Home page of Satellite Internet and Information

Satellite Internet Forum.

Welcome, Guest.
Welcome to this satellite broadband discussion forum. Wherever you are and whatever your problem we are here to help each other. Connecting to the internet via satellite is not always easy but is critically important to those in remote places or with poor terrestrial infrastructure. Both service providers and customers are encouraged to contribute. Register at the bottom of the forum home page if you wish to contribute or ask question. Read the Forum rules.
      Satellite Internet Forum : Home Page          
Pages: 1

Andrew 244 components on a 243?

(Read 7342 times)
Ex Member
Ex Member


Oct 3rd, 2007 at 10:20am  
Does anyone know if it is possible to replace the feed assembly on an Andrew 243 Ku antenna with the feed assembly from the 244?

I am not getting sufficient cross-pol on the 243 for some reason even though the dish is certainly pointed as accurately as possible.  I want to attempt to use the dual optic design to increase cross-pol performance, but because it is very difficult to get to the site I want to avoid sending an entire dish. Thus I'd like to take the dual optic feed from the 244 and put it on my 243. 

Anyone ever done this or know if it is possible?
Back to top
 
 
IP Logged
 
Eric Johnston
Senior Member
***
Offline


Personal text from Profile,
Options, Top line

Posts: 2109
Reply #1 - Oct 3rd, 2007 at 6:18pm  
Here are the Andrew antenna data sheets

244 Ku band dual optic
http://www.andrew.com/search/BN_PA-100556.1-EN.aspx

243 Ku band prime focus
http://www.andrew.com/search/BN_PA-100572.1-RU.aspx

A problem with swapping the feed assemblies is that the feed arm also is different. Ask Andrew if the focal length of the reflectors is the same and if the feed arm and feed/subreflector may be exchanged.

Some ideas that may avoid the need for change.

1. Adjust the polarisation angle accurately. This is very difficult as you need to turn the feed in 1 deg increments or less. If the elevation angle is very high, the feed may be high up and needs a ladder for access, and a single polarisation adjustment and measurement may take 30 minutes or more.  Note that you can't get accurate measurements if any part of your body or the ladder is obstructing the beam, so you have to remove the ladder and then get the hub to measure. In these situations I would suggest making four cross-pol measurements at four carefully marked angles at about 5 deg increments and then writing down the results and interpolating for the best angle. Get the hub to make really accurate measurements.  It takes time particularly when the results are like 28 to 30 dB.

2. Check the dish rim is flat. Put fishing line tight up and down and across.  The lines should just touch. If not the dish is distorted, like the sides pulled forwards or pushed backwards.  This may not affect the polarisation purity but may improve the gain.

3.  If you tip the feed horn down, and perhaps use a larger diameter feed horn, so it aims just 1/3 up the dish then cross pol performance will improve, but the gain will go down as the upper edge of the dish is hardly illuminated and the dish appears to be smaller.  The f/d ratio increases and the offset angle decreases, so the cross pol improves.

Best regards, Eric.
Back to top
 
 
IP Logged
 
Ex Member
Ex Member


Reply #2 - Oct 3rd, 2007 at 6:54pm  
The dish is on an elevated platform as you can see here:

http://www.broadwave.com/images/iraqsat.jpg

The elevation is about 42 degrees, minus the offset of 22.6 making it about 19-20 degrees.  As it is on that platform which is about 15ft high, we are using a ladder to get to the feed.  As the dish is offset and we are below the feed, I would not expect that the ladder is causing a shadow on the dish to reduce crosspol isolation.

The other problem we have is the people at the HUB are incredibly impatient and I do not think they will wait for us to make an adjustment, climb down, move the ladder, measure, then move the ladder back, climb back up, make adjustmet...etc.etc

Another strange issue is yesterday we got about 25db.  Then today when we tried, with absolutely no adjustments being made to the dish, we got 20db.  Would atmospheric conditions cause such a dramatic change?  I wouldn't think so.

We will check the dish to see if it is warped, although I highly doubt it.  I will update with our findings.

