radar reflector

[LittleSister]

The other issue rarely discussed is that the tests are carried out on radar reflectors mounted alone in a test chamber. In practice you have to mount them somehwere on the boat. The standard mounting on front of a mast will be invisible, at least in theory, to a ship approaching from behind. (The Echomax website has some theoretical parameters.)

Following on from this I wonder whether the Tri-ball type (fresnel lens?) reflectors which mount on top of the mast (and which had a slightly lower performance than the Echomax type in the Ouzo lab report, IIRC) would not in practice give a better all round perfomance in practice.

Regarding the potential for two (or more) reflectors cancelling one another out, I would guess that this is unlikely within the length of a yacht, but I would be interested to hear from someone who knows about the wavelength, etc., of radar commenting on the plausibility of this.

[timbartlett]

I appreciate the possibility of the returns from two reflectors cancelling each other out.

Would it be any good fitting two reflectors vertically one exactly above the other so that they were always the same distance from the radar source but doubling the effective area?

If you had two identical reflectors at exactly the same distance from the transmitter as each other, then they would indeed present double the RCS of a single reflector. But if you moved one of them a quarter of a wavelength further away from the transmitter, then the signal would have to travel a quarter of a wavelength further out, and a quarter of a wavelength further back, so it would travel half a wavelength further -- i.e. it would be exactly out of phase with the echo received from the other reflector.

Given that the wavelength of X-band radar is just a tad over 3cm, a quarter of a wavelength is about 8mm. According to my mental arithmetic, if the centres of the two reflectors were 48cm apart (vertically), one degreee of heel or pitch would be enough to cause 8mm of displacement (horizontally). Another one degree would cause an extra 8mm of displacement, so the signals from the two would be back in phase and adding together again.

...Regarding the potential for two (or more) reflectors cancelling one another out, I would guess that this is unlikely within the length of a yacht, but I would be interested to hear from someone who knows about the wavelength, etc., of radar commenting on the plausibility of this.

Afraid it's not "unlikely" at all: it's inevitable!
Any practical yacht is made up of countless radar reflectors, both inside and outside the hull -- anything that conducts electricity will do. Even the crew will have an effect (I've had my own personal RCS measured at Funtingdon -- IIRC I achieved a peak of just over 2sqm! I've still got the polar diagram somewhere!

As the yacht pitches, yaws, and rolls, and as it rises and falls relative to the surrounding surface, the relative position of all those things change. As I've shown above, just a few millimetres is enough to make a difference. So the boat's RCS is constantly changing -- a fact that has been borne out by practical tests. I did a lot of work on this with Funtingdon for PBO in 1999. I would like to think that the articles are still available somewhere, but unfortunately I do not have a copy.

[VicS]

The other issue rarely discussed is that the tests are carried out on radar reflectors mounted alone in a test chamber.

Essential in order to eliminate variables over which you would have no control if you attempted to test them on a boat.

[blackbeard]

......SOLAS Regulation 19 is one of the (relatively few) SOLAS requirements that applies to yachts.
19.2.1.7 requires "if less than 150 gross tonnage and if practicable, a radar reflector or other means, to enable detection by ships navigating by radar at both 9 and 3 GHz. According to MCA guidelines, any vessel over 15m in length should be able to fit a reflector that complies with ISO 8729:1997. According to them, smaller vessels should fit the largest reflector possible That "guideline", coupled with the fact that the MCA was the driving force behind 19.2.1.7 in the first place, is why it took so long for the MAIB to publish the radar reflector section of the Ouzo report: they were trying very hard to come up with a form of words that did not highlight the fact that the MCA had just made it complulsory for yachts to fit something that (a) did not exist and (b) was about as much practical use as a lucky rabbit's foot!

.......

Thanks for this.
Does, of course, raise the question of what we should actually do. A bit of a let-out in "if practicable" especially if the kit does not actually exist.
Is it reasonable to assume that fitting a good big passive reflector as high up as practicable shows a willingness to comply with the spirit, if not the letter, of SOLAS?

