected to for testing where ¼ waves and J-poles as they appear to be dimensionally stable, so should easy to trim and be suitable for mounting in the open air. Both have relatively high take-off elevations, so the effect of the hill will be minimised. Coax Colinear antenni have more variables in their construction, without test equipment the performance of a colinear would be a shot in the dark, for me at least.

Dates:
Simple J poles from 16 March 2018 to

Links:
Calculator:

Slim Jim / J Pole Antenna Calculator

Books:
“The Radio Communication Handbook” by RSGB ISBN: 1905086083,9781905086085 Chapter 16 “Practical VHF Antennas”.

Tests J-pole:
All tests were performed with the Rx in the same location and the data for each test was collected over 24 hour period, using both the Mode S Mixer data capture application and the Flight Aware flight counters.
I have included the results of a Coke cantenna as a “Control”, so that you can compare the coverage of your standard Coke can cantenna with the J Pole designs sown below. The Coke can cantenna is a great reference point, as we can all build identical antanni and compare our results no matter which country we live in.

In order to ensure all the measurements of the elements where repeatable, the elements were measured from the “centre of the bend” using an engineers steel rule. A pair of vernier callipers were used to measure the diameter of the brass rod, length of the stub and the separation of the two elements.

Tuning the J Pole without electronic test equipment needs to be performed by trimming the elements 1.0mm - 0.2mm each day and then recording the aircraft contacts for that day: a slow process. I limited the changes to one each day so that I could identify the effect of each individual change.

The final element lengths were calculated using a spreadsheet. (The most online calculators don’t provide enough resolution on the stub length). Searching the web revealed the separation to go for was between 6mm and 8mm and the feed point 6mm and 10mm
The diameter of the rod: 3.2mm
Element separation : 6.0mm (base) – 6.3mm (tip)
Feed point : 7mm

BEST RESULTS FROM:
198.8mm * 6mm * 66.5mm : TEST J16
198.1mm * 6mm * 66.2mm : TEST J19

The line below shows the difference in the measured length (slightly diagonal) versus the “true vertical length”. It was not practical to measure the “true length” with a steel rule or calliper, thus the measurement was taken from the central bend to the element tip. I found I could see the difference in coverage if the elements changed by 0.1mm - 0.2mm.
vertical length = Sqr( (67.5mm)^2 - (3mm)^2 ) = 67.23mm

1. J1, J-pole, copper wire, 207mm x 69mm , feed 10mm
2. J2, J-pole, copper wire , 207mm x 69mm, feed 7mm
3. J3, J-pole, copper wire , 201mm x 68mm, feed 7mm
4. J4, J-pole, brass rod, 207mm x 68mm
5. J5, J-pole, brass rod, 205mm x 68mm
6. J6, J-pole, brass rod, 204mm x 68mm
7. J7, J-pole, brass rod, 203mm x 68mm
8. J8, J-pole, brass rod, 203mm x 66.8mm
9. J9, J-pole, brass rod, 202mm x 66.8mm
10. J10, J-pole, brass rod, 201mm x 66.8mm
11. J11, J-pole, brass rod, 200.5mm x 66.8mm
12. J12, J-pole, brass rod, 200.5mm x 66.5mm
13. J12, J-pole, brass rod, 200mm x 66.5mm
14. J12, J-pole, brass rod, 199.5mm x 66.5mm
15. J12, J-pole, brass rod, 199.2mm x 66.5mm
16. J12, J-pole, brass rod, 198.8mm x 66.5mm
17. J12, J-pole, brass rod, 198.8mm x 66.5mm
18. J12, J-pole, brass rod, 198.5mm x 66.5mm
19. J12, J-pole, brass rod, 198.1mm x 66.2mm
20. J12, J-pole, brass rod, 198.0mm x 66.2mm
21. J12, J-pole, brass rod, 198.0mm x 66.1mm
22. J12, J-pole, brass rod, 198.0mm x 66.0mm
23. J12, J-pole, brass rod, 197.8mm x 66.0mm
24. J12, J-pole, brass rod, 198.5mm x 66.0mm

Test Equipment:
Same as used in, Antenna Testing, Home Built , ¼ waves
Rx: Noolec SMArt, Aluminum Enclosure, 0.5PPM, SMA conector (2017)
Filter: 1090Mhz surface mount, 50 Ohm, SMA connector Passband: 1086 ~ 1094 MHz
Processor: Raspberry Pi 3
Operating Sys: PiAware 3.5.3 with Mode S Mixer
USB cable: Lindy 100mm from Rx to the processor
Arial main cable: 1000mm double screened satellite cable, 75 Ohm
Arial pig tail: Lindy 100mm coax, SMA to F type panle mount
Arial connector: F Type 75 Ohm
Data comms: Pi Ethernet to the power line (Pi WiFi diabled) .
Location: Inside concrete block built garage, tile roof, arial 2m above floor
Ground 97.5m Above sea level
Note: There is only one impedance step change in the Rx line 75 Ohm to 50Ohm, which occurs at the join between the satelite cable the pig tail. Not an significant issue as the rig is not a transmitter.

