Threaded rod parallel to tracker antenna - what happens?

So, I have read not to do this. What is the expected error or loss in performance? I have a BRB900 and Perfectflite MAWD together in an av-bay with 1/4" threaded rod about 8" long. The altimeter works fine and fires ejection charges as expected. The tracker seems to give normal GPS readings in my backyard. However, I have not yet tested the tracker at long distance/range.

What is the harm in this setup?

Buckeye said:

So, I have read not to do this. What is the expected error or loss in performance? I have a BRB900 and Perfectflite MAWD together in an av-bay with 1/4" threaded rod about 8" long. The altimeter works fine and fires ejection charges as expected. The tracker seems to give normal GPS readings in my backyard. However, I have not yet tested the tracker at long distance/range.

What is the harm in this setup?

Click to expand...

My opinion and experience, no harm. It may change the radiation pattern but all the RF gets out.....

I have yet to experience interference from metal with my BRB900. Here is an experiment you could try, although to me it is a pain in the you-know-what.

Put the threaded rod and the transmitter in a nose cone and put it in your car. Have someone at home on the phone with you, and drive until it goes out of range. Take the transmitter out of the NC and see if it comes back into range. If it does, keep driving until it goes out of range and note the distance differential. There's your loss due to interference.

Of course, if it doesn't go back in range, then you have reached the units limit, and the threaded rod made no difference.

Personally, I put my transmitter in my NCs with threaded rod and even brass tubing, and let it rip. Haven't had a problem. Maybe if my rockets were landing five miles away, but on our range, I've never had to walk more than a mile.

jderimig said:

My opinion and experience, no harm. It may change the radiation pattern but all the RF gets out.....

Click to expand...

Do you think it might matter what metal the threaded rod is? Using aluminum all thread versus steel?

markkoelsch said:

Do you think it might matter what metal the threaded rod is? Using aluminum all thread versus steel?

Click to expand...

I don't think it matters.

I think a lot of the concern has been with the Eggfinder antenna, not the BRB900.

Is it the metal or the threads? I'm assuming its the metal.

Bat-mite said:

Put the threaded rod and the transmitter in a nose cone and put it in your car. Have someone at home on the phone with you, and drive until it goes out of range. Take the transmitter out of the NC and see if it comes back into range. If it does, keep driving until it goes out of range and note the distance differential. There's your loss due to interference.

Click to expand...

Must have line of site the entire time for that test to be valid.

Exactly. Putting it inside your car is going to attennuate the signal too.

timbucktoo said:

Must have line of site the entire time for that test to be valid.

Click to expand...

OK, I just conducted a test. I hung my av-bay in a tree and took a series of GPS readings from 0 to 0.8 miles away with a decent line of site. I then repeated the measurements with the all-thread removed from the av-bay. No appreciable difference in range or accuracy. If anything, with the all-thread, the GPS measurements had less scatter!

The allthread WILL reduce the range, in certain directions that are difficult to predict. It will reflect some of the signal back upon itself causing attenuation and possible dead spots, as well as blocking the signal in the direction immediately in-line with the allthreads. That applies to any RF device, not just an Eggfinder. Now, whether or not it matters in your particular application is another matter... it would depend both on the distance at which you expect to see the rocket from, and which direction you happen to be at. I don't know if anyone has done a controlled study of the effects of allthreads on the RF signal pattern, though. Our recommendations are based on best practices... that doesn't necessarily mean that doing it other ways is not going to work.

Buckeye said:

OK, I just conducted a test. I hung my av-bay in a tree and took a series of GPS readings from 0 to 0.8 miles away with a decent line of site. I then repeated the measurements with the all-thread removed from the av-bay. No appreciable difference in range or accuracy. If anything, with the all-thread, the GPS measurements had less scatter!

Click to expand...

I'm not sure what the GPS readings are telling you. Is the GPS Rx sending a NEMA data stream to another Tx and you are receiving that data on a separate Rx? If that is the case as long as the GPS Rx has a lock on the satellite constellation you are only receiving the data that the GPS Rx is sending to the Tx. If there is a change in the GPS Rx's accuracy (less scatter) when you add or subtract the metal then that would indicate that there is an affect from the metal.

