22KHz signal coupling through dual LNB

[wrdavis]

Since I installed my Sadoun 1.8 polar dish, I find I have a problem that is probably related to my use of Diseqc 4x1 switches, 22kHz switches, or the dual output C Band GeoSat LNB.

My setup.
I have two Pansat 9200HD units, two 1M motorized dishes, and two 3 circular LNB dishes. Both setups are the same: the 1M dish is wired to Diseqc #1, circular LNB Echo 5 to #4, circular LNB Echo 7 to #3, and circular LNB Echo 8 to #2. The selected output from the Diseqc 4x1 switch goes to the 0KHz side of a 2x1 Encoda 22KHz switch. The output from the Encoda switch goes through the motor and back to the Pansat. The two outputs from the dual output C Band LNB go to the 22KHz sides of the two Encoda switches.

The symptom.
If I select the Home channel on Echo 8 in the bedroom, I get 0 quality. If I go to HDTHR on Echo 8, I get 80% quality. If I select the test channel on Echo 8, I get 80%. So everything is working from Echo 8 in the bedroom except the Home channel. If I change the living room unit to the Home channel from a C Band channel, the bedroom works on the Home channel also. If I change the living room unit back to C band channel, the bedroom unit drops to 0% again.

My speculation.
The 22KHz signal from the living room unit is somehow going through its Encoda switch to the one side of the dual output GeoSat LNB. The 22KHz signal is then somehow coupled through the C band LNB and finds its way back to the other 22KHz Encode switch for the bedroom and causes it to malfunction until the living room is switched off from the 22KHz to 0 KHz.

Anyone have a similar problem and what did you do to fix it?

 

[SatelliteAV]

A European SatelliteGuys member was experiencing similar issues with sharing a common dish in his housing unit in a thread last year. We have experienced this problem with many systems over the years. I would suggest that it has nothing to do with back feeding 22KHz or DiSEqC signals.

Try different combinations of the polarity and transponders on the problematic satellites. You night determine that it is a oscillation conflict between the two receivers. See if the conflict exists by substituting another brand of receiver.

Most likely there is a voltage present on ground. Try plugging both receivers into the same power source. Bond the switches and all other electronics to the home's ground. MPEG 2 receivers are notorious for having small voltages present on the ground. While this does not normally cause any issues with single receiver installs, it can cause issues with multi-receiver set-ups.

 

[wrdavis]

Thanks SatelliteAV, gives me something else to try.

I just finished trying a different approach. I switched the channel list on both Pansats to swap the use of 22KHz between Ku & C, and then swapped the coaxes on the 22KHz Encoda switches. Figured if the 22KHz was the culprit, going to the C Band LNB; I'd make it be 0KHz going there instead. And the 22KHz would dead end at each 1M dish diseqc switch.

But now if either Pansat is on 22KHz for Ku, the other Pansat on C Band gets fouled up on C Band.

Running both Pansats from a common outlet was a quick and easy test. Thought maybe either a difference in ground or perhaps a different phase; either way one extension cord could test it. But the results stayed the same.

I don't have a different brand or model receiver to try. If it is a conflict between the two Pansats, what can be done shy of buying a different receiver? Since I didn't have a problem before installing a dual output C Band LNB, what if I had two single output C Band LNBs? One aimed at the center and the second offset to pickup the next satellite two degrees over?

 

[Lak7]

Can you make 1 unit 22KHz "On" for C Band and the other 22KHz "Off" for C Band?

 

[SatelliteAV]

We have had some success using a LNBF by a different manufacture, but in this case your options are very limited for another dual output C-band LNBF.

Separating the C-band connections to individual LNBFs would solve this issue.

To test your suggestion on the 22KHz signal back feeding, do you have a scope or a satellite meter that could indicate indicate the 22KHz/DiSEqC signals? Example: http://satelliteav.com/p/shop/?ws_pid=644&ws_cat=429

 

[Lak7]

Would DiSEqC Switches after the 22KHz "absorb" the 22KHz tone?
You would have a total of 4 DiSEqC Switches.

 

[wrdavis]

Tried Lak7's idea of one receiver each way. Still fouls things up but then when both are on the same satellite instead of opposite ones.

Finished by putting everything back the way it was to start. Actually that failure mode seemed to be the least offensive of the three combinations.

I have a Tektronix scope I could haul up and check for 22KHz coupling through the dual C band LNB. Also have a Digisat II meter, would need to check if it does a 22KHz on/off indication. But like you say, even if I proved that was the problem; there's not much I can do to fix it.

Is there a low block filter that could be put in line at the C band LNB to stop the 22KHz switching signal, yet let the digital video/audio data stream through? Would be an interesting filter, it would still need to let DC through to power the LNB, so I guess it would be a 22KHz notch filter of some sort.

I have a second combo C/Ku LNB (BCS621) that I tried in between other things. Yes with a separate LNB on the C band dish the problem goes away. But with my 2 degree offset idea I'd just have a new problem remembering where to point the dish to receive which satellite for the different rooms. And with the installed LNB's scalar ring in the way, I never managed to get a signal on the second LNB. So I suppose I'd have to chop scalar mounting rings to do a make shift second LNB work. More work than I want to get involved with at the moment.

 

[wrdavis]

Confirmed the Digisat 3 meter does have a 22KHz detect function when connected to the receiver's output.

So did a quick check to see if 22KHz was leaking through the LNB. The Digisat did not detect a 22KHz signal at the disconnected terminal on the C band LNB with the other half using the 22KHz from the Pansat.

Of course the Digisat has some threshold that a 22KHz signal must exceed to be detected by it. Could have a signal below its threshold and still high enough to foul things.

