Hughesnet Community

Warning, don't fry your modem.

cancel
Showing results for 
Search instead for 
Did you mean: 

Warning, don't fry your modem.

  Some years ago, an EMT vehicle backed over and destroyed my Hughes dish. The next morning I called my trusty tech company, and the first thing he asked me was if the coax was damaged, and next if had I cut the power to the modem.

 

  I didn't discover that the coax was broken until the next morning, and I hadn't powered down the modem for all of that time. He said that with all the rain we had that night, that the modem was probably fried, so he would make sure that the tech he sent brought a new modem with him, and for me to unplug the modem at once, even though it was probably too late by now.

 

  He was right, the modem was fried.

 

  I knew that coax was designed to carry extremely weak signals, and I didn't think there was enough power available to cause any damage. Looking back now, I think it was probably the transmit signal to the dish that probably had enough power to do the damage, and the water on the broken end of the coax caused a short on the transmit circuit of the modem. At least, that's my theory.

 

  I suspect that this may be true for the newer generation modems also. I'm going to play it safe from now on, and if anything ever happens to my dish again, I'm going to immediately cut the power to the modem.


PREJUDICE, n. A vagrant opinion without visible means of support.
The Devil's Dictionary

1 ACCEPTED SOLUTION
BirdDog
Assistant Professor

Yes, there is DC power supplied by the modem going through the coax to power the radio on the dish. Under certain situations the modem can be damaged if the power is shorted out. There are protections built in but they don't always work.

 

Also one reason we say to power down the modem from the wall receptacle, not the plug on the back of the modem.

View solution in original post

27 REPLIES 27
BirdDog
Assistant Professor

Yes, there is DC power supplied by the modem going through the coax to power the radio on the dish. Under certain situations the modem can be damaged if the power is shorted out. There are protections built in but they don't always work.

 

Also one reason we say to power down the modem from the wall receptacle, not the plug on the back of the modem.


@BirdDog wrote:

Yes, there is DC power supplied by the modem going through the coax to power the radio on the dish. Under certain situations the modem can be damaged if the power is shorted out. There are protections built in but they don't always work.

  Given the behavior of my tech guy I would guess that they seldom work, because he was absolutely sure that I was going to need a new modem.

 

  There's obviously a lot more happening on that one piece of coax than I ever thought there would be.   BTW I hate coax.  The first computer network I worked with,  was coax run overhead on already existing poles.  It was an absolute nightmare to keep working.   After we had done about everything that we could think of to try to fix it.  I told my supervisor the only thing we could to do now was to change all the poles.  :>)> 

lets be honest...your modem is supplied power by an AC/DC converter. A cheap one at that. If the modem failed before the converter did...HughesNet needs a new modem manufacturer. Plain, simple, to the point.


@gaines_wright wrote:

  Given the behavior of my tech guy I would guess that they seldom work, because he was absolutely sure that I was going to need a new modem.

People often say that about me when I suggest they buy new stuff...oh...erm...I will just be over here looking at my tractor pics....

                                                                                                                                                 MrB.png

 

GabeU
Distinguished Professor IV

@gaines_wright 

 

I remember a post from a year or two ago in which someone's modem was fried due to corrosion in/on one of the connectors and water getting inside the cable and shorting the system out.  If I remember correctly it was on an older legacy installation.  I think they may have had to replace the radio, as well, though I may be conflating that with something else.  It was most likely from an insufficient amount of dielectric grease being applied to the connector or the installation tech just forgetting to do so.  Zap!  


Ryzen 5 3400G | MSI B450M Pro-M2 MAX | 16GB Corsair Vengeance DDR4 3000 | XPG SX8200 Pro 512GB NVMe | Windows 10 Pro


@GabeU wrote:

@gaines_wright 

 

  It was most likely from an insufficient amount of dielectric grease being applied to the connector or the installation tech just forgetting to do so.  Zap!  

