Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Question about coax

  1. Nov 21, 2008 #1

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    We live in a rural area and can only get normal programming by satellite. So we use Direct TV, but we have a unique problem. For the house system, the antenna is located about 500 feet away because this the only place where we can get a low, southern shot, with the dish. This is antenna-to-base distance is beyond the reach of the DTV system using RG6/U cable.

    As I understand it, what happens is that DTV applies a potential to the center wire of the coax, which then selects the required transducer [for the desired channel] at the antenna, according to the voltage. If there is too much of a voltage drop across the cable, then the transducer and channel selection are out of sinc, which causes a loss of signal. This had me stumped for a time as we could get some channels with a decent signal strenth, but others would not work at all.

    I had this problem beat by retransmitting the signal from my office, which is much closer to the antenna. I found this nifty retransmitter developed for NASA that allows me to transmit the signal on CAT5. The video quality was excellent and everyone was happy. But, when we purchased a HD big screen TV for the house, we found that the retransmit system is vulenerable to ground differentials between my house and the office. This apparently has something to do with the scan rate of the TV. As a result, I have to keep my load in the office balanced or we get horizontal lines drifting upward on the screen. And best of all, if this happens, we can't change channels because the channel selection is broadcast from the house to my office and depends on the ground level! I assume that noise caused by the ground differential is the issue. We are also prevented from watching HD on the new TV except for on a few local stations that we get by regular antenna. But that is a minor issue: The load balancing business has to go. The people who designed the retransmitter have run out of options and ideas.

    So, my only hope is to select a coax that would reduce the DC line loss by half, while still retaining the required characterestics needed for the signal transmission. I was going to start reviewing this to see if I could make everything match-up, but before I do, I wanted to throw this out to see if this approach is fundamentally flawed. Is there any reason why I can't simply upsize the cable size, match capacitance and inductance, or at least keep the impedance to a mininum, and expect this to work? This cable will cost something like $700, so I hate to just try it and see how it goes. Also, I'm not sure what frequencies DTV uses.

    Any input is appreciated.
     
    Last edited: Nov 21, 2008
  2. jcsd
  3. Nov 22, 2008 #2
    You misunderstand the system.

    The system in the USA may be a bit different to Britain and the rest of Europe but the Low Noise Block (LNB) at the focus of the dish down converts the whole of the satellite band 10.5 - 12+ GHz (or the 4 GHz band) to 750 - 2000+ MHz. For channels on the lower part of the band the LNB local oscillator works at 9.75 GHz. For the higher part of the band the LNB local oscillator switches to 10.6 GHz. 22 kHz is fed down the coax to affect the switching. The DC voltage on the coax jumps between two states.. about 14 Volts for Horizontal polarisation and 16 Volts for Vertical (or the other way around).

    So 750-2000+ MHz (think it is called hyperband...intermediate frequency) travels down the coax and not the actual satellite frequency 10.5 - 12+ GHz. 9.75 or 10.6 GHz is subtracted to give the IF.

    The receiver then tunes to the required intermediate frequency (IF) and applies the 14 or 16 Volts and 22 kHz or not depending on what group of digital channels you are receiving. The receiver then selects the required channel within the group.

    A long coax run will attenuate the 750-2000 IF and there will be a DC volts drop which may prevent the polarisation switching. It wont affect the 22 kHz.

    Solution is to use and in-line amplifier for the IF near the dish and feed the DC via a thick power cable...bit messy.

    Alternative would be a better coax such as LMR600.. LDF4-50 Could be expensive. Maybe the amplifier idea would be much cheaper.

    Half way house would be a medium loss coax (LMR400) and an amplifier....also low DC loss so no messy separate power cable required.

    Low loss cables tend to 50 Ohm and satellites systems use 75 Ohm but I can't see it making much difference.

    So I would get some LMR400 and an in-line amplifier near the dish. The amplifier is powered by the DC on the coax and also loops the 22 kHz (if used in USA) to the LNB.
     