This C Band project is sure making use of functions on my meter I never fiddled with before. LOL

 

[Lak7]

Can you connect the meter "in-line"?
LNB---Meter---Switch

 

[Anole]

I have a Tektronix scope I could haul up and check for 22KHz coupling through the dual C band LNB.

Sounds like a good tool for the task.
Just be careful connecting it - make sure the 'scope ground isn't hot relative to the coax.

If you were to find 22khz, I wonder what a 2x4 multiswitch would do for it?
Certainly never heard anyone complain about this problem when using one.
Might cure your problem (whatever it is), even if it has nothing to do with 22khz.
...and try grounding the switch, too.

(each of the two outputs of your LNBF hooks to a single switch input - you then have four outputs to connect to your receivers)

 

[wrdavis]

Can you connect the meter "in-line"?
LNB---Meter---Switch

Yes, might try that. But looks like the wife wants to watch TV now, not play with TV.

If you were to find 22khz, I wonder what a 2x4 multiswitch would do for it?
Certainly never heard anyone complain about this problem when using one.
Might cure your problem (whatever it is), even if it has nothing to do with 22khz.

Interesting idea, never used one before. So would I leave the existing selection as is and the 2x4 multiswitch is just inserted near the dual LNB?

Like this.

|-----|out #1-------LNBV .... RX1-----to existing 22khz & 1x4 switch bedroom
|
|dual LNB , , , , , , 2x4 multiswitch
|
|-----|out #2-------LNBH ....RX4-----to existing 22khz & 1x4 switch livingrm

 

[Anole]

I have two Pansat 9200HD units, two 1M motorized dishes, and two 3 circular LNB dishes. Both setups are the same: the 1M dish is wired to Diseqc #1, circular LNB Echo 5 to #4, circular LNB Echo 7 to #3, and circular LNB Echo 8 to #2. The selected output from the Diseqc 4x1 switch goes to the 0KHz side of a 2x1 Encoda 22KHz switch. The output from the Encoda switch goes through the motor and back to the Pansat. The two outputs from the dual output C Band LNB go to the 22KHz sides of the two Encoda switches.

Okay, so you have five sources.
Four Ku band LNBs are on a 4x1 diseqc.
That output then goes through a 22khz Ecoda which lets you select those four LNBs or the one C-band LNB.
Somewhat like this hookup in the Switches Simplified FAQ.

Scroll down a little in the FAQ and read about Multiswitches.
There's no picture for hooking up a 2x4, but the words should make it clear. - :up

 

[pedrogarcia]

If you are using DiSEqC 2.0 switches or above they do have bi directional communication. It may be worth changing out one with a 1.x to see if this solves the problem.
Personally I see this as a receiver quirk and as Brian says induced ground voltage on class 2 equipment may well be the fault

 

[wrdavis]

Great information Anole.

I'm going to have to start doing drawings like you, instead of trying to to type a picture using _,-, & | characters.

Looks like two ways of attacking my problem.

First, if it is 22KHz coupling through the LNB and back into the other Pansat's switching. I'm using Ecoda 22KHz required to pass the 22KHz signal to my Diseqc 4x1 switches. So I could put the Traditional 22KHz switch in series with the output from the Ecoda to the C Band LNB. The Traditional ones don't pass the 22KHz signal, so IF that's the problem, should fix it.

Second, try the 2x4 multiswitch.

With shipping being the delay, seems reasonable to buy both, try them and see what works.

Took my scope this morning out to check for 22KHz coupling through the LNB. I see some, but have no idea if its enough to cause the problem. Even in shadow on the roof, the sky was light enough to wash out my photos. So the outdoor ones were enhanced to see the waveform better. All have the same time and voltage scales, AC coupled.

 

[pedrogarcia]

wrdavis, great photos thanks for sharing them. Sorry though, it does not convince me that 22khz is the problem. I look forward to your next steps

 

[wrdavis]

I agree, the signal through the LNB looks pretty small; to cause what I see as interfering from one system to the other.

I've tried running both systems from the same power outlet, and saw no difference.

The 22khz switch and 2x4 multiswitch idea fix is waiting for the mail to arrive in a week or so.

In the mean time I just put both systems on the same channel either Ku or C band to ignore the problem. But if that's the only thing I can do; might as well drop to only one system on c band and put a signal splitter to the other. No guarantee that would work, if somehow the diseqc switch is effected.

 

[SatelliteAV]

Have the switches and dishes been bonded together and grounded?

 

[pedrogarcia]

wrdavis, Brian has a valid point here. Your receivers are Class2 so currently they are grounded only via the coax as far as the metal cased C band LNB which is bonded to the dish but not grounded effectively or to the switches etc.
One other thought if your dual c band cables are cable-tied tight together induced RF on one may affect a weaker signal on the other particularly is the cable dielectric is crushed.

 

[wrdavis]

No they are not ground bonded together. Could try that while awaiting the parts I ordered. Doubt its RF coupling. The signal goes to zero on a wanted signal if the other Pansat has selected the opposite dish. But could try shortening the coax and separating them as well.

But today was spent re-installing s/w and op sys on my wife's computer.

 

[wrdavis]

Status update.

First, tried bonding all the antennas together in a star pattern from the C Band dish. The diseqc switches, 22KHz Ecoda switches, and the C Band LNB were either daisy chained or wired in another star pattern at each antenna, in case coax ground shield was the problem. Result the same (if one receiver is on C Band, and the other receiver is on Ku, the Ku signal goes to zero) .

Second, removed 14-16 feet of parallel coax cables going to the C Band dish, in case RF coupling was the problem. Result the same.

Third, drove an eight ground rod into the ground and grounded the items from step one, in case charge build up was the problem. Result the same.

Still waiting mail delivery for the traditional non 22KHz passing switches and 2x4 multiswitch to see if either corrects the problem. Maybe next week.