Hughesnet (Venezuela) HN7000 issue

[takoateli]

Guys, The day before yesterday we had a strange new event. We had a big storm pass through and the signal faded out as it always does, and when the storm passed our RX went back up to acceptable levels but connectivity didn't return. We had a TX error number 17 (TX unable to obtain an available transmission rate), and resetting/restarting the modem didn't fix it. I manually forced ranging and that immediately got us going again. Yesterday we had a TX error, unfortunately I didn't grab a screen shot and I don't remember the error. I think the number might have been 10. Forcing the ranging got it going again.

On thing to keep in mind about this system is the antenna was installed more than a year ago, and it was realigned to a different satellite at least 8 months ago, and it worked pretty good up until recently. There was a point where with this setup on this satellite with the antenna pointed just as it is we got super performance. Skype worked great for video calls in both directions. Here's a timeline of what's been going on:

1) For many months, probably at least 6, I've noticed the RX signal fade to the point of losing connectivity on blue sky days with no wind (possible antenna movement) and no obstruction of the antenna. The base is very rigid and the antenna mount is very stout. There is no movement of the antenna that would cause this. But the signal would come back up on it's own and required no intervention on my part. Resetting the modem or forcing ranging didn't speed the recovery. The outage would last between 1-3 minutes. Most users didn't notice unless Skype was running and they'd see they're going on and off line, and the online gamers noticed it.

2) performance has dropped in recent months, but I don't know what our contention ratio is and our provider won't tell us. I can't be sure the degredation in performance is due to our equipment here.

3) It's only in the past few days that we've been having errors and loss of connectivity which requires intervention.

 

[grohgreg]

You're missing the operative word; dielectric. I doubt any of those spreadable products you mentioned are dielectric. If true, they potentially reduce your signal levels - both transmit and receive.

I don't doubt you followed instructions. Problem is you've got the wrong feedarm for that TRIA (or vice versa). Your TRIA is an OSIRIS-TGx, it gets mounted on a round feedarm. That flat feedarm is designed to support a "pure" OSIRIS. The feedarms are of differing lengths, and require different support struts. But understanding the logistics difficulties you must face down there, I understand you must "make do". That's why I pointed you to that General Dynamics document. There may hopefully be enough mounting hole variations that you can adjust to the optimum focal angle and distance.

//greg//

 

[takoateli]

You're missing the operative word; dielectric. I doubt any of those spreadable products you mentioned are dielectric. If true, they potentially reduce your signal levels - both transmit and receive.

I don't doubt you followed instructions. Problem is you've got the wrong feedarm for that TRIA (or vice versa). Your TRIA is an OSIRIS-TGx, it gets mounted on a round feedarm. That flat feedarm is designed to support a "pure" OSIRIS. The feedarms are of differing lengths, and require different support struts. But understanding the logistics difficulties you must face down there, I understand you must "make do". That's why I pointed you to that General Dynamics document. There may hopefully be enough mounting hole variations that you can adjust to the optimum focal angle and distance.

//greg//

Greg, Thanks again! I get it, but I'm not putting any of those products into the connector, on the threads between the male and female connector or any place else inside the connector. Those products I mention are only applied over the connectors once they're seated firmly. So the electrical properties of the products I have at hand are not a factor. Since I don't have dielectric grease I'm not putting something else as a substitute in there. I learned from the Google that dielectric grease is something very special and nothing else you find sitting around the house approximates it.

On the WISPA (wireless ISP group) there was a discussion about the dielectric grease and it's use. Some folks and companies still use it and some don't. It's a good thing, and I'd use it if I had it, but I believe if you seal the connector well on the outside then sealing it on the inside is redundant.

The cables I'm using now came off a satellite internet system installed by the national telco CANTV and the connectors on the ODU were only sealed with self-vulcanizing tape, no dielectric grease or anything else. Not that that means it's the right way to do it, but it's the way things are done here in the jungle.

I got the measurements of my dish and feed horn as per the General Dynamics drawings and I'll compare them in a bit.

Thanks again!!!

