Some thoughts on carbon fibre and trackers

[stealth6]

This post will likely illustrate how much I DON'T know about this subject. But I just had a thought and figured I'd put it out there for those in the know to either ponder, ridicule, or perhaps (though unlikely) hail me as a genius for thinking of this.

Ok, so we all know that enclosing trackers in carbon airframes or nosecones is a bad idea, and that that RF signal will be effectively blocked when doing so. But what if you had a metal tip on that cone? Could you tap the tracker antenna into that and would that work? Would that allow the signal to get through and act as an effective antenna?

Or perhaps better still - what about embedding a copper wire into the outside of the airframe or nose cone? This could easily be done so that it was completely flush, and the outer surface of the rocket could be perfectly smooth (just cut a channel, set the wire into it, epoxy it into place, and sand flush). Again, you could tap the tracker antenna into that so that your antenna would effectively be "outside" the carbon fibre. You could also perhaps do this with a few wires rather than just one so that they would wrap around the whole rocket or cone.

I was thinking of this because I've used a technique in the past for woodworking whereby I inlay wire into the surface of a piece. It's a neat effect. This would be using the same basic technique but for an entirely different purpose.

Stupid idea? Absolutely brilliant? Already been done?

discuss...


s6

 

[CarVac]

If the wire is parallel to the carbon surface, as it would be if it's embedded in the surface, then most of the radiated energy will get absorbed into the conductive carbon. If perpendicular, then voila, you have a really great antenna, but a rocket with a wire sticking perpendicularly out the side.

The metal tip would be hard to turn into a good antenna, I think.

 

[Peter Olivola]

Does the tracker have a built in antanna? Can the antenna be disconnected and another antenna attached? Do you know the tracker's operating frequency and how to calculate the proper antenna dimensions?

 

[troj]

An aluminum tip could be used as an antenna, but it's going to be a poor one, for any wavelengths we use -- it's the wrong size.

As far as bonding it to the airframe goes, that'll work fine....as long as the side with the antenna is oriented towards what you want to receive the signal. Think of it like trying to see a lantern next to a Giant Sequoia, but you can only see the light directly from the lantern, no reflections. If you're on the side opposite the lantern, you won't see it at all, because the tree blocks it. The only way you'll really get a good view is when it's oriented such that the lantern is directly between you and the tree.

-Kevin

 

[Benbug]

So glad to hear of someone else thinking along these lines!

I'm currently partway through a 54mm min. diameter build and have been experimenting with embedding ceramic coated 0.6mm wire into the surface of carbon tube.

I decided to try several variations in antenna design and build them all into a test rig (all 70cm band for use with the telemetrum) : a regular 17.5 mm straight antenna plus a variation @ 35mm, another two variants of 17mm and 35mm set in a subtle helix around the airframe and a 35mm dipole. I've only got as far as constructing the test rig and won't get to try it out until I can get some time off work - I may be able to cram a test into the NZ long weekend coming up.

After doing a heap of research into the "line of sight" issue with RF , I've found it isn't as straight forward as one might think. Radio has all kinds of anomalies and quirks that don't quite fit the analogy to visible light. My brother works for a top Silicon Valley company and his radio specialist colleagues had the same thoughts I did in regards to embedding the antenna flush to the airframe - "There's only one way to find out..." Hence why I've built the test rig.

The boffins did make one suggestion - the trench in the surface of the carbon should have a "u" shaped cross section that at the surface of the airframe, is at least 3 times the width of the antenna wire (in my case, the trench is 1.8 mm wide and 0.6 mm deep). With the ceramic insulated wire centered at the lowest point in the "U" the trench works as a parabola. They did agree that the carbon would still soak up some of the signal, but not to a huge degree. I suggested using heat fixed micro-foiling on the surface of the trench as a reflector and insulator - they thought that should do the trick, so long as the foil didn't prevent the epoxy used to flush the trench from adhering to the carbon.

Stealth6, it sounds like this idea has been knocking around in your head - much like it was in mine. I was finally inspired to go ahead and try it after seeing the success Bdale Garbee had with his dipole fin design - https://www.gag.com/rockets/airframes/YikStik3/index.html

Although the layout is different, the two concepts are still in the same family. I'll post some results from my test rig as soon as I have them.

Bb

 

[dixontj93060]

I had a problem at Midwest Power back in November finding out that my Formula 98 nosecone apparently does have some CF in it as it totally killed my tracker signal. Simple solution, I turned the carrier board for my BRB around and got out an 1/8th drill and put a small hole in the nosecone bulkplate so the antenna could stick out of the bulkhead--problem solved; worked perfectly.

