Long line drop qestion- how much is too much?

[Koby]

Hi, I'm new here and need help from the experts.

I am doing a new install in a very tough location. (I should mention I want to get HD). The only good line of sight I have is approx. 300ft from the house. Direct Tech (local D* installers) came out, said, "No way in hell..." and left. I've also spoke with some private contractors who have told me similar-- that I will get 0 signal over that distance.

I've read enough to know that just because they say it's impossible doesn't mean it can't be done so that leads me to you guys. Is this 300ft too far? I think the LOS is going to be good from that spot, so assuming I have high levels at the dish, will enough signal reach my receiver?

If it's questionable, what gadgets are available to help? Signal boosters? etc...

Any help is appreciated. Thank you! :grin:

[HDTVsportsfan]

Welcome to DBSTalk.

300 Feet is probably to long for a reliable signal, if any. Even with the best cable and connectors. Sonora makes several signal locker products that would probably help w/ this. But I have never used any of there products.

I'm sure some will come along and offer better insight.

[Koby]

Just for laughs, here is a Google Earth screen shot of my little predicament:
http://img187.imageshack.us/img187/3517/dbsta0.th.jpg (http://img187.imageshack.us/my.php?image=dbsta0.jpg)
This is all on a hillside with rises up to the southeast, further robbing me of elevation. Like I said, this is a tough location. Actually it's a dream location for everything else, just a nightmare for dish LOS.

[TimG517]

If it helps any I have probably close to or more than 300 feet of cable running from my dish to my 3rd floor TV's (2 of them).

The dish is outside in front of my townhouse. The cable goes in a window, under the front door and up the first flight of stair. It then goes around a corner, through my kitchen and up the 2nd flight of stairs. Then one cable goes around a corner, through the hallway into my bedroom. From there it goes up the wall and over the two doorways and into my tv.

The other cable goes around the 3rd floor vestibule and into the spare room, up the wall and over the 2 doors around a corner and into the TV.

The install guy said he used more than 500 feet of cable in the initial install (didnt' charge me for that :)

I have zero issues with poor signal, signal loss or anything. My HD DVR box does the same routing, only stops on the 2nd floor.

[veryoldschool]

Just for laughs, here is a Google Earth screen shot of my little predicament:
[/URL]
Note the this is all on a hillside with rises up to the southeast, further robbing me of elevation. Like I said, this is a tough location. I should mention that it is a dream location for everything else, just a nightmare for dish LOS.
A Google Earth can't even show my house, just the trees... :lol:
300' is "do able". Good solid copper core cable is needed. There are two things you need to look at:
#1 The DC power driving the LNBs
#2 The RF signal loss through the RG6

#1 needs the solid copper core cable and can be helped with a Sonora locker: http://www.sonoradesign.com/product_info.php?manufacturers_id=2&products_id=360 (http://img187.imageshack.us/my.php?image=dbsta0.jpg)

#2 may not be a problem, but if it is then 2 of these will work: [URL]http://www.sonoradesign.com/product_info.php?products_id=194&cPath=41&target=homeowners

[doctor j]

To insure success go ahead and buy RG-11 cable.
There is solid copper RG-11. (Belden 7731a)
It's more expensive but 300 to 400 ft could work.
THE KEY as VOS pointed out is1) line voltage loss (if <16 volts LNB's can't switch and 2) signal strength loss (dBm) (power) which can be helped by larger dish if needed.
If problems the line amps VOS referenced will help.
I would not attempt without the sonora polarity locker HRPID1422.

Doctor j

[houskamp]

Just for laughs, here is a Google Earth screen shot of my little predicament:
http://img187.imageshack.us/img187/3517/dbsta0.th.jpg (http://img187.imageshack.us/my.php?image=dbsta0.jpg)
Note the this is all on a hillside with rises up to the southeast, further robbing me of elevation. Like I said, this is a tough location. I should mention that it is a dream location for everything else, just a nightmare for dish LOS.
wow... whats the feild to the north of you? looks like a shorter run (would have to be far enough into it to get LOS)
otherwise a tower of some type might be easier to install.

[HDTVsportsfan]

That tree line might be why he can't go there. Maybe he needs to stay onm the side he is on now. Or it's not his property. :grin:

[Koby]

The field to the north is accessible, but a slightly less desirable because the line would be less out of the way as would be if I run it through the woods. Also the tree line it is quite tall so two locations are more or less equidistant.

Question about HRPID1422: Am I understanding that this would be located about half way along the cable run? If so I would need to run power out to the device, correct?

[joe diamond]

Another route:

Consider a tower......they come in ten ft sections. In some areas you can get one for free if you remove it.