The test location was the same for all tests The location is not ideal thus the results of one anttena should only be compared to results of another antenna in this test.

Results J Pole:
The J-Poles take a great deal of time to tune and only when tuned do they perform better than a basic ¼ wave. 1.6mm copper wire was found to be too flexible to form a stable J-Pole, due to the tension in the coax connections. 3.2mm brass rod was found to be easily worked and did form a dimensionally stable J-Pole. A variation in length of 0.1mm is all that it took to take the j-pole for good to average coverage.

Conclusion:
Without the use of electronic test equipment, it is possible for a home builder to construct ¼ wave ground plan antenni, which can be relied upon to give good or acceptable coverage. The J-Pole is likely to give worse coverage than a ¼ wave antenna unless it is tuned with electronic test equipment or if no test equipment is available, painstakingly slowly by test, trim test.

The recommendation is for a new home builder to first build a ¼ wave ground plane “spider”, using accurate measurements and good construction techniques, as this can be relied upon to give good coverage.
After 3 weeks of tuning my simple J-Pole I found it performed slightly better than my best ¼ wave (which was never turned up!). I suspect the J-Pole will go off tune as soon as it is placed in a PVC pipe radome.

Control Test:
¼ wave Coke Cantenna /
garage /
13-16/03/2018 /
Tues 20:00 – Friday 20:00
Arial element length: 68.5mm x 1.0mm brass wire
Diameter: it’s a coke can
Height: 69mm
Connector: F Type

J-Pole Tests 1.6mm copper
Tests J1 → J3

Initial designs calculated using the M0UKD online calculator.
The initial tests were performed using 1.6mm copper wire. Copper wire of this diameter was found to be too flexible as the gap between the elements changes if the tension on the coax cable changed. 3.2mm brass rod was found to be more physical ridge and stable thus enabling measurements and tests to be repeatable.

J Pole 3mm Brass rod
Tests J4→J23
3.2mm brass rod bent around a 6mm drill to form a 6mm space between the “Inside” edges of the vertical elements.
Satelite coax attached to elements using the innards of a main terminal box connector. The nards where removed and then cut in half to form the two connectors. Coax cable comes in from the top.
Top of the terminal is 7mm from the “Inside” edge of the bottom bend.
The whole antenna is isolated via a plastic terminal connector (with all the metal internal connectors removed) and a wooden dowel.

Test J6:
garage /
2018-03-19 19:54:09
to Tuesday 20th (22 hours 15 minutes)

Change: cut 1mm from main.

204mm * 6mm * 68mm and
feed 7mm from base
3.2 mm Brass rod
velocity factor = 0.990

Total messages / min : 8.390 x 10^3
Total messages: 11200,000
Max distance: 113 n miles (Azi 110 deg)
Max dis 2nd lobe: 100n miles (Azi 200 deg)

Test J10:
garage /
2018-03-23 18:39:05
to Saturday (1 day 37 minutes)

Change: cut 1mm from main.

201mm * 6mm * 66.8mm and
feed 7mm from base
3.2 mm Brass rod

Total messages / min : 8.182 x 10^3
Total messages: 12100,000
Max distance: 108 n miles (Azi 120deg)
Max dis 2nd lobe: 97n miles (Azi 220deg)

Test J16:
garage /
29/03/2018 /
to Friday (1 day 1 hours 25 minute) “Good Friday”

Change: cut .4mm from main.

198.8mm * 6mm * 66.5mm and
feed 7mm from base
3.2 mm Brass rod

Total messages / min : 12.177 x 10^3
Total messages: 18571,000
Max distance: 134 n miles (Azi 120deg)
Max dis 2nd lobe: 118n miles (Azi 210deg)

Test J19:
garage /
1/04/2018 /
to Monday (24 hours 1 minute) Easter Monday

Change: cut 0.4mm from main.