To do an accurate assessment of the affect the metal in any particular avbay configuration is beyond the scope of this discussion, or maybe not, but it is pretty involved and would require some pretty sensitive measurement equipment that most folks flying rockets probably don't have access to. You have demonstrated that adding or subtracting metal in close proximity to an RF device (your GPS Rx) changes that devices behavior. The behavior of the antenna in relation to the metal is complex but the overall changes are usually negative when compared with an omni-directional radiator.

cerving said:

The allthread WILL reduce the range, in certain directions that are difficult to predict. I don't know if anyone has done a controlled study of the effects of allthreads on the RF signal pattern, though. Our recommendations are based on best practices...

Click to expand...

AG0D said:

To do an accurate assessment of the affect the metal in any particular avbay configuration is beyond the scope of this discussion, or maybe not, but it is pretty involved and would require some pretty sensitive measurement equipment that most folks flying rockets probably don't have access to. The behavior of the antenna in relation to the metal is complex

Click to expand...

Well, this IS the scope of this discussion, so let's talk about it. I have data. Here are the results of my simple experiment. Shown below are the decimal minutes of my BRB900 measurements that correspond to the placemarks on the image. The maximum range is about 40m for scale. Yellow icons are with all-thread. Red icons are without all-thread.

With all-thread:
.7272 N, .8150 W, Range = 0
.7288 N, .8133 W, Range = 0.25 mi
.7267 N, .8195 W, Range = 0.5 mi
.7255 N, .8193 W, Range = 0.8 mi

Without all-thread
.7262 N, .8133 W, Range = 0
.7397 N, .8130 W, Range = 0.25 mi
.7427 N, .8112 W, Range = 0.25 mi
.7280 N, .8058 W, Range = 0.5 mi
.7217 N, .8154 W, Range = 0.8 mi



As I mentioned, there is less scatter with the all-thread. Anomaly? Disproves the best practices? Anyway, I don't think any of these readings would prevent me from finding my rocket, which is the bottom line.

Your result probably does not have statistical significance unless you repeat it several more times.

You may have a "lucky" incidence where your all-thread config gives better signal strength on those satellites that can give a better fix of your position. Or your all-thread may have censored otherwise strong satellites and forced the ublox to select other weaker satellites that gave a better fix (closer to the horizon for instance).


The best way to verify this is to use a tool like ublox u-center which will give you signal s/n for the satellites in the constellation. I don't know if the BRB900 outputs the sentence for this measurement or not.

By the way, your range has no effect on the GPS fix standard deviation so you can repeat this experiment without the distance factor. Easy to run.

Edit: I did more analysis. The latitude fix variation was the biggest difference between your runs. The longitude variation was about the same. This suggests there was a constellation difference between the 2 test runs, might have nothing to do with the all-thread or may have.

But your bottom line is right, the all-thread wont impede you finding your rocket.

Buckeye said:

Well, this IS the scope of this discussion, so let's talk about it. I have data. Here are the results of my simple experiment. Shown below are the decimal minutes of my BRB900 measurements that correspond to the placemarks on the image. The maximum range is about 40m for scale. Yellow icons are with all-thread. Red icons are without all-thread.

With all-thread:
.7272 N, .8150 W, Range = 0
.7288 N, .8133 W, Range = 0.25 mi
.7267 N, .8195 W, Range = 0.5 mi
.7255 N, .8193 W, Range = 0.8 mi

Without all-thread
.7262 N, .8133 W, Range = 0
.7397 N, .8130 W, Range = 0.25 mi
.7427 N, .8112 W, Range = 0.25 mi
.7280 N, .8058 W, Range = 0.5 mi
.7217 N, .8154 W, Range = 0.8 mi

View attachment 278418

As I mentioned, there is less scatter with the all-thread. Anomaly? Disproves the best practices? Anyway, I don't think any of these readings would prevent me from finding my rocket, which is the bottom line.

Click to expand...

The GPS readings aren't the issue, it's the RF output from the transmitter. The only way to properly test this would be to put it in a proper RF-anechoic testing chamber and measure the RF output at a large number of points at various distances around the device with a properly calibrated and very expensive meter. This is how antennas are tested. Maybe somebody out there has access to this equipment and would be willing to perform a little experiment for the benefit of the community, but unfortunately Eggtimer Rocketry doesn't have the resources to perform such testing (it's VERY expensive) and I doubt that any of the other rocketry electronics vendors do either. So for now, the best thing we can do is to state best practices... keep metal away from the antennas whenever possible.

jderimig said:

Your result probably does not have statistical significance unless you repeat it several more times.