+

Consider cutting (topping) some trees to get a LOS.

I did one (top secret location) where we topped a tree & spiked the mast to the top of the topped tree. It really messed up the tree. If you don't kill the tree you will need to prune an odd limb from the front of the dish. Really high winds mess up the signal.

Use a transit set to the az, and el for your zip code for the 5 lnb dish. See what you get.

Tough one.

Joe

[MIMOTech]

I know how to do it.....Not a problem.....PM me.

[HDTVsportsfan]

Not something you want to share with the rest of the class?

[litzdog911]

...
Question about HRPID1422: Am I understanding that this would be located about half way along the cable run? If so I would need to run power out to the device, correct?

No, the Polarity Locker would be located where the cables enter your house.

[veryoldschool]

No, the Polarity Locker would be located where the cables enter your house.
He snuck that edit in late. :lol:

[K4SMX]

Sonora HRPID1422 (WB68 multi-switch required!) or WB616 powered multi-switch + Belden 1694a RG6: Medium $
Belden 7731a RG11: Bigger $

Both should work, but RG11 will have less RF signal loss. Question is: how many feet of RG6 or RG11 would that be?

If you only have one to four tuners in use (2 tuners per DVR) and you want to try RG6, you can run your tuner lines straight to the dish for an initial test. I'd try a single run on one receiver just to see what happens. It may work. If not, and you then opt for either the Sonora/WB68 combo or the WB616, you must run all 4 lines to the dish from the multi-switch. We're talking 1,200'.

You're asking a lot of the power supply in the receiver to send signal voltages that distance on RG6, and the Sonora/WB68 combo or the WB616 handles the load of the 300' as well as the LNB/internal multi-switch. Personally, I'd use a WB616 powered multi-switch over the Sonora/WB68 combo, if you decide on RG6 and you need the externally powered signal voltages.

Use PVC conduit and not direct burial for either type of cable.

[VaJim]

Buy a chain saw.:grin:

[doctor j]

Reviewed my installer certification manuals.
2 calculations 1) voltage loss 2) signal loss

The manual calculates voltage loss as V=IR
receiver voltage 18.6v may be 19.6v.on some newer models. The Sonora HRPID1422 polarity locker outputs 20v. Minimum voltage 16v to allow switching of odd/even (RHCP/LHCP) polarization channels.

therefore 2.6 to 4.0 volts to play with.
Belden 7731a (RG-11)specs as only 1.5 ohms/1000 ft. 1694a (RG-6) at 2.8 ohms/1000 ft.
Manuals gives #'s as RG-6 CCS (copper coated steeel)-4.5 ohms/100 ft ; RG-6 Solid Copper 1.6 ohms/100ft ; RG-11CCS at 1.5ohms/100 ft . No mention of solid copper RG-11.

750 milliamp power draw of Ka lnb.

Thus 4 volt drop at 1.6 ohm/100 ft and 750 milliamp is 1.2 volt per 100 ft or 330 ft.
Belden #'s give 3500 ft for 7731a and 1900 ft for 1694a!!!

dBm signal loss (more at higher freq.)(Ka high is 1650 to 2150MHz)per manual is 9.91 dBm /100 ft at 2250 MHz for RG-6 and 5.21 dBm at 2250 MHz for RG-11.
Belden specs are 9.1 dBm/100 ft at 2250 MHz for 1694a (RG-6 solid copper) and 6.9dBm /100ft for 7731a (RG-11 solid copper).

Phase Four (AT-9 or AU-9) dishes are conservatively rated as -25 dBm power reception. (I've seen several readings in the -17 to -20 dBm range). The receivers need a -55dBm input for reliable reception. Thus about 30 dBm signal loss to play with.

With above losses with the most conservative readings gives 300 ft for RG-6 or 430 ft for RG-11 without signal amps needed.

I know this is more than you ever wanted to know but it does explain why high quality cable is helpful and why depending on the receivers to power the LNB ( maybe only 2.6 volt drop to failure) vs the Sonora locker (4.0 volt drop) is a bad idea.

If you've read to this point you are a true Directv Geek!!

Doctor j

[Tom Robertson]

....

If you've read to this point you are a true Directv Geek!!

Doctor j
Guilty as charged. :)

[veryoldschool]

If you've read to this point you are a true Directv Geek!!

Doctor j
Or just have an understanding of what you're posting.
750 milliamps may be a bit too much as most specs are for 500 milliamps.