Unfortunately due to the remoteness of the site, we do not hhave another feedhorn onsite and it will take some time to get another onsite.  Time that we do not have as we are drastically behind schedule because of all the issues we have had.

Andrews did not indicate the focal points of the two dishes, however I had planned on replacing the entire feed arm when doing this, so unless the curve of the parabola is different it should be fine.  They said that the main reflectors are identical so I do not see a problem.  Of course the reason for doing such a conversion is that it is much easier to ship a feed assembly than an entirely other dish/feed/etc.

In another note, do you know of any other providers with coverage in Iraq that uses extended Ku but not on Badr-4?  I would like to try working with another satellite just as a sanity check, but the extended Ku makes it difficult to find someone who can provide it. 

If there is another provider that can do extended Ku on another bird and we can connect to it and it works.  We will definitely go with the other provider as that is the fastest solution to get this site online.

I appreciate all the help

--Chris
Back to top
 
 
IP Logged
 
Eric Johnston
Senior Member
***
Offline


Personal text from Profile,
Options, Top line

Posts: 2109
Reply #3 - Oct 4th, 2007 at 9:37am  
If you can use a ladder so as to reach up under the feed arm to make the polarisation adjustment that should be fine, as you won't be blocking the beam.  The person adjusting needs to be using a phone to talk to the hub while the adjustment is made.  The movement required to get into the deep cross polar null is very small as the deep null is only a degree or two wide.

The assembly of the feed horn to the OMT is important.  The two flanges are both circular, they must be clean and they must align exactly straight and co-centric.  If there is a circular rubber O ring check the flanges are parallel.  The 4 screws need to be all done up the same.  The diameter of the screws should be such that there is no possibility of a sideways misalignment.

Someone who installed a transmit reject filter the wrong way round had a poor cross pol performance.  
...
Note the longer end of the TRF is towards the feed (in this specific correct example).  Note the faces of the two circular waveguide flanges at the circular joint are parallel.   The four screws are tightened evenly.

Your present feed configuration is shown below:
...
The polarisation appears to be set for vertical receive with almost zero adjustment angle applied. Maybe your satellite is almost due south ?

Check your required polarisation adjustment angle with:
http://www.satsig.net/maps/satellite-dish-pointing-iraq.htm
If the answer is that a positive polarisation adjustment is required, then from the nominal horizontal or vertical starting position, rotate the feed clockwise, while facing towards the satellite.

Best regards, Eric.
Back to top
« Last Edit: Oct 4th, 2007 at 10:38am by Eric Johnston »  
 
IP Logged
 
Ex Member
Ex Member


Reply #4 - Oct 4th, 2007 at 11:05am  
Wow Eric.  I appreciate the detailed response. 

Regarding the pic of our dish, that was before the polarization was applied.  We require about +23 degrees polarization for our site and have adjusted as such.

I will have the guys inspect the feed horn as you indicated and see if they notice anything out of the ordinary.  We do not have a Tx reject filter, so that possibility is out.

I hopefully will attempt with another satellite tomorrow providing a standard Ku BUC arrives onsite (we are using extended right now).  We'll see how that goes.

Assuming the feed is ok, my idea to adapt the 243 antenna to a dual optic design of the 244 should increase the isolation correct? 

Again, thank you for all the info.

Back to top
 
 
IP Logged
 
Eric Johnston
Senior Member
***
Offline


Personal text from Profile,
Options, Top line

Posts: 2109
Reply #5 - Oct 4th, 2007 at 12:32pm  
You must have a transmit reject filter (TRF) between the OMT and the LNB or you should have an LNB with an inbuilt transmit reject fiter (e.g. like http://www.invacom.com/products.htm used with their OMT junction or integrated BUC)

The purpose of the TRF is to protect the LNB from spurious transmit energy which will, at best, desensitise the LNB and, at worst, burn out its input stage.  The TRF is mounted on to the OMT.  Any 14 GHz power that comes up this arm gets reflected back down agin once it hits the filter elements.  The distance up and down seems to be important so the TRF should be mounted as shown above, not the wrong way round with the short end against the OMT, nor with the 90 deg H bend first and the TRF next and then the LNB.

