1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

SWM - Large building installation problems

Discussion in 'DIRECTV Installation/MDU Discussion' started by satguyj, Jul 20, 2009.

  1. satguyj

    satguyj New Member

    1
    0
    Jul 20, 2009
    I'm in a LARGE pickle right now. I have layers of IT mangment asking me why TV isn't working, and I just don't know. I'm a not an installer or RF enigneer, just a guy. ANY HELP would be apprecaited.

    Here's my situation:

    We have a SWM install done by our technology partner. Bottom line is all 8 receivers can't see the dish.

    Here's our setup from top of the building down:

    - 5 LNB and paired dish AU9-SL5-SWM
    - Quad shielded RG 6
    - ~200 ft of RG 6 from roof to IDF Closet

    - DirecTV PI-28 Switching Power Supply
    - A Sonora Designs LA 141R swm amp w. power supply in IDF closet
    - ~200 ft of RG 6 from IDF to IDF Closet

    - A Sonora Designs LA 141R swm amp w. power supply in IDF closet
    - ~200 ft of RG 6 from IDF to IDF Closet

    - Two (2) Sonora Designs LA 141R swm amp w. power supply in IDF closet
    ~200 ft of RG 6 from IDF to Head End rack.

    - A DirecTV SWM 1x8 splitter
    - 8 DirecTV H21 receivers


    We seem to have VERY sporadic ability to get signal. When we do, we can do activation, etc.. Wiring and alignment is all expected to be good, though we don't have a way to test for small line damage. A tone on each segment came back fine. Having only 1 receiver attached to the red "out" on the 8 way SWM splitter, worked for a day, now doesn't.

    I've been told repeatedly that this SHOULD work, but it doesn't.

    Is there anyone who has any luck with getting a SWM system working reliably @ 800 ft with SWM amps, etc....


    If SWM isn't an option, what would it take to get a 4 wire system reliably working?
     
  2. carl6

    carl6 Moderator Staff Member DBSTalk Club

    12,437
    929
    Nov 15, 2005
    Seattle, WA
    You might try posting this question in the MDU forum. There are several people there who are experienced in large building installations who may be better able to answer your question. Yours is well beyond the average/typical/normal configuration most are experienced with.
     
  3. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    SWM8 output = -30dBmV
    200' RG6 ~ -20 dB [-50 dBmV]
    Sonora amp = +14dB [-36 dBmV] add amp here
    200' RG6 ~ -20 dB [-56 dBmV]
    amp = + 14 dB [-42 dBmV] add second amp here if needed
    200' RG6 ~ -20 dB [-66 dBmV]
    amp = +14 dB [-52 dBmV]
    amp = +14 dB [-38 dBmV]
    200' RG6 ~ -20 dB. [-58 dBmV]
    8-way splitter ~ -12 dB. === -66 dBmV [without added amp(s)] one amp= -52 dBmV
    Receiver min level ~ -55 dBmV.

    [may be helpful] http://www.sonoradesign.com/images/LA141R_spec.pdf
     
  4. doctor j

    doctor j Hall Of Fame DBSTalk Club

    1,706
    50
    Jun 14, 2006
    Birmingham, Al.
    Sonora has 20 dB gain amps also.

    Doctor j
     
  5. Cap'n Preshoot

    Cap'n Preshoot Godfather

    452
    0
    Jul 16, 2006
    I agree, you have way too much loss in way too much cable length between existing amplifiers which themselves are inadequate.

    RG6 at satellite frequencies is marginal in residential applications (in my opinion). To deploy RG6 in a commercial installation is inexcusably bad design! The cable lengths might be OK using RG11, but with RG 6 at those lengths you're going to experience exactly what you're experiencing.

    I won't try to explain "slope" to you in this post, but suffice it to be said that ALL coaxial cable exhibits non-linear attenuation across the wide band of frequencies required and RG6 is particularly bad in this regard. For this reason, even using RG11, you need to be using slope-compensating amplifiers. Since the best of these from Sonora can only compensate for 5 db of slope, your design spacing should be at 5db slope spacing, not 14db loss spacing (which your installer didn't do anyway). Indeed Sonora has higher gain amps, but they do not offer any greater than 5db slope compensation, and only in those so-advertised. You need slope-compensating amplifiers, and twice as many as you presently have (at half the current spacing). Granted you will have a higher noise figure in 4 amps vs. 2, but you'll only have a max of 4 amplifiers in cascade, so don't sweat it.

    To summarize: Twice as many amplifiers, use slope-compensating amplifiers, place them half as far apart as they presently are and use RG11 between the dish and your first amplifier. RG6 the rest of the way is probably OK, though I wouldn't have used it myself. I use RG6 for horizontal drop cable, not vertical backbone (riser cable). I guarantee this will solve your problems. Fire the moron that originally installed it.

    .
     