[houskamp]

realy sad (geek) part is I still remember those formulas.. (dad was an EE) :lol:

[joe diamond]

doctor J,

Thanks for the grind out. I have been using 250 ft for RG 6 max runs only because the dual cable with ground comes in 500 ft boxes and the single comes in 1000ft rolls. I just didn't want to get into underground splices or above ground splice boxes. Now I have little feel for the EE part.

Could you suggest a few texts on the satellite world with emphasis on
SMAT / MDU design & installation?

Thanks,

Joe

[doctor j]

doctor J,



Could you suggest a few texts on the satellite world with emphasis on
SMAT / MDU design & installation?

Thanks,

Joe


Not any readily available info.
What I have comes from the SBCA satellite installer courses for residential, commercial/SMATV and MFH1/MFH2 systems.

Some info/diagrams at the NAS site

http://www.nasproducts.com/
look under applications

Some basic stuff at Hometech

http://www.hometech.com/learn/index.html


By the way VOS the instructors are specking the Ka LNB's at 750 mAmp.
I have not made any direct measurements, just using those #'s for design drawings.
Designs from Directv are usually very conservative.

Doctor j

[Koby]

Wow, great info everyone!! I really appreciate the help! I have read every word, and even understood some of them too ;)

Here's a thought... what if I use RG-11 for the initial run down from the dish into HRPID1422, then run RG6 to the actual receivers. This would minimize the loss over the long distance, and then on the other side of the splitter I will have the advantage of the more flexible cable which will make for an easier time routing it.

BTW, as mentioned above, I only need 4 inputs: 2 dual turner DVRs.

[Brian Hanasky]

Buy a chain saw.:grin:




:lol: You'll never get an invite to join Greenpeace....

just kidding that was funny.

[K4SMX]

Wow, great info everyone!! I really appreciate the help! I have read every word, and even understood some of them too ;)

Here's a thought... what if I use RG-11 for the initial run down from the dish into HRPID1422, then run RG6 to the actual receivers. This would minimize the loss over the long distance, and then on the other side of the splitter I will have the advantage of the more flexible cable which will make for an easier time routing it.

BTW, as mentioned above, I only need 4 inputs: 2 dual turner DVRs.
I was going to suggest transitioning to RG6 once you get to the house. There are adapters/connectors for this (http://www.solidsignal.com/cat_display.asp?main_cat=07&CAT=Connectors). The "gold-plated" installation would be 4 runs of Belden 7731a RG11 from the dish to the house, a WB616 powered multi-switch (http://www.solidsignal.com/prod_display.asp?prod=WB616), and Belden 1694a (https://www.tselectronic.com/belden/1694a.html?tse_Session=2a6a631f756e811ae63568006d9 6953f) around the house. You could probably hold off on the WB616 until you would ever need additional receiver connections, but it would provide a solid DC power source in the meantime.

[veryoldschool]

By the way VOS the instructors are specking the Ka LNB's at 750 mAmp.
I have not made any direct measurements, just using those #'s for design drawings.
Designs from Directv are usually very conservative.
Doctor j
It's always good to have "headroom".
Spaun is going to be in trouble if 750 mAmp is the draw, since they spec their products for 450 mAmp.
Regardless, it is never good to design a system at the limit to start with. :)

[AntAltMike]

Here's something I mentioned in a previous thread which may be of interest to those running long coaxes,

A few months ago, I installed a system using what I will call a conventional, 13v-18v/22 KHz tone switch to support three H20 receivers. Within the next week, the receivers got software updates but then could not complete the reinitialization. After hacking around with these receivers for half an hour, I went back through the "guided set-up" and noticed that one step had me choose whether there was a 4x4 switch or a 6x8 switch, and these receivers were all in the default setting of 6x8. When I changed them to 4x4, they all were able to function again.

I am guessing that those receivers had been given software that used pure DiSEqC commands for selecting its inputs when set for 6x8, as an alternative to some hybrid combination of voltage/22KHz, with DiSEqC that might alternatively be used for accessing "flexiports, but if the set-up option of 6x8 is still a set-up menu choice, and if an AU9 LNB can respond to DiSEqC commands, then one might be able to support longer coax runs by setting up such receivers in 6x8 mode, as the 18 volts switching threshold of approximately 15.5VDC might not have to be exceeded. Unfortunately, I do not have the means (or the time) to explore this possibility further.

[veryoldschool]

Wow, great info everyone!! I really appreciate the help! I have read every word, and even understood some of them too ;)

Here's a thought... what if I use RG-11 for the initial run down from the dish into HRPID1422, then run RG6 to the actual receivers. This would minimize the loss over the long distance, and then on the other side of the splitter I will have the advantage of the more flexible cable which will make for an easier time routing it.

