Satellite Internet Forum

[David]

I have just setup a NewSat 2 way internet system (LinkStar), it came standard with 30ft cables as my signal was a little weak (4 zeros) I was told if I shortened the cables it would work better. This was a big mistake as i cut one cable into 2 and now if I use the short cable on transmit it will not work, luckly I still have one long cable so I use this on TX. My question is why is this so?
I had thought maybe the connectors I put on cut side was a problem but I have since tried 2 others and it still will not work. Even though it works (with one long and one short) I only get 2 zeros and lots of packet loss, I am wondering if maybe I need a long cable for receive as well? Any comments would be appreciated!

David
(P.S I should not have cut one in the first place!)

[ninosat]

it came standard with 30ft cables as my signal was a little weak (4 zeros) I was told if I shortened the cables it would work better. This was a big mistake as i cut one cable into 2 and now if I use the short cable on transmit it will not work, lucikly I still have one long cable so I use this on TX. My question is why is this so?
I had thought maybe the connectors I put on cut side was a problem but I have since tried 2 others and it still will not work. Even though it works (with one long and one short) I only get 2 zeros and lots of packet loss, I am wondering if maybe I need a long cable for receive as well? Any comments would be appreciated!

Signal loss across 30' of quality coaxial cable is negligible from the perspective of catastrophic signal loss. But excess cable is always frowned upon. The cables should have been cut to length as a normal course of the installation. Your problem could have originated in bad cable in the first place, another incorrectly installed connector somewhere, the grounding block, or in the antenna pointing angles themselves.

I suspect the do-it-yourself connectors are currently the problem, causing a cable short between the shielding and the conductor. Also, the precut lengths of cable probably has interior grade connectors on it. Either replace the outdoor ones with exterior grade, or weatherproof the hell outa them. Doesn't sound like you have a cable grounding block either. Once you fix the cable path, we can revisit the issue of signal strength.

I don't personally know what the zeros represent to a LinkStar user, so - if necessary - you could bring up any recurring signal strength issue in the LinkStar discussion board of Satellite Internet forum.

[hulijan]

Be very careful. Do NOT ever cross over or short circuit the TX and RX cables for a LinkStar modem. It will burn out the modem.

The quality of the F connector workmanship is critical. The centre pin should stick out about 2mm and have a nice smoothly rounded end if possible so that it inserts readily into the socket. The outer screen/braid must make good electrical connection to the F connector outer part and there must be no fragments of braid wire causing a short circuit. Use silicone electrical contact grease to liberally lubricate the pin and pack the braid/outer crimp/screw on area. Seal with self amalgamating tape.

A bad connection in the outer of one cable can cause very strange effects.

If you change the length of the transmit cable the hub (or you) will need to change the TX attenuator setting. If the level is too high it overloads the BUC and distorts the transmit signal.

The receive quality should be 0.00001 ideally. 0.001 is not good enough. Try adjusting the pointing by 1/2 turn of a nut. Repeat the process for an hour or so till you are certain you are in the middle. This really matters as the transmit beam (which you can't measure) is narrower than the receive beam and you must get the transmit beam centered perfectly also. It is not easy to do and it is not obvious, until you have had it explained, why such precision is necessary.

If you join two cables end to end using 2 F connectors and a 'bullet', then beware. The quality of the tiny springs inside the 'bullet' is often very poor and the contact pressure is weak - leading to arcing and failure. Avoid joints if at all possible. If really necessary, join the cables alongside one another making sure the centre wires are well twisted together. Wrap the braids together side by side with silicone grease in between and with a length of sturdy copper wire wound round tightly, and almost up to the end of the dielectric. Put electrical silicone grease all over and wrap in plenty of self amalgamating tape. Fix in position pointing the joint upwards inside a strong plastic bottle which open downwards. Don't worry about the slight impedence mismatch.

[David]

Thanks!

If you change the length of the transmit cable the hub (or you) will need to change the TX attenuator setting. If the level is too high it overloads the BUC and distorts the transmit signal. I think this is maybe my issue, as I believe the F connector is OK, is there a method to set this to correct value? Also what is the attenuator is this the power level?

