Category: Technical

VA3SIE Bike Mobile Station

By , July 16, 2011 5:58 pm

I had to rebuilt my bicycle mobile antenna yesterday, the old one failed.  This was not unexpected I had made an error while building it and I’m surprised it lasted as long as it did (3 years).  So I thought this is a good opportunity to build it better and share the details.

The Design

The DBJ-2 (construction article here) is a roll-up dual-band J-pole.  It offers significant gain (7dB – 10dB+) compared to the stock rubber ducky antenna.  It is better than a ground plane for a bicycle because it does no need a counterpoise and the high current sections are higher up off the ground so there is less ground loss.  Also because it is a base loaded long antenna it is easy to mount on a bicycle.

Construction Plan

Construction Plan

This design is interesting!  While a standard 2m J-pole does resonate on 70cm, the radation pattern is not ideal.  Most of the energy is directed out at more than 45° elevation resulting in a 4dB to 6dB loss compared to a groundplane. 

This antenna design incorporates the following elements.  At the base is a (shorted) 15″ matching stub, tapped for 50Ω.  This stub is λ/4 on 2m and 3λ/4 on 70cm.  At the top of this matching stub is the first ¼″ notch.  This notch stops isolates the current on the other side of the twinlead, ensuring that the bottom 15″ does not radiate (since the RF current flows on both sides of the twinlead in this section).

Above that is a λ/2 UHF radiator.  Above that is a λ/4 shorted stub constructed from RG-174 coax.  This coax stub is open circuit at UHF.  At the top end of the coax stub is an additional 18″ of 300Ω twinlead which completes the λ/2 VHF radiating element (which includes the UHF radiating element, the coax stub and the remainder of the twin lead).

The coax stub is slightly inductive at VHF and the twinlead has a velocity factor, so the lengths shown above are all shorter than the value you might expect.

Construction Details

I constructed this antenna using TV twinlead from Princess Auto

Stage One

I started by completely removing the twinlead plastic from 1″ to 1 ½″ from the bottom of the coax and twisting together the wires at the bottom end of the coax.  This gives me a ½″ section to slide the feed wires along to find a good 50Ω match. 

Next, I removed the three ¼″ notches on the ground side of the twinlead.  But rather than snipping out both the wire and the plastic I snipped the plastic with a cutter at either end of the ¼″ notch and then slit it down the side with a hobby knife, pried the plastic open with a screw driver and pulled out the wire with needle-nodes pliers.  This way, when I tape up the plastic afterwards it is still very structurally strong.

 

Stage 1 Build (stripped matching section, twisted together base wires, cut notches

Stage 1 Build (stripped matching section, twisted together base wires, cut notches

 Stage Two

After that, I prepared the coax stub.  I clipped a 4¾″ section of RG-174 and I stripped the outer cover from ¼″ at each end to expose the shield.  At the top end, I pulled back the shield strands and straightened them out, and I removed the inner dielectric, then I pulled the shield strands back out to mingle with the inner strands.  Thus, the stub is shorted at the top end.  At the bottom end I pulled the shield strands out and then clipped them off, exposing the inner dielectric.  I removed ½ of the dielectricm exposing some of the coax inner wire.   I realized that during soldering ½ again of the exposed dielectric would melt back.  This completes the 4¼″ stub preparation.

I also clipped out two ¼″ notches and this time I removed the plastic, to expose two ¼″ wires to connect the stub between.

Finally, I soldered the stub into the twinlead and taped the ends securely.  In order to ensure a good connection, I did not twist any of the wires of the RG-174 inner/outer or the wires in the twinlead.  I mingled all the wires together and then soaked in some solder at 750º.  Capillary action created some nice strong sections without too much solder.  I then taped the notched areas securely.

Shorted the top stub end and exposed the inner on the bottom, notched twinlead then soldered stub in

Shorted the top stub end and exposed the inner on the bottom, notched twinlead then soldered stub in

 Stage Three

Finally, I prepared the RG-58 feed coax by exposing ¾″ of inner and outer separated by ⅛″ of dielectric and I trimmed some of the outer wires so that the outer and inner were roughly the same diameter, easier to work with.  Then I taped up all the remaining notches and weak points, and I connected the RG-58 feed point to the ½″ matching area, ready to perform the final testing & trimming.

