Tuning a Telemetry Antenna with a VNA

[OverTheTop]

Tuning a Telemetry Antenna with a VNA

As part of my Vertical Trajectory System I am adding another telemetry downlink from the rocket, and for that I need an antenna. I want something robust and simple so I chose to just make a simple dipole that can be taped to the outside, or inside, of the rocket airframe. This was accomplished by just using a SMA pigtail and stripping the coax. This however had to be tuned to the correct frequency.

I grabbed my Nano VNA, and quickly calibrated it for a frequency range of 1-3GHz, to suit my target frequency of 1.28GHz.

Connecting the commercial antenna to it quickly confirmed the center frequency of that antenna at 1.24GHz.




Next up I grabbed my roughly-trimmed dipole and connected it, revealing a frequency of 1.1GHz in air (look for the resonance dip in the yellow trace. That is where more energy is going out into the air).




I then taped the antenna to the airframe, simulating its final mounting to some degree. The frequency was shifted to below 1GHz due to the fiberglass dielectric in proximity to the dipole.




Based on the reduction of frequency due to the fiberglass and the low frequency from the first attempt I calculated the length required and cut the dipole. The result is an antenna on the 1.28GHz I wanted.




I took the antenna and taped it to the stick of a Q-tip for now. It will be quite adequate for bench testing.




I really can’t wax lyrical enough about the VNA. Such a handy tool to have around, and the whole process of tuning the antenna took around 15 minutes. Totally amazing!

 

[Hobie1dog]

You are amazing

 

[cerving]

I've been looking at getting of one those, it's a amazing tool for the price ($129). That and a RF Explorer pretty takes care of just about anything you need to do radio-wise.

 

[OverTheTop]

I've been looking at getting of one those, it's a amazing tool for the price ($129). That and a RF Explorer pretty takes care of just about anything you need to do radio-wise.

It really is worth the money. Don't hesitate to push the button. Very easy to use, but you might need to purchase some more adapters. $129USD sounds a bit expensive, unless you are talking the larger screen. Make sure you get the V2 that has native capabilities to 3GHz IIRC, rather than the earlier version that just uses harmonics to get up that high.

 

[Voyager1]

Thanks for presenting this information OTT. This is really interesting stuff and close to one of my current design projects - albeit, a little inactive at present due to other commitments. I totally agree with the usefulness of the Nano VNA V2. It's great! Also, as Cris mentioned, the RF Explorer is a great little spec analyser, as we use one for confirming the frequency and power output from our member's telemetry Txs.

In your case, where you form the dipole elements from the stripped coax, although the coax (looks like RG316 or similar) is, I assume 50 Ohm, the dipole feed point will be around 70+ Ohms when the dipole is straight. However, I believe that as you bend the dipole elements towards each other, the characteristic impedance reduces as the bend increases. This is useful because it might bring the feed impedance closer to the desired 50 Ohms of the coax and the VNA, but the penalty might be in a reduction in overall gain. Additionally, there might also be an increase in the SWR either side of that balance point. As for the far field radiation pattern, there is probably no significant change due to the bending of the dipole in that orientation. Curving towards or away from a ground plane might be a different case, though.

It would be interesting to compare this configuration with the dipole along the main axis of the body. In this way you might be able to determine the effects of the dipole bending vs the dielectric on the characteristics. Also, what are the dipole characteristics when curved in free space vs being curved around the dielectric body cylinder.

As an afterthought, would you consider using two dipoles with cross polarisation? One one each side of the body. They could be coupled through a simple 2-1 power divider/combiner.

Keep up the good work.

 

[OverTheTop]

Good stuff Voyager. Yes the impedance does drop as a make a V with the dipole. That was likely the configuration I was going to run with in the end. What I have is good enough for bench testing for now. I seem to remember having the antenna in a V formation also means less change in signal level due to wrong polarisation matching with the receiver. Not sure the mechanism but I seem to recall reading it is so.

I am considering going circular polarisation but keeping it simple initially.

At the moment I have a lend of a spectrum analyser from work so I don't need to buy one just yet. There is a superfluity of them there currently so I am permitted to borrow one.

 

[Voyager1]

At the moment I have a lend of a spectrum analyser from work so I don't need to buy one just yet. There is a superfluity of them there currently so I am permitted to borrow one.

Lucky lad! Having retired from the profession (if you ever do!) a couple of years ago, I no longer have access to the R&Ss and Agilents, etc. I just have to make the best of these cheap units, but they are very good value for money.

 

[OverTheTop]

Work also have three Agilent VNAs, ranging from a cheap $10k FieldFox, and upwards. The NanoVNA suits my needs well, so I don't need to borrow one of those! Takes up less bench space too .

I had been looking at one of those RF Explorer units but have not bitten yet.

 

[cerving]

It really is worth the money. Don't hesitate to push the button. Very easy to use, but you might need to purchase some more adapters. $129USD sounds a bit expensive, unless you are talking the larger screen. Make sure you get the V2 that has native capabilities to 3GHz IIRC, rather than the earlier version that just uses harmonics to get up that high.

I just pulled the trigger on one of the new 4" ones, got it for $90. Should be fun to play with!

 

[cerving]

Work also have three Agilent VNAs, ranging from a cheap $10k FieldFox, and upwards. The NanoVNA suits my needs well, so I don't need to borrow one of those! Takes up less bench space too .

I had been looking at one of those RF Explorer units but have not bitten yet.

When you get an RF Explorer, spend the extra money and make sure that it handles whatever bands you may be dealing with. I got one for the 900 MHz band, it turns out that it only handles 910-920 MHz so it wouldn't even handle the full band that we use for Eggfinders; I had to purchase an add-on module to extend it.

 

[mbeels]

I really can’t wax lyrical enough about the VNA. Such a handy tool to have around, and the whole process of tuning the antenna took around 15 minutes. Totally amazing!

Wow, very nice work. I absolutely can't believe that it is possible to buy a functional VNA for $129.

 

[OverTheTop]

I paid about $110AUD for mine IIRC. I think that was with cal kit as well.

Don't forget to buy some connector savers as shown in my original thread too.