Microwave communications

I need to design a system that communicates in the microwave range (>
5GHz) using a horn antenna... Does any body know if there are any off
the shelf chipsets, or can point me in the direction of any reference
designs or manufacturers. The system would be very low power.

Cheers
Ross

On Wed, 30 Jul 2008 23:31:32 -0700 (PDT), ross.g.marchant@gmail.com
wrote:

>I need to design a system that communicates in the microwave range (>
>5GHz) using a horn antenna... Does any body know if there are any off
>the shelf chipsets, or can point me in the direction of any reference
>designs or manufacturers. The system would be very low power.

In which countries should this system operate. Are you prepared to do
the type acceptance testing required in some countries ?

A horn antenna would suggest line of sight communication and quite
high effective radiation power (EiRP), so this would rule out any
licence-free bands and a license is required for the system.

You did not specify the frequency range, but it is hard to get any
chip sets running directly at say 10 GHz. In practice you would have
to use a receiver preamplifier on the microwave frequency, then mix
down the received microwave signal to a lower intermediate frequency
(IF) and then do the processing at a lower frequency. 

For transmitting, generate the signal at some lower frequency, mix it
up with a local oscillator, apply some power amplification before the
antenna.

Paul

On 31.07.2008 08:31, ross.g.marchant@gmail.com wrote:
> I need to design a system that communicates in the microwave range (>
> 5GHz) using a horn antenna... Does any body know if there are any off
> the shelf chipsets, or can point me in the direction of any reference
> designs or manufacturers. The system would be very low power.
> 
> Cheers
> Ross

Try this one:

http://www.microwaves101.com/cgi-bin/messageboard/yabb.pl

Best Regards,
Andreas

<ross.g.marchant@gmail.com> a &#4294967295;crit dans le message de news: 
74d5f1cf-3115-45c0-b38e-7d0154644fbb@q28g2000prh.googlegroups.com...
>I need to design a system that communicates in the microwave range (>
> 5GHz) using a horn antenna... Does any body know if there are any off
> the shelf chipsets, or can point me in the direction of any reference
> designs or manufacturers. The system would be very low power.
>
> Cheers
> Ross

Mimix has some chips for the 24GHz band (U1010, etc). Hittite too.

Cheers,
Robert
www.alciom.com
The mixed signal experts

On Jul 31, 5:20=A0pm, Paul Keinanen <keina...@sci.fi> wrote:
> On Wed, 30 Jul 2008 23:31:32 -0700 (PDT), ross.g.march...@gmail.com
> wrote:
>
> >I need to design a system that communicates in the microwave range (>
> >5GHz) using a horn antenna... Does any body know if there are any off
> >the shelf chipsets, or can point me in the direction of any reference
> >designs or manufacturers. The system would be very low power.
>
> In which countries should this system operate. Are you prepared to do
> the type acceptance testing required in some countries ?
>
> A horn antenna would suggest line of sight communication and quite
> high effective radiation power (EiRP), so this would rule out any
> licence-free bands and a license is required for the system.
>
> You did not specify the frequency range, but it is hard to get any
> chip sets running directly at say 10 GHz. In practice you would have
> to use a receiver preamplifier on the microwave frequency, then mix
> down the received microwave signal to a lower intermediate frequency
> (IF) and then do the processing at a lower frequency.
>
> For transmitting, generate the signal at some lower frequency, mix it
> up with a local oscillator, apply some power amplification before the
> antenna.
>
> Paul

Thanks for the replies.

We are investigating a microwave link through a small steel pipe that
is about 3km long. From my basic calculations the cutoff frequency for
this waveguide was about 2GHz so I was thinking about >5GHz to
minimise transmission losses. I think that a horn antenna would be
good to direct the energy, but do they transmit at a particular
'mode'? Would common 2.4GHz stuff like 802.11a / b / g work better? I
haven't really looked into this kind of thing before, so I don't
really know how much work would be required :)


Hmmm gotta get a newsreader... google groups just pasted my email all
over the net

On Thu, 31 Jul 2008 20:33:36 -0700 (PDT), geomet
<ross.g.marchant@gmail.com> wrote:

<snip>
>
>We are investigating a microwave link through a small steel pipe that
>is about 3km long. From my basic calculations the cutoff frequency for
>this waveguide was about 2GHz so I was thinking about >5GHz to
>minimise transmission losses. I think that a horn antenna would be
>good to direct the energy, but do they transmit at a particular
>'mode'? Would common 2.4GHz stuff like 802.11a / b / g work better? I
>haven't really looked into this kind of thing before, so I don't
>really know how much work would be required :)
>
Why not just run some fibre thru the pipe?  Ought to be a lot easier
than messing with microwaves.  Probably get a lot better bandwidth,
too.

-- 
ArarghMail807 at [drop the 'http://www.' from ->] http://www.arargh.com
BCET Basic Compiler Page: http://www.arargh.com/basic/index.html

To reply by email, remove the extra stuff from the reply address.

On Thu, 31 Jul 2008 20:33:36 -0700 (PDT), geomet
<ross.g.marchant@gmail.com> wrote:

>We are investigating a microwave link through a small steel pipe that
>is about 3km long. From my basic calculations the cutoff frequency for
>this waveguide was about 2GHz 

If the dielectric constant of the gas significantly differs from that
of air, did you take this into consideration when calculating the
cut-off frequency ?