Prepared RG-58 coax, taped up points and attached feed point

Prepared RG-58 coax, taped up points and attached feed point

Installation

I connected the antenna to my bicycle.  It is supported at the back of the bicycle by two fiberglass sections of a Shakespeare Wonderpole TPS-20 fishing pole to which it is simply taped.  There is a coax balun in one of the bicycle panniers and then the coax runs to the front of the bicycle where it connects to my VX-8r H/T which is attached to the handlebars with bungee cords.  The microphone (and integrated GPS antenna) is also connected to the bicycle frame with a bungee cord.

Radio, Balun & Antenna installed on Bicycle

Radio, Balun & Antenna installed on Bicycle

Trimming

I used an MFJ-259B antenna analyzer (on loan from Anthony at the QRP Club – Thanks, Anthony!!) to analyze the antenna.  It was resonant at 136MHz.  I guess this is due to the velocity factor of the twin lead.  Moving the feed point across the ½″ sliding area produced no change in the resonant point but did change the SWR.  The SWR was better (around 1.2:1) at the top of the sliding range.  So I proceeded to trim off one inch at a time from the top of the antenna to bring it into the 2m band.  I ended up having to remove 7″ to get the resonant frequency to 147MHz.  With the feed point at the top of the sliding range, the match was 1.1:1.

Since I removed this length from the area above the UHF trap, I expect that the antenna is not a good match on UHF.  I estimate that it will be around 410MHz.  I don’t have a UHF analyzer to be able confirm this guess.  Next time I build this antenna I will scale the entire antenna to be 15% shorter in all the twinlead sections, assuming that I use the Princess Auto twinlead again.

Testing

Finally, I took it for a test drive on my bicycle!  I use a bluetooth headset to connect to the radio wirelessly, this offers superb freedom while cycling.  I compared the APRS map from this cycle trip with an identical trip with my old J-pole.  The results were encouraging.  Excluding some minor variations, the same digipeaters heard me as well as they used to.  Also I was able to hear some fairly distant repeaters and to have conversations on some close ones so the antenna was working and was working well!!

Microphone, Radio, APRS Map - Testing the Antenna

Microphone, Radio, APRS Map - Testing the Antenna

As an aside, as I approached the Rockliffe Airport, I monitored airband and heard 3 aircraft arranging themselves into a landing pattern for runway 09.  The end of runway 09 is right on the cycle path, so I stopped my bike there and continued to monitor.  A few minutes later and three aircraft all landed in succession right over my head.  I could hear the propeller blades feathering in the wind.  Wow!!  What fun! :mrgreen:

So what about the hum ?

By , December 1, 2009 10:49 pm

The spectrograms in this post can be clicked to listen to them, click the little white ‘X’ at the right and below the player to close the player once you have finished listening to the sample

Tonight I participated in the ‘News and History‘ net on VE3TWO, a local VHF 2m amateur repeater in Ottawa.  Patrick, VA3CMD, was controlling the net and he commented that he was hearing a hum on the lower power stations.  I also heard the same thing.  We had two high power stations (Patrick, VA3CMD and Larry VE3WEH), and three lower power stations (Martin, VA3SIE, David VE3ZZU and Jamie VA3JME).

The interesting thing is that we could not detect any hum on the strong stations but it was noticeably present when the lower power stations were transmitting.

So lets take a look at the spectrogram using Audacity!

Here’s the spectrogram for a strong station (VA3CMD).  Notice that there is no hum present in the gaps in Patricks speech:

Notice that there's no hum present in the gap.

(Click spectrogram to listen) Notice that there's no hum present in the gap.

Now lets take a look at the spectrogram for a weaker station, in this case Jamie VA3JME:

Lower power station has a hum present at 700Hz->900Hz.

(Click spectrogram to listen) Lower power station has a hum present at 700Hz->900Hz.

We can see a hum present at roughly 900Hz, and it’s even present during the gaps in speech.

So I isolated the hum sound by copying/pasting a sample from a gap in Jamie’s speech a few times to create a longer sample and then I applied a band-pass filter with a fairly sharp roll-off from 700Hz to 1200Hz.

So now we have the hum isolated:

Applied 700Hz -> 1200Hz Bandpass filter

(Click spectrogram to listen) Applied 700Hz -> 1200Hz Bandpass filter

Finally, just to prove that the hum is being introduced by the repeater, I switched over to the repeater input during Jamie’s transmission, and from the spectrogram we can see that the hum is present on the repeater output but not on Jamie’s signal on the repeater input frequency:

The hum ps present on the input but not on the output.

(Click spectrogram to listen) The hum ps present on the input but not on the output.

Wonder what causes it?

Panorama Theme by Themocracy