  6. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    You need to tell that to Sonora then. They spec "this amp" @ 14 dB.
     
  7. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    I'm not sure if this may not be overkill.
    Slope variations between [using] RG11 and RG6 between 1000 to 2000 MHz over 800' look to be only 6 dB.
    800' RG6 = -52 dB @ 1 GHz to -75 dB @ 2 GHz.
    800' RG11 = 23 dB @ 1 GHz to -34 dB @ 2 GHz.

    "Clearly" RG11 has less loss than RG6, but only using it on the first leg and "doubling" the amps, doesn't sound like the best plan.
     
  8. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    What RG11 would do for this layout:

    SWM8 output = -30 dBmV
    200' RG11 ~ -13 dB [-43 dBmV]
    Sonora amp = +14dB [-29 dBmV]
    200' RG11 ~ -13 dB [-42 dBmV]
    amp = + 14 dB [-28 dBmV]
    200' RG11 ~ -13 dB [-41 dBmV]
    amp = +14 dB [-27 dBmV]
    200' RG11 ~ -13 dB. [-40 dBmV]
    amp = +14 dB [-26 dBmV]
    8-way splitter ~ -12 dB. === -38 dBmV
    Receiver min level ~ -55 dBmV.
    As this shows, each amp compansates for the loss of 200' of RG11.
    No more amps are needed and one would be moved down stream from where two are doubled up. Without slop compansation, levels would be ~ -21 dBmV @ 1 GHz to -38 dBmV @ 2 GHz at each receiver.

    RG11 looks like the best "fix".
     
  9. eakes

    eakes Godfather

    350
    1
    Sep 21, 2007
    I would agree that replacing RG6 with RG11 is the best fix. However, two additional amps (same kind as existing) should make it work. Add one at the end of the first 200 feet (total of two amps) and add one just ahead of the splitter.
     
  10. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    Were you in the same lecture as I was? :lol:
     
  11. houskamp

    houskamp New Member

    8,636
    3
    Sep 14, 2006
    Without all the technical stuff..
    Take one reciever and start going back on the line to see where it quits at.. you might just have a bad amp somewhere..
     
  12. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    That's too easy, but there does seem to be more loss than amplification [needed] in this system.
     
  13. houskamp

    houskamp New Member

    8,636
    3
    Sep 14, 2006
    Just my "practical" side.. Wouldn't surprize me to find that someone pulled a plug or something stupid.. seen it many times before, especialy in one that has so many points to have it happen..
    BTW, I have 100' rg6 to PI and then 150' rg59 to my upstairs reciever and it still works fine. and IIRC someone tested 400' once too..
     
  14. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    "I think" it was Carl6 and was "something like" 360' [barrelled all together too :lol: ]

    "Also" Carl used no splitters and was right on the limit of cable loss from the -30 dBmV output.
     
  15. doctor j

    doctor j Hall Of Fame DBSTalk Club

    1,706
    50
    Jun 14, 2006
    Birmingham, Al.
  16. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
  17. doctor j

    doctor j Hall Of Fame DBSTalk Club

    1,706
    50
    Jun 14, 2006
    Birmingham, Al.
    Ah, I didn't go to older post.

    But as above they also have 20 and 28 dB gain amps if more "umpgh" needed along the way.

    Doctor j
     
  18. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    I think we both could have done a better job than was done here [so far]. ;)

    What some seem to be missing is the SWM aspect, which also seems to limit the amps available, or at least require the SWM8 to be moved way down stream and have the amps feed the SWM8 [which also may be a better way to do this].
     
  19. doctor j

    doctor j Hall Of Fame DBSTalk Club

    1,706
    50
    Jun 14, 2006
    Birmingham, Al.
    If I read your "situation" correctly, you have a residential SWM-LNB 50 to 60 floors up and are trying to run 8 H21 receivers 800 ft away!??

    Had you discussed this with the MDU installers , I believe they would have recommended a 4 (or 5 or 6 with 95 +/or 72.5) line commercial SWM MFH-2 setup. Polarity lockers/amps/ equalizers much more standardized and reception more likely.

    see example:

    http://216.254.116.117/nas/adm/pdf_files/MFH-2 HR MDU.pdf

    http://216.254.116.117/nas/adm/pdf_files/MFH-2 Large Garden Style MDU.pdf

    or

    http://www.sonoradesign.com/images/LA28A_1245_spec.pdf

    Doctor j
     
  20. veryoldschool

    veryoldschool Lifetime Achiever Staff Member Super Moderator DBSTalk Club

    42,679
    348
    Dec 9, 2006
    More info would help but:
    "I have layers of IT mangment asking me why TV isn't working"
    Seems like a company.

    "5 LNB and paired dish AU9-SL5-SWM"
    Hard to know if this is a SWM8 [but he also mentions a 28 volt PI] or a SWMLNB.
     

Share This Page