 

Just a tidbit of information, while typical coax connectors are not IP rated, the application of an additional dielectric will change impedence and negatively impact throughput, as the cable, connectors, transceiver, and receiver are all designed to be impedence-matched. Just an interesting thought to consider. Which is more important? Ingress protection, or maintaining transmit power and throughput without any significant loss or gain (depending on how you look at it)


 

GabeU
Distinguished Professor IV


@vladams2015 wrote:

Just a tidbit of information, while typical coax connectors are not IP rated, the application of an additional dielectric will change impedence and negatively impact throughput, as the cable, connectors, transceiver, and receiver are all designed to be impedence-matched. Just an interesting thought to consider. Which is more important? Ingress protection, or maintaining transmit power and throughput without any significant loss or gain (depending on how you look at it)


The connector in question was likely not weather sealed properly, as well.


Ryzen 5 3400G | MSI B450M Pro-M2 MAX | 16GB Corsair Vengeance DDR4 3000 | XPG SX8200 Pro 512GB NVMe | Windows 10 Pro

obviously so, however the point is that a type F connector (while not IP rated per se) are standard IP64 and applying a dielectric is poor practice...but one is left to consider: do we care more about a cable being submersible, or about ensuring an impedence match. 

GabeU
Distinguished Professor IV

I have no idea, but what I, HughesNet, and the installers care about is a proper installation, which requires the proper application of dielectric grease and weathersealing of all outdoor connectors.  


Ryzen 5 3400G | MSI B450M Pro-M2 MAX | 16GB Corsair Vengeance DDR4 3000 | XPG SX8200 Pro 512GB NVMe | Windows 10 Pro


@GabeU wrote:

I have no idea, but what I, HughesNet, and the installers care about is a proper installation, which requires the proper application of dielectric grease and weathersealing of all outdoor connectors.  


  LOL  "I, HughesNet"?  Sort of like "I, Robot"?  :>)>

 

  Seriously, it's been over 30 years ago, but I don't remember doing any of that stuff.   We didn't have any problems with connectors either.   Most of our problems were caused by lightning and ground loop currents.

GabeU
Distinguished Professor IV


@gaines_wright wrote:

@GabeU wrote:

I have no idea, but what I, HughesNet, and the installers care about is a proper installation, which requires the proper application of dielectric grease and weathersealing of all outdoor connectors.  


  LOL  "I, HughesNet"?  Sort of like "I, Robot"?  :>)>   


In all the hubub I missed your reply.  But, yeah, HughesNet became sentient.  Didn't you get the memo?  😛 😛 😛


Ryzen 5 3400G | MSI B450M Pro-M2 MAX | 16GB Corsair Vengeance DDR4 3000 | XPG SX8200 Pro 512GB NVMe | Windows 10 Pro


@GabeU wrote:

@gaines_wright wrote:

@GabeU wrote:

I have no idea, but what I, HughesNet, and the installers care about is a proper installation, which requires the proper application of dielectric grease and weathersealing of all outdoor connectors.  


  LOL  "I, HughesNet"?  Sort of like "I, Robot"?  :>)>   


In all the hubub I missed your reply.  But, yeah, HughesNet became sentient.  Didn't you get the memo?  😛 😛 😛

 

Perhaps not quite sentient, but I do see my fair share of "artificial" intelligence looming

maratsade
Distinguished Professor IV

"I have no idea, but what I, HughesNet, and the installers care about is a proper installation"

 

Translation:

 

"What Hughesnet, the installers, and I care about..."

 

Three entities mentioned.  "I" is normally placed at the end of this type of subject. 


@GabeUquoted
:

LOL  "I, HughesNet"?  Sort of like "I, Robot"?  :>)>   


In all the hubub I missed your reply.  But, yeah, HughesNet became sentient.  Didn't you get the memo?  😛 😛

 

  I'm not surprised.  The URL of their login contain the words  "skylogin", which  makes me think of "skynet".  :>)>  

maratsade
Distinguished Professor IV

You have discovered my secret identity as Skynet. This may not bode well for you.