I appreciate that the ideal would be an active reflector, but the OP can't afford this (nor can I)! They have another problem too; preferred location is above mast head, but using the recommended bracket they blank out an unacceptable arc of all-round-white and tricolour masthead lights. Alternative is to mount on a pole at pushpit but it's difficult, on a small boat, to reach recommended height; although 3 metres or a bit more should be feasible. According to Echomax, they should NOT be mounted in front of mast or on spreaders, even though these seem to be popular locations for passive reflectors.

[Twister_Ken]

I appreciate the possibility of the returns from two reflectors cancelling each other out.

Would it be any good fitting two reflectors vertically one exactly above the other so that they were always the same distance from the radar source but doubling the effective area?

For the price of doing that, why not fit an active transponder?

[RAI]

If you had two identical reflectors at exactly the same distance from the transmitter as each other, then they would indeed present double the RCS of a single reflector. But if you moved one of them a quarter of a wavelength further away from the transmitter, then the signal would have to travel a quarter of a wavelength further out, and a quarter of a wavelength further back, so it would travel half a wavelength further -- i.e. it would be exactly out of phase with the echo received from the other reflector.

Sorry to be sceptical about this. A container vessel has billions of surface reflections cancelling each other out but still ends up with a very large radar cross section - because the reflections also add up. The likelihood of perfect cancellation must decrease with increasing reflective area. Additionally, the radar paints a target many times on each sweep, each paint increasing the probabilty of target detection. Radar would not work if multiple reflections always cancelled each other out and never added to each other.

[st599]

Sorry to be sceptical about this. A container vessel has billions of surface reflections cancelling each other out but still ends up with a very large radar cross section - because the reflections also add up. The likelihood of perfect cancellation must decrease with increasing reflective area. Additionally, the radar paints a target many times on each sweep, each paint increasing the probabilty of target detection. Radar would not work if multiple reflections always cancelled each other out and never added to each other.

You get constructive and destructive interference.

In Tim's example, items with a difference of (an odd multiple of) a 1/4 wavelength will totally cancel out. Similarly, at multiples of 1/2 wavelength, a constructive interference occurs. In between you get a bit of both and will end up with some signal.

Think Young's Slits.

The size of the Radar target doesn't change this.

For an example, think of your radio fading.and getting stronger when driving in town. Caused by multipath

[onesea]

Also sceptical about reflections cancelling each other out. I am sure it CAN happen but feel it is unlikely.

Having used radars regularly I am often surprised how radar can miss some targets and reflect form others.

For me the more metal there is up there the better chances of being "seen".

As reports say allot comes down to the diligence and skill of the radar operator. That goes for the ship OOW and any yachty operators out there!

[st599]

Also sceptical about reflections cancelling each other out. I am sure it CAN happen but feel it is unlikely.

Not only likely, but 100% inevitable. Do not think of your radio signal as a laser beam, it isn't colimated.

The beam width means that the transmitted signal will arrive at the target a) via a direct path and b) via a large number of reflected paths off waves, other objects, the troposphere etc. This induces a current in anything conductive it hits which causes another radio wave to scatter in all directions. Some of this induced signal will return to the Radar a) via a direct path and b) via a large number of reflected paths off waves, other objects, the troposphere etc.

Any 2 reflections which are 180 deg out of phase will cancel. Targets in real conditions will fluctuate in received signal.

However, a big target will be hit many times per sweep, on some the target will cancel and be undetectable, on some it wont. These are fed into a integrator - as long as some are detectable it will be shown.

Without going into Rayleigh and Rician fades, read:

http://books.google.co.uk/books?id=h...erence&f=false

http://books.google.co.uk/books?id=D...erence&f=false

http://books.google.co.uk/books?id=F...erence&f=false

http://en.wikipedia.org/wiki/Multipath_fading



PS you now know how stealth planes work

[st599]

Sorry to be sceptical about this. A container vessel has billions of surface reflections cancelling each other out but still ends up with a very large radar cross section - because the reflections also add up. The likelihood of perfect cancellation must decrease with increasing reflective area. Additionally, the radar paints a target many times on each sweep, each paint increasing the probabilty of target detection. Radar would not work if multiple reflections always cancelled each other out and never added to each other.

You don't need perfect cancellation, just enough to go below the minimum detectable signal to noise ratio.