Postioning of multiple Omidirectional antenna's (900mhz)

[cpscdave]

Howdy folks,
Looking for some advice from people who probably have more experience with this.
This site I'm at has 5 wireless serial radios (industrial wireless). They operate at 900MHZ.
Each radio is configured to have a wireless channel that is offset from the rest. They are currently configured to use channels 1,4,7,10,13. Each is configured to be a Single to Many. IE each of the 5 transmitter talks to multiple receivers.

4 of the radios are working as well as you could probably expect. We see mostly clear communications, with the odd bit of noise.
The last radio gets some communications through but mostly its noise. A couple of the receivers are slightly better but still pretty noisy.

Below I'll provide a list of some of the things we have tried to resolve this. But for now our next step is to completely reprogram all the transmitters/receivers to be on a different channel and bring them on one at a time to see if we can't pinpoint where the trouble is occurring.

Now to my main question. I've attached a picture of how the attenna's are arranged

It seems odd to have the 3 "stacked" above one another. Is this going to cause problems? They don't know which antenna this radio is using. I tried swapping two of the antenna's, but didn't seem to make a difference.

I'm thinking their issues boil down to line of sight issues. But that's just based on running out of other ideas :) If you have any other suggestions as to what we can do they would also be appreciated.
Thanks,
-Dave

Things we have tried:
-Verified that the serial communications is working from the reciever to the controllers. This appears to be working 100%
-Changing the receiver for a new one
-Changing the transmitter for a new one
-Moving the antenna on one of the more troubled receivers
-Swapping the antenna on the troubled radio for one that isn't having problems.

 

[berkeman]

Is that a metal building? It would work better to stack them on a vertical tower above that building, if possible...

 

[cpscdave]

Yes that is metal siding. I'll see if we can make that change.
How about the arrangement? Does stacking them above one another cause issues?

 

[berkeman]

Does stacking them above one another cause issues?

It helps to cut down desensitizing issues, with TX signals from one antenna not overloading the RX input stage of another nearby antenna. If the radios have great filter skirts on their RX stages, this may not be an issue, but if it's easy to do, then a vertical stack would probably be best.

 

[davenn]

and if they are supposed to be operating Omni directionally, they should ALL be above the roofline

there's no way you can achieve an omnidirectional radiation pattern in their current mountingDave

 

[Jeff Rosenbury]

Is the top one (on the stove pipe) the problem child?

I see a flange nearly touching the antenna. That can't be good.

Antennas are designed to work with a ground plane. Each antenna has its own needs. But placing the ground plane (metal) too close doesn't help.

Simply twisting the metal flange away from the antenna might help.

Mounting all of them in line on the stove pipe with the proper distance (use the brackets I see) might be a good solution electrically. Of course I don't know what that will do to the pipe's function. (It shouldn't bother it, but I'm not a chimney expert.)

The siding isn't ideal, but antennas are typically mounted on metal towers covered in coax, so they should handle it as long as they have the proper spacing.

 

[berkeman]

Mounting all of them in line on the stove pipe with the proper distance (use the brackets I see) might be a good solution electrically.

As long as the stove pipe is non-conducting...

 

[Jeff Rosenbury]

As long as the stove pipe is non-conducting...

Given the mounting brackets supplied with the antennas and that they are usually mounted on metal towers, they are likely designed to use the tower (or stove pipe) as a ground plane. I could be wrong about that, but I would hook it up and test it expecting it to work.

 

[davenn]

Given the mounting brackets supplied with the antennas and that they are usually mounted on metal towers, they are likely designed to use the tower (or stove pipe) as a ground plane. I could be wrong about that, but I would hook it up and test it expecting it to work.

these are supposed to be omni directional antennas
they will not work as designed when they have huge hunks of metal directly by them

it will severely distort the radiation pattern
it is also likely to result in large return losses (aka high SWR)

Antennas mounted on the sides of steel lattice towers are usually directional and they may or may not ( depending on the design)
take advantage of the tower metalwork as a reflectorDave

 

[cpscdave]

these are supposed to be omni directional antennas
they will not work as designed when they have huge hunks of metal directly by them

it will severely distort the radiation pattern
it is also likely to result in large return losses (aka high SWR)

Would just affect the transmission in the direction of the steel siding? or all directions?
The open side is the direction of the receivers.

Regardless I've also been concerned as to the height of the antennas. I'll see if I can convince them to move them all up.

Is the top one (on the stove pipe) the problem child?

Wish I could tell you. They don't know which antenna is which. This project was undertaken before I became involved. It's landed on me as everyone else (including the manufacturer of the radios) has pretty much told them "Its not working? Tough muffins"

 

[davenn]

Would just affect the transmission in the direction of the steel siding? or all directions?
The open side is the direction of the receivers.

Primarily on that side closest to the metal walls and chimney/air exhaust flue

( this is the radiation pattern problem spoken of)the other significant problem is, as I also mentioned in that previous post,
is the effect close metal objects will have to the function of the antenna

(This is the tuning / SWR/ Return loss problem)

 

[Jeff Rosenbury]

It is difficult (impossible?) to mount the antennas in a vertical line without having a conductor (the coax) running beside them. The mounting brackets should take this into consideration and provide a decent spacing for a decent VSWR.

Building an antenna with a good front to back ratio takes some work. A small back plane won't do it. ( A wall of siding might.)

You are correct that the VSWR may be off if this isn't true. You are also correct that they will be non-omnidirectional. But they should be close enough if the conductor is narrowish.

Thus, mounting and testing is in order.

Another option is to mount the radios apart from one another. This could be an obvious logistical problem.

Placing them within their nearfields will create an unintentional array with unintended results. Even in a line, there could be a problem. (I've built linear arrays, but the distance between the antennas is hard to match.)

Cpsdave have you checked the coax to see if they are good? It might not be an antenna problem.

 

[cpscdave]

We changed the wireless channel on the problem radios and everything is working well.
Kinda wish I had thought of trying this at the beginning of the week :)

It's odd that this worked since the signal to noise ratio on the radios was decent.

But thanks for all the insight. Next project where I get control from the start I'll put all this info to good use!

 

[sophiecentaur]

Antennas are designed to work with a ground plane.

Some Antennas are designed to work with a ground plane.

 

[davenn]

Building an antenna with a good front to back ratio takes some work. A small back plane won't do it. ( A wall of siding might.)

irrelevant ... they are supposed to be omni-directional, no need to bring building of directional antennas into it

You are also correct that they will be non-omnidirectional. But they should be close enough if the conductor is narrowish.

HUH ? doesn't make sense

We changed the wireless channel on the problem radios and everything is working well.
Kinda wish I had thought of trying this at the beginning of the week :)

It's odd that this worked since the signal to noise ratio on the radios was decent.

But thanks for all the insight. Next project where I get control from the start I'll put all this info to good use!

Awesome Dave ... antenna mounting can be a bit of an artform :)
it does make a huge difference to the overall operation of the system

Dave