Slotted waveguide communications system

[kedam]

Hello,

I'm looking for some insightfull answers for the following topic:
I have a (slotted) waveguide where multiple (50) antennas (for sending and receiving) are located in a slot that runs along the waveguide (200 meters long). On both ends of the waveguide are probes located for the wireless signal (5GHz) to send through it. My questions are as follows:

1) If there are little metal pieces in the waveguide, do they reflect the signal or absorb it? And will this result in a signal drop only on that specific location or will it effect the entire (200m) waveguide? (overal signal strength)

2) If the position of the antenna in the slot (of the waveguide) isn't optimal (tilted etc), will this effect the way of transmitting signals from the antenne through the waveguide (to some receiver)? Will this be more difficult to send and receive data? why?

3) Can a damaged antenna effect the overal signal quality in the waveguide? why (not)?

 

[berkeman]

Welcome to the PF.

Can you post a better drawing of the setup? I don't see anything that is 200m long in your sketch (and it is too small and fuzzy to make much out.

What is the application? What are the dimensions involved, and what is the range of frequencies that are being used?

 

[kedam]

Welcome to the PF.

Can you post a better drawing of the setup? I don't see anything that is 200m long in your sketch (and it is too small and fuzzy to make much out.

What is the application? What are the dimensions involved, and what is the range of frequencies that are being used?

Thank you,

In short, the system is used for communication purposes. The picture that I previously posted contains a slice of the slotted waveguide with a antenna in it. The waveguide itself is just 200m long. See attachment for some better pictures. The frequency that is being used is 5GHz.

 

[berkeman]

Is this antenna array used for both Tx and Rx? Or just used in one direction?

What is the purpose of having the antennas spaced out like that? Are they an integral number of wavelengths apart? Is there some beam-shaping intended with this array? It will be a constant shape given the fixed dimensions, right?

 

[sophiecentaur]

Is this system for communicating in a tunnel, perhaps ? I get the impression that it's not a transmitting array so much as a near field distribution system, in which each individual antenna serves a limited area. The far field radiation pattern in free space of such a set up could be very difficult to determine. It would be a sort of traveling wave antenna with, perhaps a major lobe facing near the direction of the feed axis and with a lot of sidelines.
Presumably each antenna is coupled fairly lightly into the transmission line so it's unlikely that a faulty element would have a great effect on the others (with their 2% of the power).
PS, the little bits of metal are probably for matching the guide to take account of the load from each antenna.

 

[kedam]

Is this antenna array used for both Tx and Rx? Or just used in one direction?

What is the purpose of having the antennas spaced out like that? Are they an integral number of wavelengths apart? Is there some beam-shaping intended with this array? It will be a constant shape given the fixed dimensions, right?

The antenna is used for both receiving and transmitting (sensor) data. The purpose of having 2 (about 2 meters from apart) antennas/driving module is to ensure that there is always communication (when a module switches out of a waveguide to another). And yes, the shape is constant.

 

[sophiecentaur]

The antenna is used for both receiving and transmitting (sensor) data. The purpose of having 2 (about 2 meters from apart) antennas/driving module is to ensure that there is always communication (when a module switches out of a waveguide to another). And yes, the shape is constant.

You know exactly what this question is about but, from what you have written, the way this thing operates is not at all clear to me. Can we have some background please?

 

[kedam]

You know exactly what this question is about but, from what you have written, the way this thing operates is not at all clear to me. Can we have some background please?

Of course,

The system is used in a car manufacturing plant for an electric monorail system. Basically, it is a way of transporting components (doors, cars etc.) through the factory at computed speeds, depending on production rate. Each module (with a car component on it) needs to be able to receive and send data with a segment controller (top computer) to let the computer know where it is located in the factory, which speeds, and how much current the servomotor consumes etc.

 

[sophiecentaur]

OK thanks. It looks an interesting system. So each antenna stands on its own and feeds the nearby sensors and there may be blank spots in between the antennae. The way that the elements are all fed will tend to make the fields from successive elements be nearly in phase at some distance from the wire and there would be no serious nulls. Right next to the wire, there will be regions where the signals cancel and produce nulls.
The performance will be largely the same when used to transmit or to receive (reciprocity) and it is often easier to bear this in mind and just look at the Transmit situation. This may not be enough if there are high levels of interference involved - but that's a separate issue.

 

[kedam]

OK thanks. It looks an interesting system. So each antenna stands on its own and feeds the nearby sensors and there may be blank spots in between the antennae. The way that the elements are all fed will tend to make the fields from successive elements be nearly in phase at some distance from the wire and there would be no serious nulls. Right next to the wire, there will be regions where the signals cancel and produce nulls.
The performance will be largely the same when used to transmit or to receive (reciprocity) and it is often easier to bear this in mind and just look at the Transmit situation. This may not be enough if there are high levels of interference involved - but that's a separate issue.