    Last edited: Nov 22, 2008
  4. Nov 22, 2008 #3

    dlgoff

    User Avatar
    Science Advisor
    Gold Member

    I did some googling for LNB amplifiers for DirecTV and found one that provides "Amplification to compensate for runs up to 250 feet of RG6 coaxial cable".
    http://www.satellitedish.com/cata0012.htm
    It's the Spaun USA WBA-425F 4-Way Wide Band Amplifier. I don't know how many LNB you have on your system. This one is for four. It's a little expensive IMO.

    edit: oops you said 500 feet. Man that's a long way.
     
  5. Nov 22, 2008 #4
    --------------------------------------------------------------------------------

    Thinking more.........................

    You don't need a low angle take-off.

    Satellites are typically 20 - 30 degrees above the horizon even in northern Canada. The dishes are 'off-set' feed so that they are mounted nearly vertically but the satellites are at a steep angle. Try the dish on or near the house.
     
  6. Nov 22, 2008 #5

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Great post! Thanks.

    I had two amplifiers inline. The channels that worked had a perfectly acceptable signal - signal strength of about 80%. Based on your explanation and what I observed, it would make sense that the polarization signal was the problem. We couldn't get about half of the channels.

    Will the DC loss be reduced by half using the LMR400? DTV only guarantees functionality to I think 250 feet [maybe 200].

    How do I identify the connection point for a separate polarization signal? Also, I don't understand how this ties in at the base unit.
     
    Last edited: Nov 22, 2008
  7. Nov 22, 2008 #6

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Also, as I said earlier, this is the only place on the property that I can get a shot at the satellite. We live in the hills and are surrounded by up to two-hundred foot trees to the East and South.
     
  8. Nov 22, 2008 #7

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Thanks. Yep, been down that road. 500 feet is a long way to go.
     
  9. Nov 22, 2008 #8
    It is not a separate polarisation signal. It is the actual power feed to the LNB. The LNB will contain a voltage regulator which will derive a fixed voltage (10 Volts say) from the incoming 14 or 16 Volts. The 10 volts will power the LNB. The LNB senses whether it is fed with 14 or 16 volts and switches polarisation. The receiver (base unit) provides the DC.

    You might not be getting some channels because of the DC voltage drop along the long coax is not allowing the LNB to switch polarisation and/or the much greater attenuation at the higher IF requencies towards and beyond 2000 MHz. In both cases a lower loss cable will help.

    You might need special N connectors for LMR400 and adaptors to F type.. Could be expensive.

    The inner conductor of LMR 400 is quite thick so DC losses should be much less.

    Looks like the DC loop resistance of LMR400 is 1.5 Ohms (inner and outer..total DC path) over 500 ft. That could be reduced further by running a parallel power cable and connection to the LNB body and the receiver ground. I think RG6/U will be about 4 Ohms. Hopefully it make all the difference but I CAN'T Guarantee it. Some versions of RG6 have a copper coated steel centre conductor and will have a much higher DC loss.

    A 500 mA current (LNB + amplifiers) will drop 0.75 V with 1.5 Ohms but it will be 2 Volts with 4 Ohms. The receiver sends about 18 V for one polarisation so can cope with a 0.75 Volt drop. 2 V will be too much.

    You need to look at prices for LMR400, LMR600 (better) and connectors.
     
    Last edited: Nov 22, 2008
  10. Nov 22, 2008 #9
    Other idea would be a power inserter like this...

    It blocks the DC from the base unit but passes the IF signal. It allows an separate DC supply unit to be used which could send a higher voltage to compensate for the voltage drop. You would need a 1 Amp 0 - 25 Volt variable voltage supply. You would have adjust the voltage for HOR or VERT. Not very convenient but would solve the DC volt drop problem. OR.. two of these devices back to back at the receiver and a fixed 2 Volt supply between the two.. In other words add an extra 2 - 3 Volts to the receiver output to compensate for the DC drop. It would be handy if you could just turn up the output voltage from the receiver but that mught be easier said than done.


    http://www.summitsource.com/f-type-...volt-2-amp-21450-mhz-part-8001ifd-p-7309.html


    Except that that one says it only works to 1450 MHz. mmmm

    I guessing a lot of this as I don't know the full parameters of the system there.
     
  11. Nov 22, 2008 #10

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    You have been very helpful. Thank you.