Also do you need to make any changes when altering the receive cable?

I have no joins in cable straight from LNB to LynkStar, I have changed dish angle a few times to try and improve zeros but no real gain from .001. I tried posting Telnet sheet shot not sure how to insert a photo. Does the transmit cable effect zero level? I presume not but was wondering?

David

[hulijan]

If you have cables 30ft long then shortening the length will give no significant benefit. The output level of the noise floor from the LNB is plenty high enough for 30ft cables - unless the cables are faulty. If the signal quality is poor, the dish size is too small for the location, the feed horn/filter has water in it, the dish is distorted or mispointed, wrong focal length setting, beam obstructed, xpol or adjacent satellite interference, water or corrosion in cables, bad joints etc.

If you are getting 0.001 check to see if you have the right carrier. Does tclgets show pcr=true ?

If you are on the right satellite you need to peak up. Adjust by 1/2 turn and determine if you are going in right direction etc. The polarisation angle needs setting accurately as xpol interference can degrade the rx signal - have you set the polarisation correctly ? Is the dish in focus ? Is the dish rim flat to 1mm ?

The TX cable will not affect the receive performance unless the outer conductor of one or other of the cables is partial or completely open circuit. You can set up the system using the receive cable alone. There is some merit in this to avoid spurious transmissions before you are ready.

The hub should give you an initial TX attenuator setting. Something like savebootparms -t -28 is about right.

I have just visited a LinkStar site installed 10 months ago. The cables had no foil braid at all for 20 ft from the antenna - the aluminium film had changed to white powder all along - due to moisture. A 'bullet' joint midway along in the TX cable had arced and overheated on both sides and the centre wires of the cables either side had corroded to nothing. All four F connectors had poor/loose connection of the outers and the centre pins were easily pushed back so they did not insert properly or were too short.

[David]

If you are getting 0.001 check to see if you have the right carrier. Does tclgets show pcr=true ?

Thanks for your help. I only log in at night so sorry for delay. Yes, tclgets shows pcr=true.

System is only new, I doubt any issue with cable corrosion, I do not have spectrum analyzer just a noise sounder it seems to be very accurate when adjusting dish angle but maybe not good enough, I have entered sat long/lat on web calculator and set polar etc to correct levels I have tried a few degrees each way. I am thinking of buying 2 new cables to eliminate possible cable issue. It still seems to work OK even at 2 zeros and lots of RSUncorrectedErrorcount (it caps out at 65535 in about 5 minutes). I am heading north west closer to Darwin (in Alice Springs at the moment) and as NSS6 is over Indian ocean I expect to get a better signal. I am still at a loss as why the cable I cut in half (doesn't matter which one) doesn't work on TX. It does connect but no through put (well very little). Dish size is 120cm plastic by the way.

David

[hulijan]

See the image below for coverage map of NSS6 - Australia beam. You should get about 2 dB improvement in Darwin vs. Alice Springs.

NSS6 Australia Beam

NSS-6AustraliaSpot.jpg (185.33 KiB) Viewed 309 times

If the RSUncorrectedErrorcount caps out at 65535 in about 5 mins that is very bad. You are looking for zero uncorrected errors in 5 minutes.

If you have a thin plastic type dish check its rim is flat using fishing lines up/down and across - the gap should be 1mm max. Severe distortion can occur in the heat with heavy BUC etc. I've seen 4 dB gain reductions occur due to the weight of the BUC etc. If your dish has two little short side struts at the back, push or pull the sides of the dish while tightening the end screws in these short struts. One end of each short strut has an elongated hole. These struts are there to prevent the dish edge being pulled forwards or pushed backwards by the weight of the BUC.

[harri]

In installing Linkstar, there's always specified cable length, which is always 30m. Now if your distance is more than 30m you may have to use an external power supply for your Tx.

But in this case where you got the cable length cut to about 15m, you may have to check the Tx power you configured in the Linkstar Modem, ss this can affect Tx. Another thing is whether you checked the continuity of the cable and you're sure it has no wound.

Then the connector termination can also affect your Rx. If after all these and your Rx is still low, then re-alighn the antenna and do a peak to pole test.
Good Luck
Harri