 

[CCotner]

...finding out that my Formula 98 nosecone apparently does have some CF in it as it totally killed my tracker signal...

Let me guess, it was one of those shiny profusion fiberglass ones, right? Graphite-laden.

 

[JordanT]

Let me guess, it was one of those shiny profusion fiberglass ones, right? Graphite-laden.

Wait a minute...I though graphite was okay, but CF was okay (particle vs strand). I thought the Blackhawks have graphite nose cones to allow a tracker signal.

If the signals are effectively blocked by the CF, could you simply embed multiple antennas at 180 or 120 degrees?

 

[jason-and-his-toys]

I'm using the Comspec transmitters and I just attach them to the Kevlar recovery cord. I put one on the main line and one on the drogue line. They are on different frequencies so when the drogue comes out you get the strong signal on that transmitter and still nothing on the main one. When you get both signals at apogee it becomes an "oh crap" moment when you now know you're going for a long hike to get your bird back. Yup been there done that.... On my TRA L3 / CAR L4 certification flight! :blush:


If it augers in for some reason it isn't going to matter if you had a signal or not on the way up. Nothing will be working at the end of the "flight".

I guess my point is, as long as you design your tracking to get "in clean air" after your drogue deploys your good. Now if your wanting GPS data on the way up you may have a problem.

Jason

 

[cwbullet]

Like Jason and many others on this board, I attach my tracker to the shock cord with tape - electrical tape. I have never had one come off.

 

[bobkrech]

Wait a minute...I though graphite was okay, but CF was okay (particle vs strand). I thought the Blackhawks have graphite nose cones to allow a tracker signal.

If the signals are effectively blocked by the CF, could you simply embed multiple antennas at 180 or 120 degrees?

Graphite is a conductor. Carbon fiber is a conductor. Conductors block RF.

Bob

 

[GDJ]

Graphite is a conductor. Carbon fiber is a conductor. Conductors block RF.

Bob

In that case,if the nose cone is carbon fibre, how about making the nose cone into the antenna?

 

[CarVac]

I'm pretty sure you need a good conductor to make an antenna, while you only need a bad conductor to ruin one.

The world likes to make things difficult like that.

 

[GDJ]

Hmm....theoretically if the copper wire "antenna" were to be spaced away from the carbon fibre nosecone (ie laid on top of the epoxy layer)wouldn't the nosecone become a reflector or a driven element?

 

[G_T]

If you want the antenna to do a good job radiating the signal, then it should be resonant at the desired frequency. Otherwise much of the energy gets reflected back to the transmission circuitry and can potentially degrade/damage/destroy the final amplifier circuit depending on how it was designed.

Things such as a parabolic shaped trench for the wire to act as a reflector are the product of faulty reasoning. The dimensions of the trench would have to be on the order of multiple wavelengths otherwise one is looking at near field phenomenon rather than far field. The parabolic tranch is an attempt at a far field solution. In the near field, the trench will couple capacitatively and inductively with the "antenna" and cause severe detuning. It may end up being a good matching element to permit the nosecone to absorb the RF and convert it to heat - not exactly what is desired.

An antenna in the neighborhood of things which conduct suffers several effects: (1) detuning or altering of resonant frequency(ies) (2) frequency dependent alteration of radiation pattern and polarization (3) frequency dependent alteration of standing wave ratio (4) frequency dependent alteration of antenna efficiency.

Without extensive modeling or extensive testing, it is best to use the stock antenna or a manufacturer recommended antenna substitute, positioned as far away from conductors (even poor conductors such as a graphite infused nosecone) as possible, and oriented as far away from parallel to any conductors as possible.

Gerald

 

[bobkrech]

In that case,if the nose cone is carbon fibre, how about making the nose cone into the antenna?

If you could get it to resonate at the transmitter frequency it could work however I think it is beyond the average rocket hobbyist who lack the expertize in MW antenna design.

Two (and only two) patch antennas on opposite sides of an airframe will give omnidirectional coverage. The pros do it with flex circuit type patch antennas but again the problem is that you may not find a COTS solution and most rocket hobbyists don't have the expertize to design one or willing to spend the $$$ to pay for a fabrication run.

Bob

 

[MCS]

I haven't actually used a patch antennae in a rocket, but I love the concept. In theory it should work great.

If I were building a patch antenae I would start by copying the work done by PSAS. The antennae is the green and gray cylinder in the middle
photograph.
https://psas.pdx.edu/communications

Lots of good documentation and notes on their website including spread sheets.