Greg

 

[grohgreg]

I understand and accept your logistics difficulties, but I have to drive this point home about protecting connectors. Whereas your approach is well-meaning, it can prevent neither the differing expansion rates nor condensation. Moisture WILL get trapped between your tape/goop and the metal. If you've ever noticed white powder on the center conductor (which carries the signal) and/or the threaded portions (which represents both electrical and signal ground), that's corrosion. Same thing with spark plugs and spark plug boots. Dielectric grease IS intended to be smeared on/in both male and female connectors to produce an anaerobic condition that will not support moisture BUT still pass an unimpeded. If it's not dielectric, it cannot satisfy the latter.

As for the rain thing you just mentioned, that's another symptom of bad transmitter isolation (POL). After losing the signal, the modem uses the System Info to auto-range prior to resuming lock. Specifically it uses dish geoloco and satellite location to calculate range and distance, then sends a test signal to verify the parameters are within a default window of acceptability. When the results fall outside that window, the modem won't lock. Why? To prevent satellite interference. Forced ranging widens that window, and therefore temporarily ignores the parameters. When the POL angle is incorrect, it the window of acceptability is skewed.

As for losing the signal on blue sky days, that's another symptom of inaccurate pointing angles. Geosynchronous satellites are NOT actually stationary in the sky. They're simply orbiting in sync with the earth's rotational speed. But while the earth is rotating, it's also wobbling on it's North/South axis. The result gives the appearance from the ground, that the satellite is moving back and forth. The range of this apparent motion is predictable, it's called "the box". Many satellite companies publish "center of box" (COB) data on their websites. Ideally the dish should be pointed dead center of this box. That way, as the satellite "moves" back and forth, the dish continues to intercept the signal with minimal loss. But if the dish happens to be pointed at one SIDE of the box, it may experience signal loss as the satellite "moves" to the opposite side. 12 hours later the satellite is back on the "strong signal" side of the box. So if what you're seeing is demonstrating a 12/24 hour pattern, you've got a COB issue. Repoint required.

//greg//

 

[takoateli]

I understand and accept your logistics difficulties, but I have to drive this point home about protecting connectors. Whereas your approach is well-meaning, it can prevent neither the differing expansion rates nor condensation. Moisture WILL get trapped between your tape/goop and the metal. If you've ever noticed white powder on the center conductor (which carries the signal) and/or the threaded portions (which represents both electrical and signal ground), that's corrosion. Same thing with spark plugs and spark plug boots. Dielectric grease IS intended to be smeared on/in both male and female connectors to produce an anaerobic condition that will not support moisture BUT still pass an unimpeded. If it's not dielectric, it cannot satisfy the latter.

As for the rain thing you just mentioned, that's another symptom of bad transmitter isolation (POL). After losing the signal, the modem uses the System Info to auto-range prior to resuming lock. Specifically it uses dish geoloco and satellite location to calculate range and distance, then sends a test signal to verify the parameters are within a default window of acceptability. When the results fall outside that window, the modem won't lock. Why? To prevent satellite interference. Forced ranging widens that window, and therefore temporarily ignores the parameters. When the POL angle is incorrect, it the window of acceptability is skewed.

As for losing the signal on blue sky days, that's another symptom of inaccurate pointing angles. Geosynchronous satellites are NOT actually stationary in the sky. They're simply orbiting in sync with the earth's rotational speed. But while the earth is rotating, it's also wobbling on it's North/South axis. The result gives the appearance from the ground, that the satellite is moving back and forth. The range of this apparent motion is predictable, it's called "the box". Many satellite companies publish "center of box" (COB) data on their websites. Ideally the dish should be pointed dead center of this box. That way, as the satellite "moves" back and forth, the dish continues to intercept the signal with minimal loss. But if the dish happens to be pointed at one SIDE of the box, it may experience signal loss as the satellite "moves" to the opposite side. 12 hours later the satellite is back on the "strong signal" side of the box. So if what you're seeing is demonstrating a 12/24 hour pattern, you've got a COB issue. Repoint required.

//greg//

I did notice a pattern to the problems, not always at 12/24 hours, but there was enough of a pattern to it that I noticed!

Thanks!
Greg

 

[takoateli]

I started tweaking the POL just by watching the "stream error rate". I started around 4% and now it's bouncing between .04 and .09. So Greg I think your right on, that the problem is alignment. The other errors have stopped too.

Greg