Thanks for the help
So to go back to my questions:
1) The metal pieces will reflect the signals and this can result interference only at the side which bounce back towards the source or will it also effect the signal strength at both ends of the metal piece?

2) So the positioning of the antenne in the waveguide slot will be of importance for absorbing the signal? How can this be explained?

3) Because of the small amount of coupling, the damaged antenna will be af small importance for the overal performance of the waveguide and will not effect the other antennas?

Also, is it possible to adjust (drop) the signal strength at a specific interval of waveguide? by using for example metal pieces?

 

[sophiecentaur]

1. The metal pieces are just part of the transmission line. They disturb the wave a bit and behave like added Cs and Ls on a regular transmission line, as far as I can see. They prevent unwanted reflections of the signal back down the guide due to the disturbance that the antenna causes. The antennae are like holes in a water hose and each one let's some power out, which gives you an array of elements, fed from the waveguide. I think the point of having a slot is that you can choose where to put the elements. Nothing comes out of the slot except where the elements are placed (basic waveguide theory)
2. I am using the Transmission model, for convenience ( you will just have to accept reciprocity - it's ok trust me) The position of all elements will affect the radiated power everywhere BUT only the nearby one or two element will have much effect.
3. Losing an element will reduce the field near it but, at a distance, the other nearby slots will give some coverage.

This sort of arrangement is NOT PERFECT. There will be dead spots but the fact that the signal traveling down inside the guide is traveling at almost the same speed as the signal along the outside, means that the leaked signals tend to augment each other in most of the surrounding space - much better than some other arrangements.
There are alternative ways of doing this sort of thing. It is possible to use 'leaky feeder', for VHF signals, which consists of a coax cable with a very open weave outer. That's a very cheap and cheerful way of feeding signals into a tunnel. Same basic principles the signal leaks out all along the feeder and provides fairly good coverage all along a tunnel. I believe this has been used in road tunnels to provide VHF Radio service for motorists and in coal mines, where the risk of fire is very high.

 

[Baluncore]

I assume here that the mobile units move their two couplers with them and benefit from spatial diversity when changing waveguides. As a mobile units travels, at least one of the two TR coupling elements will be in a guide.

If the lips on the waveguide slot are dimensioned correctly, (λ/4 = 15mm ?), then the slot will not radiate except when a coupling element passes through the slot into the guide. The waveguide will appear to be a long transmission line with a small percentage of the energy removed by each coupler present. At the far end, the waveguide can be terminated in a resistive load or matched receiver to prevent standing waves on the line.

Any metal particles or swarf that gets into the guide will cause some reflection and therefore a reduction of signal in the waveguide. The slot could be covered with two rubber or fabric lips to prevent entry of small metal particles, that would still allow the coupler to move along the slot, (similar to the methods used with glass scale DROs). Maybe a dedicated guide-cleaning unit could routinely roam the network collecting stray iron particles with a magnet or vacuum particles out of the slot.

The orientation of the couplers in the slot will not be critical since they are only coupling a few percent of the signal. Coupler dipole orientation will only need to be within about 30° since the small dipole couplers have a very wide beam pattern.

A bent coupling antenna should not be a problem unless it actually contacts the metal lips.

The waveguide couplers at the ends must be better adjusted in order to match the impedance while exciting the optimum mode in the waveguide.

 

[kedam]

And why do metal particles reflect and the antennas absorb the signals? (basic question)

Is it possible that the signals coming out of the waveguide can interrupt a magnetic field of permanent magnets? (there are located next to it, they are used for switching reed contacts)

 

[Baluncore]

And why do metal particles reflect and the antennas absorb the signals? (basic question)

Mirrors are made from metal because conductive metal is a good reflector of EM waves. Current flows on the surface of the metal due to the incident EM wave, which radiates a reflected EM wave traveling away from the conductor. Inside a conductor they cancel, outside they travel in opposite directions.

The 'antenna' probes are designed to intercept only a small percentage of the energy. They are impedance matched so they will carry that intercepted energy into the RF electronics without reflection.

Is it possible that the signals coming out of the waveguide can interrupt a magnetic field of permanent magnets? (there are located next to it, they are used for switching reed contacts)

Permanent magnets are effectively a DC field, while the RF field is an AC field.
The RF field changes too quickly to effect mechanical reed switches.
In effect, the permanent magnet and RF are on different frequency bands.

 

[sophiecentaur]

And why do metal particles reflect and the antennas absorb the signals? (basic question)

There is a fundamental difference between a random piece of metal and an antenna. A random piece of good conductor will have currents induced in it by any incident EM wave. But the energy is not (cannot be) absorbed because there is no (virtually) resistance so it will all be re-radiated (it has to go somewhere). On the other hand, an antenna is connected to a resistive load (the receiver input) and is designed to channel signal power into that resistance. A random object that's made of a resistive material will also dissipate some of an incident EM wave.