BTW, as mentioned above, I only need 4 inputs: 2 dual turner DVRs.
This is going to be a significant project and should have some flexibility.
1) RG-11 from the dish to the house [in conduit]. This might need 2- 1,000' rolls.
2) 300-400' 1 1/2" [or larger] PVC conduit.
3) Sonora locker [HRPID1422].
4) 2 Sonora amps [optional] to add headroom for RF loss or rainfade.
5) Zinwell WB68 multi-switch.
6) RG-6 for everything from the Sonora locker to the receivers.

Now let's look at another option:
Dish mounted SWM8
1- RG-11 cable to the house.
SWM power inserter
Sonora SWM amp: http://www.sonoradesign.com/product_info.php?products_id=331&cPath=42&target=homeowners
RG-6 with "simple" splitter to both DVRs.
This would be limited to 4 DVRs [or eight tuners total], but if the RG-11 is run through conduit, a second run and SWM8 could be added for another 8 tuners.

[joe diamond]

It's always good to have "headroom".
Spaun is going to be in trouble if 750 mAmp is the draw, since they spec their products for 450 mAmp.
Regardless, it is never good to design a system at the limit to start with. :)
Very old,
I had this conversation with my son recently. I noted that engineers of all persuasions would build in a service factor to protect from understating a design capability. He used the patient but superior voice to explain.....well, dad, you know a lot of early work was done on slide rules...these were and are just a close guess at results so it is a good practice to build some "headroom."

Now that they use computers will end users be expected to use equipment only to the design limits?

Joe

[AntAltMike]

Now let's look at another option:
Dish mounted SWM8
1- RG-11 cable to the house.
SWM power inserter.
Can the SWM be powered by its stacked lead? If not, then he'd need a second coax for powering.

[veryoldschool]

Now that they use computers will end users be expected to use equipment only to the design limits?
Joe
Some think the sign of a good engineer is to design things to the minimum. I remember one with a fiber optic task. She spent months working it out and submitted her design, only to be asked to remove a [99 cent] transistor. Was she ever pissed [frustrated].
"Headroom" is a luxury these days.
If your market will pay for it, then you can have it. If price is the major factor, then you can't.
This all comes down to the constant fight between the engineering and marketing departments and then which management supports [more].

[veryoldschool]

Can the SWM be powered by its stacked lead? If not, then he'd need a second coax for powering.
Yes, FTM #1 [or I would have added a power cable too]

[AntAltMike]

DBS satellite system design needs headroom because the unique installations are done without electronic measuring equipment, and because the company that furnishes the hardware does so in anticipation of it generating a stream of revenue that it can't generate if it doesn't work, and it is also on the hook for the expense of making the system work if it doesn't

[joe diamond]

DBS satellite system design needs headroom because the unique installations are done without electronic measuring equipment, and because the company that furnishes the hardware does so in anticipation of it generating a stream of revenue that it can't generate if it doesn't work, and it is also on the hook for the expense of making the system work if it doesn't
Yep,
That is another aspect of the head room overbuild.An exact tested specification would allow a little planning but, as you say......a little extra capacity won't hurt. That hook can rip ya!

If it is not there trucks roll and money flies.

An extension I have seen is intentional OEM..original equipment manufacture components. A mechanical part that looks good but is under built to get the unit sold low and out the door soon. When the unit blows the replacement part is a well engineered and very expensive replacement.
I watch as the quality of the 18"dishes comes down. The Ka/Ku seems good but kicked out the door. (tested in the field)
I think a cheaper Slimline without the dibble system could be built with a better (much cheaper) test meter.

Joe

[1948GG]

Okay, folks have gone about the distance with various ways to make it work...

Without asking one of the most basic questions I would ask: Is there any AC power available around/near the proposed dish site?

Either the WB616 (powered multiswitch) or the SWM8, are, 'power wise', simply voltage/current 'sense' switchers. They don't require anything more than microwatts of current drain to 'know' what to do. I haven't had enough time yet to really put the SWM8 'through the paces' as of yet, I might know more in a few weeks. But I have put the WB616 through all kinds of yanking, and have several customers out to 400+ feet distance, on RG6 (buriable) cable. And, one really only needs FOUR cables to 'daisy-feed' to another WB616 at the house (a WB68 probably would work, but I avoid those like the plague), to feed 16 tuners from that point.

The WB616 powered either right at the dish (AC power available from the remote garage where the dish is at), or by RG6 coaxial power feed from <100' away down the 'driveway' (i.e., AC powered lamps along the road up to the home). This easily powers the WB616, and would do likewise with the SWM8 (but simply that part wasn't available until recently).