You may have a "lucky" incidence where your all-thread config gives better signal strength on those satellites that can give a better fix of your position. The best way to verify this is to use a tool like ublox u-center which will give you signal s/n for the satellites in the constellation. I don't know if the BRB900 outputs the sentence for this measurement or not.

By the way, your range has no effect on the GPS fix standard deviation so you can repeat this experiment without the distance factor. Easy to run.

Click to expand...

Yeah, the range was noted just to ensure that both configurations had the same range capability, at least to 0.8 mi.

Surely, this is not rigorous data collection, but some typical field use by a guy walking in the local park on his day off! I wanted to put a few numbers behind the anecdotes mentioned in previous posts.

These are the sentences offered by my BRB900:

Actually I do (at my day job, we need to pass FCC and global EMI standards). The best practice is to keep metal away from an antenna if you do not know what you are doing.

Almost every commercial antenna (cell phones, commercial radio, broadcasters etc) have their antenna's mounted to metallic structures. Even our handheld Arrow yagis have their resonant elements attached to metal booms, electrically! In most cases there is little practical effect on the usability of the system which is what Buckeye and many other rocket people have discovered.

One notable exception case is if a nearby metal object is resonant or very close to resonant at the transmitting frequency. Then this object will couple with the device antenna and pull the antenna load impedance off the design value and cause some SWR losses there. But that assumes that the device's antenna is correctly impedance matched in the first place. Most are not, but it doesn't practically matter, they do the job. Also, all this antenna theory goes out the window when the device is on the ground right? When the antenna is in the air signal reception is generally not a problem with even poor antennas.

So will nearby metal affect our RF output in some way? Absolutely! But if the effect is hard or impossible to measure in the use application does it practically affect the RF output?

cerving said:

The GPS readings aren't the issue, it's the RF output from the transmitter. The only way to properly test this would be to put it in a proper RF-anechoic testing chamber and measure the RF output at a large number of points at various distances around the device with a properly calibrated and very expensive meter. This is how antennas are tested. Maybe somebody out there has access to this equipment and would be willing to perform a little experiment for the benefit of the community, but unfortunately Eggtimer Rocketry doesn't have the resources to perform such testing (it's VERY expensive) and I doubt that any of the other rocketry electronics vendors do either. So for now, the best thing we can do is to state best practices... keep metal away from the antennas whenever possible.

Click to expand...

I find your dithering results interesting, however as mentioned by John and others in the collective opinions upstream, there's way more unaccounted for in your tests. Still, it's interesting.

Getting back to the active radio question/discussion, the XBee 900MHz radios do extremely well when at altitude. I can say that in all our RTx system field testing there's not been anything notable from flights or other tests with or without allthread in proximity. BRB, Marsa, Missile Works, and RocketTrack users will all be using this platform for telemetry. The following info is a partial hijack, but at the same time pertinent to this discussion.

I've confirmed with Greg Clark that all BRB900 systems are using the XBee legacy XSC firmware, thus my upcoming RTx system and BRB900 systems won't clash at all (other than the rare occasion when hopped packets are on simultaneous frequencies, and even then the radios account nicely for this rarity).

I've spent a lot of time pondering the underlying methods of supporting and maintaining the MACHY addressing for data "packetizing" used by the RTx system, and because the release of this system is very close I felt this thread would be a good venue point out the domain of this addressing where my radio system configurations will get deployed. I'm putting this out with the intent that it be known by all XBee system suppliers/users and independent RocketTrack builders/users.

The RTx system will employ the "non-meshing" MACHY addressing scheme of the XBee Pro 900HP radio modem with the -DP (point-to-point) firmware.
My plans are to deploy radios as follows:

Network Preamble: 5
Network ID's: 0 thru 32767

RTx radios will also be using the unique 64-bit DH/DL addressing of each radio module as yet another layer of guaranteeing addressing uniqueness and maintaining interference free data. I'll be setting up each radio system with a "many-to-one" capability, thus if additional "rocket radios" are desired, they can all be addressed to a unique base radio ID. Accessing the pertinent network ID is supported thru the LCD interface and can be retrieved if needed for system replacements or expansions.