    I will post and let you know how it worked out. Also, I'm always open to additional suggestions.

    I understand that there are no promises here.
     
  12. Nov 22, 2008 #11

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Oh yes, just fyi, I should mention that I don't think the problem was ever signal attenuation because the channels would come and go. When they worked, the signal strength was still about 80%. Then the channels would just disappear. That was part of what really had me stumped. And it even seem to be related to the time of day, which I assumed to be a temperature issue.

    This of course was all after I had exhausted the input from the DTV techs, replaced the antenna and cable, and replaced the boxes.

    This has been quite the ordeal... but fun.
     
    Last edited: Nov 23, 2008
  13. Nov 23, 2008 #12
    From a website...

    Polarisation switching is controlled by dc voltage supplied by the receiver. 12.5v to 14.5v gives vertical and 15.5 to 18v gives horizontal polarisation. A higher voltage than that may damage the LNB but most are OK up to 20v. A voltage that is too low (less than 12.5v) will prevent the LNB from working correctly.


    http://www.satcure.co.uk/tech/lnb.htm


    If it is the polarisation switching then the Horizontal channels would be missing OR intermittent if the voltage at the LNB was near the switching point. If you know which channels were missing then you look up up what polarisation they are using.

    If they were permanently missing then turning the LNB through 90% would restore them and prove it is a polarisation switching problem.

    King of Sat and Lyngesat.. (lyngsat) have info on all the satellites and channels.

    http://en.kingofsat.net/satellites.php


    Somebody is having a prob with HOR. Somebody has suggested it may be due to a long cable run.



    http://www.satelliteguys.us/free-air-fta-discussion/124560-problem-horizontal-polarization.html


    If it so not the polarisation switching then it could be the 22 KHz band switching. If the 22 Khz source in the receiver (base unit) has a high impedance output the capacitance of a long cable run will reduce the level of the 22 kHz. That would be a tricky one to solve. A low loss cable wouldn't help. If it a low impedance ,,near the 50 -75 Ohms of the coax then then the 22 KHz will suffer very little loss.
     
    Last edited: Nov 23, 2008
  14. Nov 23, 2008 #13
    If you are considering getting another cable check out the "RG11." The cable industry uses it for drops more than 200 feet long. The cable is as thick as a thumb, but exhibits very low attentuation.
     
  15. Nov 23, 2008 #14
    RG11 is a bit lossier than LMR400 but being 75 Ohm might have a higher DC resistance. But it may be all copper rather than the copper coated aluminium of LMR400.
     
  16. Nov 24, 2008 #15

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I will probably play with this over the holiday and order something next week. Additional suggestions are always appreciated, and I will let you know how it goes, but it may take a little time.

    I have to trench a ~400 foot long ditch. :grumpy: Of course, on the up side, the coax will be large enough to add support to the foot bridge across the creek. :biggrin:
     
  17. Nov 25, 2008 #16
  18. Nov 26, 2008 #17
    Andrew make LDF4-50 and much thicker cables. Problem is the cost.

    LMR600 OR variants might be a cheaper option. Maybe with one amplifier (or none) at the dish end.
     
  19. Dec 2, 2008 #18
    Have you considered cat5 shielded cable - designed to reduce electrical interference. If the new TV is creating interference with the cat5 signal this will help.
     
  20. Oct 21, 2009 #19

    Ivan Seeking

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Thanks again for all of the input, but we have a change in plans. As a stopgap, I will be running a separate power wire for the DTV and retransmitter, from the house to the office. I can buy 500 feet of low-cost direct burial wire and hide it pretty easily. This will eliminate the ground differential issue between the two buildings and the need for load balancing in my office. Then, rather than going for DTV HD, we will begin shifting to internet TV. Already, many of my favorite programs are available online in HD.

    Between Hulu, Netflix, the YouTube shows, and a number of other services, online TV is coming along very quickly.
     
    Last edited: Oct 21, 2009
  21. Oct 21, 2009 #20
    Good choice. Even a major cable company Comcast, will soon allow internet customers to watch some of the cable content from the computer next year.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Question about coax
  1. RADAR Power and COAX (Replies: 8)

Loading...