>so I was thinking about >5GHz to
>minimise transmission losses. 

Components made for 2.4 or 5.7 GHz WLAN systems should be of interest.

>I think that a horn antenna would be
>good to direct the energy, but do they transmit at a particular
>'mode'? 

Microwave power is usually connected to the waveguide using an
electric or magnetic probe. The magnetic probe might be a better
choice, since it can be made smaller, thus causing less resistance to
the gas flowing in the tube.

There is no need to use a directional horn, since the waveguide is
directing the power anyway. The omnidirectional probe will launch
power in both directions along the waveguide, but placing a metallic
screen (net) at a critical distance from the probe will direct the
power in one direction only and thus, increase the signal by 3 dB.
Such a small increase might not justify the cost of the screen.

At least the fundamental propagation mode does not like sharp bends,
frequencies well above the cut-off frequency in some higher
propagation mode might propagate better through sharp bends.

With a very high frequency signal compared to the waveguide cut-off
frequency, there can be multiple propagation modes just like in a
multimode optical fiber. In these the maximum _symbol_ rate for a
specific distance is limited by the dispersion due to multiple
propagation modes having different propagation times. To get around
this, WDM (Wavelength Division Multiplexing) is sometimes used on
VHF/UHF/SHF OFDM is often used to reduce the symbol rate by sending
multiple streams in parallel at different frequencies.

With 2 GHz cut-off and 5.7 GHz signal there are only a few possible
propagation modes, so you might get away without OFDM, but at least
you should be able to change the frequency slightly, if there are some
unfavorable propagation modes canceling each other at a specific
distance. 

However, if the dielectric constant of the gas might vary e.g. due to
change of temperature, pressure or consistence, the location of nulls
might change, calling for frequency changes. To avoid this, OFDM or
some wide band spread spectrum system could be used.

Paul
 

On Aug 1, 3:47=A0pm, Paul Keinanen <keina...@sci.fi> wrote:
> On Thu, 31 Jul 2008 20:33:36 -0700 (PDT), geomet
>
> <ross.g.march...@gmail.com> wrote:
> >We are investigating a microwave link through a small steel pipe that
> >is about 3km long. From my basic calculations the cutoff frequency for
> >this waveguide was about 2GHz
>
> If the dielectric constant of the gas significantly differs from that
> of air, did you take this into consideration when calculating the
> cut-off frequency ?
>
> >so I was thinking about >5GHz to
> >minimise transmission losses.
>
> Components made for 2.4 or 5.7 GHz WLAN systems should be of interest.
>
> >I think that a horn antenna would be
> >good to direct the energy, but do they transmit at a particular
> >'mode'?
>
> Microwave power is usually connected to the waveguide using an
> electric or magnetic probe. The magnetic probe might be a better
> choice, since it can be made smaller, thus causing less resistance to
> the gas flowing in the tube.
>
> There is no need to use a directional horn, since the waveguide is
> directing the power anyway. The omnidirectional probe will launch
> power in both directions along the waveguide, but placing a metallic
> screen (net) at a critical distance from the probe will direct the
> power in one direction only and thus, increase the signal by 3 dB.
> Such a small increase might not justify the cost of the screen.
>
> At least the fundamental propagation mode does not like sharp bends,
> frequencies well above the cut-off frequency in some higher
> propagation mode might propagate better through sharp bends.
>
> With a very high frequency signal compared to the waveguide cut-off
> frequency, there can be multiple propagation modes just like in a
> multimode optical fiber. In these the maximum _symbol_ rate for a
> specific distance is limited by the dispersion due to multiple
> propagation modes having different propagation times. To get around
> this, WDM (Wavelength Division Multiplexing) is sometimes used on
> VHF/UHF/SHF OFDM is often used to reduce the symbol rate by sending
> multiple streams in parallel at different frequencies.
>
> With 2 GHz cut-off and 5.7 GHz signal there are only a few possible
> propagation modes, so you might get away without OFDM, but at least
> you should be able to change the frequency slightly, if there are some
> unfavorable propagation modes canceling each other at a specific
> distance.
>
> However, if the dielectric constant of the gas might vary e.g. due to
> change of temperature, pressure or consistence, the location of nulls
> might change, calling for frequency changes. To avoid this, OFDM or
> some wide band spread spectrum system could be used.
>
> Paul

Thanks for letting me pick your brains :)

Actually its not a gas pipe, its used for pumping so it will be full
of air, also no sharp bends. I shouldn't give too much away, but we
need to send a sensor system to the end of the pipe. Other systems use
a cable with power and data, however we are looking at something
battery powered with some kind of wireless communications, as cable is
both heavy and wears out easily. With the probe, does it need to
energise off the waveguide (pipe),? There is do direct connection to
the pipe. I see what you mean about the horn - not really needed. I
will start looking at the WLAN stuff. Also I saw on wikipedia the
802.15.3a spec is for MB-OFDM so maybe that would work too (actually
google says its dead). Would anything using PSK for the modulation be
a problem? Anyway, thanks again :)