A missing TRF might explain the poor cross pol.  The 14 GHz reflection (if it is not absorbed in a burnt out front end stage) will be at the LNB and the distance back to the OMT junction will be wrong.

All the above about TRF filters and poor cross-pol is based on someone else's experience that installing the side arm end to end, with the 90deg angle at the OMT caused them severe x-pol and putting it the right way so it stuck out sideways, immediately solved their problem.  I have never verified this myself.

Best regards, Eric.
Back to top
« Last Edit: Oct 4th, 2007 at 9:51pm by Admin1 »  
 
IP Logged
 
Ex Member
Ex Member


Reply #6 - Nov 21st, 2007 at 12:49am  
now , whts the better way to make a better link : better upload and better download , is it using cross pol with tx reject filter ? or use this dro drom invacom ?
Back to top
 
 
IP Logged
 
Ex Member
Ex Member


Reply #7 - Nov 21st, 2007 at 6:11pm  
hello dear Eric , i need to know the difference between PLL and DRO LNB , can we use DRO LNB to recieve SCPC service ? or we must use PLL with cross pol ? and what about some links i saw , they are using crosspol without any TX filter and they are using 8Watt BUC for 3 years without any problems ?
Back to top
 
 
IP Logged
 
Eric Johnston
Senior Member
***
Offline


Personal text from Profile,
Options, Top line

Posts: 2109
Reply #8 - Nov 21st, 2007 at 6:41pm  
Dielectric Resonator Oscillator (DRO) LNBs are unstable in frequency.  If you are trying to tune to a wide carrier, like 27.5 Msps, then a frequency error of 1 MHz does not matter.  The receiver can find the carrier.  This is normal for satellite TV type services.

Phase lock loop (PLL) LNBs use a crystal oscillator or an external 10 MHz reference and the final local oscillator frequency is phase locked back to the reference.  It has far higher frequency stability and low phase noise characteristic.  If you are trying to tune to a narrow 64 ksps SCPC carrier then you need a PLL LNB.  It would be difficult to find the carrier is the frequency tolerance was 1 MHz.

A few SCPC receivers may be able to sweep slowly  over +/- 2 MHz searching for an SCPC carrier and try to recognise it amongst many other similar ones.  I had some old BPSK Wegener DR95 music/audio receivers that would do this.  

Assuming the carrier can be found there is then the problem of phase noise.  Low bitrate SCPC with QPSK or higher order of modulation is sensitive to phase noise, so PLL is preferred.

Filtering at the front of the LNB protects it against transmit power leakage.  Leave the filter out and you risk slight or severe degraded receive performance and, in the worst case, damage to the LNB.

It is not certain if the lack of (or wrong position of) a transmit reject filter may also degrade the feed transmit x-pol performance.  All I can say is that someone reported, in one instance, that putting the Andrew TRF and 90 deg bend wrong way round, end to end, reduced their tx x-pol performance significantly.  I can't repeat that experiment.

Best regards, Eric.
Back to top
 
 
IP Logged
 
Ex Member
Ex Member


Reply #9 - Nov 21st, 2007 at 7:06pm  
a great help from you dear Eric as  always ..
so what i understand , its a matter of high and low symbol rate .. in dvb always its high like 27.500 , and we can use DRO without problems , but when we come to SCPC , its often low symbol rate , DRO cant help locking at the signal .. it has nothing to deal with frequencies or local ocillator of the LNB's .. right ?
i faced another situation when i tried to replace my perfectly running DRO with cross pol and PLL LNB , my recieving is DVB-S2 with symbol rate 27.500 , of course without TX filter , 4WATT BUC , i couldnot ever lock at the signal !!!! even 360 dgree rotating the polarization and spending so many time ..that was a test to compare which is better in ebno value ..
Back to top
 
 
IP Logged
 
Pages: 1