-->MCS


.

 

[bobkrech]

I haven't actually used a patch antennae in a rocket, but I love the concept. In theory it should work great.

If I were building a patch antenae I would start by copying the work done by PSAS. The antennae is the green and gray cylinder in the middle
photograph.
https://psas.pdx.edu/communications

Lots of good documentation and notes on their website including spread sheets.

-->MCS


.

Good reference. GoFast used 2 patch antennas 180 degrees apart giving spherical coverage which insures uninterruipted hemisherical coverage of the GPS signal. More is not better. They first tried 4 at 90 degrees but that doesn't work i'm guessing do to a phase interference effect.

One of my coworkers is an expert in the field and holds several patents in microwave antenna design and she's my goto person when I need antennna info.

Bob

 

[UhClem]

I haven't actually used a patch antennae in a rocket, but I love the concept. In theory it should work great.

If I were building a patch antenae I would start by copying the work done by PSAS. The antennae is the green and gray cylinder in the middle photograph.
https://psas.pdx.edu/communications

The PSAS antenna has the oddest feed system I have ever seen. To understand why it is so odd, you need to know a few things about patch antennas.

The first is that it is a resonant system and the patch will resonate at a frequency determined by the length of the antenna and the material properties of the substrate and surrounding materials. Rogers Duroid which is frequently used has a dielectric constant of 2.2 so the length of the antenna will be lambda/2 * 1/2.2.

The second is the the impedance at the feed point is very dependent on the design. Wider antennas have lower impedance. Normally a square patch will have an impedance well above the usual 50 Ohms expected by transmitters.

(The PSAS antennas are so wide that their impedance is less than 50 Ohms.)

There are several ways to match the impedance:

1) You could include impedance matching structures in the microstrip feeds.
2) The impedance at the feed point can be reduced by offsetting the feedpoint from the edge. The closer to the center the lower the impedance.
3) You can use an array of patches.


The PSAS antenna has a corporate feed with multiple feeds to a single wide patch. I have never seen this before as the usual method is to feed multiple patches. I suspect that it isn't very efficient as a result of the impedance matching requirements in the feed.

Design and fabrication of patch antennas is pretty straightforward. The tricky part is bonding a protective radome to the antenna.

 

[GDJ]

If you could get it to resonate at the transmitter frequency it could work however I think it is beyond the average rocket hobbyist who lack the expertize in MW antenna design.

Two (and only two) patch antennas on opposite sides of an airframe will give omnidirectional coverage. The pros do it with flex circuit type patch antennas but again the problem is that you may not find a COTS solution and most rocket hobbyists don't have the expertize to design one or willing to spend the $$$ to pay for a fabrication run.

Bob

So you are referring to a simple dipole antenna?

 

[bobkrech]

In discussing this thread with a EE coworker who is also a ham, he suggested a good home brew way to determine whether your NC is microwave transparent is to put the NC (less any other parts) into your microwave oven and turn the oven on for a few minutes. If it gets hot, it absorbs microwaves (especially in the 2.3 Ghz WiFi band since microwave ovens operate at 2.45 GHz.) and therefore is not microwave transparent.

Caution: You might want to do this test in stages and watch carefully. If sparks occur, it is conductive and not microwave transparent. If you don't see sparks, the NC may still be microwave absorbing and if you get the NC too hot, you could melt the resin or otherwise damage a perfectly good NC.

Bob

 

[CarVac]

In discussing this thread with a EE coworker who is also a ham, he suggested a good home brew way to determine whether your NC is microwave transparent is to put the NC (less any other parts) into your microwave oven and turn the oven on for a few minutes. If it gets hot, it absorbs microwaves (especially in the 2.3 Ghz WiFi band since microwave ovens operate at 2.45 GHz.) and therefore is not microwave transparent.

Caution: You might want to do this test in stages and watch carefully. If sparks occur, it is conductive and not microwave transparent. If you don't see sparks, the NC may still be microwave absorbing and if you get the NC too hot, you could melt the resin or otherwise damage a perfectly good NC.

Bob

A few minutes?

If the nosecone is indeed microwave transparent, you'd be running the oven unloaded for a few minutes, which is apparently a Bad Thing.

 

[bobkrech]

A few minutes?

If the nosecone is indeed microwave transparent, you'd be running the oven unloaded for a few minutes, which is apparently a Bad Thing.

Really????

 

[GDJ]

Perhaps a small mug of water should be added?

 

[CarVac]

Really????

Yeah, my microwave has a sticker admonishing against it; apparently it can cause resonances and localized overheating in the chamber.