Signed: I, Skynet.


@maratsade wrote:

You have discovered my secret identity as Skynet. This may not bode well for you.

Signed: I, Skynet.


  ROTFL!  I guess I better get my 1000 ton hydraulic press and my vat of molten steel out of the barn.  :>)>


@vladams2015 wrote:

Just a tidbit of information, while typical coax connectors are not IP rated, the application of an additional dielectric will change impedence and negatively impact throughput, as the cable, connectors, transceiver, and receiver are all designed to be impedence-matched...........

 


   In this case the coax was actually cut in two, so the connector wasn't a factor.

 

  It's been over 30 years siince I worked with coax ( I didn't enjoy the experience ).  I don't remember using any sort of grease on the connectors.   We eventually replaced all  the coax in the plant with fiber optics.  Which is about the only thing that will work in a central Florida phosphate plant   We all celebrated when the last piece of coax was gone. 

 

 


 

 I don't remember using any sort of grease on the connectors.   

 

There is still a going trend to use dielectric grease in outdoor applications of F-type connectors, and for whatever reason some companies insist on it being appropriate. When you run 100 ft of RG6 with a dielectric in the connector on the main connection, and all four connections in the loop, and don't expect significant loss...you are being naive at best. Companies like HughesNet, Viasat/Exede, etc should use LMR400 as a baseline, with N-type connectors, and certainly no dielectric. Techs often overlook (or don't understand) how a dielectric grease will impact impedence in the connector. We rely on internal dielectrics in the cables to maintain a matched impedence and minimize signal loss and transmit power gain...yet we make it common place to impact impedences and increase the loss/gain while using a (sure its solid copper core) cable that has its own inherent signal loss issues regardless of impedence matching.

 

It is truly a shameful practice in the RF world. If HN can't pinpoint problems with my system, I will be building an LMR400 cable from the dish to the modem, removing dielectrics, and changing all connectors to N-type and see what happens. Guarantee my speed will increase and transmit power will decrease. There are a lot of factors at play that should be customizable in these systems, but alas it is a cookie cutter product that is meant to work for Joe in the inner city with a 20 ft cable run, and Susie in the farmland with a 100 ft run. 

Not to mention HN installers do not care about both horizontal and vertical distances between the modem and the dish. Feedback, resonance, and crosstalk is a thing.

 

 


 

   

Interesting.

 

  I worked in a phosphate plant in central Florida ( Which has more lightning than any place in the US ) that was built mostly on reclaimed land from a salt water bay.

 

  The plant had been in operation since 1915, and the combination the salt water, and years of acid spills, turned the whole place into a giant battery.  I used to joke that if you bolted a 4/0 welding cable to a beam at GTSP and then ran it to #4 CAP ( about 3/4 mile away ), you could probably weld with the ground loop current.  BTW the place was well grounded.  I was amazed to see the number of ground rods that were put in when they built a new plant.

 

  When we put in our first Distributed Control System, it was connected with coax as double token ring.  We called it the broken ring.  All of the coax had to be quickly replaced with fiber optics, which added another layer of hardware that could fail, since a FO to coax converter had to be added to each DCS site.  It was still a broken ring, it just wasn't broken as bad.

 

  We all breathed a sigh of relief when the DCS company finally supported ethernet.

 

  Boy, writing this has brought back a lot of old memories, although I worked there for 31 years it's all been 16 years ago.

 

"If men were the automatons that behaviorists claim they are, the behaviorist psychologists could not have invented the amazing nonsense called “behaviorist psychology.” So they are wrong from scratch--as clever and as wrong as phlogiston chemists."

Lazarus Long

 

Fiber and 1000Gb is definitely the way of the future. RF will always have its place, but I would rather see a company like this use quadrax or even pure silver twinax and start utilizing some aerospace tech instead of staying with the antiquated tech they still stick to because it has worked for so long.

I understand the inherent costs, but if you can provide speeds faster than the competitors simply by using different cabling...dudes, why aren't you doing it?