198.1mm * 6mm * 66.2mm and
feed 7mm from base
3.2 mm Brass rod

Total messages / min : 11.519 x 10^3
Total messages: 16600,000
Max distance: 135 n miles (Azi 120deg)
Max dis 2nd lobe: 126n miles (Azi 200deg)
Max dis 3rd lobe: 109n miles (Azi 280deg)

[abcd567]

@g7pnu
Happy Experimenting! Keep it up!

[triggers]

@g7pnu
Great experimentation

I’m not that far from you in Benson. I think if you can get an antenna outside that will improve your range a great deal and give you a better comparison. Can’t you put the J pole straight outside without a radome? A bit of hot glue on the coax connection should keep it waterproof.

@g7pnu

Coax Colinear antenni have more variables in their construction, without test equipment the performance of a colinear would be a shot in the dark, for me at least.

I agree, but a coax collinear is worth building. I’ve made many but only got a couple to work properly. I don’t have any test equipment. I have a 12 element on my brother’s roof in Farnborough. https://uk.flightaware.com/adsb/stats/user/triggers#stats-15612 which has been up there for more than a couple of years.

[g7pnu]

Hi @triggers,
I had spotted that you weren’t far away from me. As you were local I was using your high flight statistics as my goal but I had no real idea of the antenna type, height above ground or if it was inside or outside.
I really like the idea of a super J, so yesterday I added a 1/4 wave stub plus 1/2 wave (in 1.6mm copper wire) to the top of the J pole I had been working on. The first test was worse than the simple j pole but had I spotted @abcd567 cut off the insulation of the stub and saw an improvement. I did the same and saw an improvement too but the results are still worse than a simple J. I’m finding the copper wire elements are difficult to measure accurately and they don’t retain thier shape as well as brass. Trimming the various elements and stub may be a problem. I don’t have an answer but from the simple J tests it only takes ~0.2mm to go from good to poor. If I fail with the super J may be I will have to eat my words and try a coax colinear.
Have a good weekend all.

[triggers]

@g7pnu

I have the FlightAware antenna here at home. I did have an 18 element coax collinear but during the winds we had last winter, I lost half of it when the 22mm plastic pipe I was using as a radome snapped

I had no luck with the super J.

I have a coax collinear which has just come off test made from 8mm diameter copper tube as the outer and the internals of the coax. I stripped the outer sheath and the braid from the coax and pushed the dielectric and inner core into the copper tube. When soldered together it is quite stable. I also wanted to see if a decoupling sleeve was needed. My experiments don’t seem to show that a decoupling sleeve makes any difference.
I’ll post a photo when I get a chance.

Keep us informed of your next build.

[g7pnu]

@triggers,
I tried again with the super J. I can’t find a way to make it stable enough to trim the elements consistently, I coming to the conclusion that as home build super J may be an art form rather than a science.
I would to see how you built you stable coax colinear when you have time to take a picture.
Cheers

[belzybob]

Based purely on logic, if you found that the spider performance was equal to the FA antenna then I’d be looking closely at your main station receive path.

[biekerc]

I have made more or less made the same observations. In an indoor situation a spider can give you good results, that the FA anenna is quite sensitive to location, I also had to change gain (and added a filter, just for fun).

However, my reception is geographically limited to 150nm, so in a different location the commercial antenna probably picks up signals further away, and my impression is that although the number of planes appeared to be almost similar, the message count of the commercial antenna was and is higher.

[g7pnu]

Like you both I found that the 1/4 spider performs really well, needs little or no tuning and doesn’t take long to put together. I will resist buying an antenna as I find the fun is in the making, so I will compare my attempts to the good old abcd cantenna, rather than the FA ant. 1/4 waves can look really neat and perform well, I have a longer sma to N type cable on order, so I hope to connect up my new brass 1/4 wave when the boat from China arrives with the cable, three weeks This time I will fit a screw to the main element so that I can tweek the length up and down. The only thing I’ve found with spiders is the radials have to be more than 1/4 wave from the centre, when I cut my radials to .25 the performance fell off. I have read some rsgb and arrl books which say the length of the radial needs to be .28 from the centre rather than .25. What length of radial works for you?

[g7pnu]

@trigers is this the sort of thing you made with copper tube? (this example was 2.4Ghz in brass)

[triggers]

@g7pnu

@triggers is this the sort of thing you made with copper tube? (this example was 2.4Ghz in brass)

Yes, exactly like that.