The point is, that the current drop is, with the utilization of the WB616, extremely minimal. All the box needs to do is be able to, again, 'sense' the 13v/18v and 22Khz, and to do so (since it's a current loop device) is a couple of microamps at best, NOT the huge (in comparison) current either the dish electronics or a non-power switch like the WB68 needs. The 'power' for the dish comes from the WB616, NOT the receivers.

So, with that basic design, the actual distance from the home is not the big deal it once was. No need for 'heavy' coaxial like RG11 or better, because the current flow (again, microwatts) or the 22Khz tones (again, not much there current wise either), because all the switch (WB616) needs to do is 'sense' it.

For anyone who's done work with industrial electronics, particularly instrumentation and control systems, the 'analog' is a solid state relay system versus the old-style mechanical relay controls of before 1970's or thereabouts. They used several milliamps of power (usually) to operate, just like the 'standard' DirecTV receiver/tuners and the dish electronics need. But put in a 'remote power' feed, like the WB616, and bingo. The power needs are from the WB616 to the dish (a few feet?) are met, and between the WB616 and the receivers, almost nothing.

Got it?

How far? I've tested out to 1000' of RG6 on my 'test bench', from the DirecTV receivers (both H20, HR20, and older SD stuff), connected to the WB616.

Works like a champ. Amount of current flow between the two? Almost too low to measure (microamps).

But I doubt that DirecTV would call any such installation 'basic' by any means.

By the way, even if there isn't any AC power plugs available around the dish site, I'll bet that it's a LOT cheaper to run an AC power feed the distance rather than the RG11+. The power needs of the small 'brick' ac/dc power module of the WB616 is 1Amp @120v. That's pretty small pickings.

[veryoldschool]

Got it?

How far? I've tested out to 1000' of RG6 on my 'test bench', from the DirecTV receivers (both H20, HR20, and older SD stuff), connected to the WB616.

Works like a champ. Amount of current flow between the two? Almost too low to measure (microamps).

But I doubt that DirecTV would call any such installation 'basic' by any means.

By the way, even if there isn't any AC power plugs available around the dish site, I'll bet that it's a LOT cheaper to run an AC power feed the distance rather than the RG11+. The power needs of the small 'brick' ac/dc power module of the WB616 is 1Amp @120v. That's pretty small pickings.
You given some good information, but it's all related to the WB616 to the receiver side, which I doubt is where the issue is with this install.
The SWM is limited to around 200' [output without amp] and requires two-way communication that the WB616 doesn't.
Running 120 volts out [300-400'] to the dish, might be an option, but compared to one run of RG-11, might be more costly along with all of the RG-6 runs needed from the WB616.
The SWM is designed to be mounted out near the dish, can be fed power through a common coax, and the range can be extended with the Sonora amp I linked to.
As with any project like this, all the options should be evaluated.

[1948GG]

You given some good information, but it's all related to the WB616 to the receiver side, which I doubt is where the issue is with this install.
The SWM is limited to around 200' [output without amp] and requires two-way communication that the WB616 doesn't.


Well, the '2-way' communications with the WB616 is voltage/tone, whereas the SWM8 is some actual 'data' talking <40Mhz, like a 2-way cable system. But, in reading the 200' 'specs' of the SWM8, it looks to me more like a current/distance problem with the power inserter to the box, and if that (like I said, I need some more weeks of beating on the thing) if so, that can be gotten around by powering it through the one of the legacy ports, if not through the actual SWM port. But I'll need to do some further testing on that portion of the 'puzzle'.



Running 120 volts out [300-400'] to the dish, might be an option, but compared to one run of RG-11, might be more costly along with all of the RG-6 runs needed from the WB616.
The SWM is designed to be mounted out near the dish, can be fed power through a common coax, and the range can be extended with the Sonora amp I linked to.
As with any project like this, all the options should be evaluated.

It all depends on how many receivers are going to be in the home. I guess that if only one, or if the 200' limit on the SWM is 'variable' (with of course only SWM capable receivers, plus only up to the 8 tuners possible), then it may get interesting as to which way is best. Luckily, like I pointed out, the systems I've designed all had some AC fairly close, or the dish was actually put on top of the garage. Plus the number of receivers in the home were rather a lot (and a mixture of old and new types).

FYI, I've been all over the 'wilds' of New Jersey, several years ago, hauling a Ku-band uplink, doing scientific/geological studies. Some really nice areas, very wooded. Most tops of hills have cellphone towers and such.

But it comes down to the local 'lay of the land'.

[AntAltMike]

For residential, self-installations by people who aren't all that concerned with the possibility of electrocuting themselves or others, it may be cheaper to run the 110 VAC to the powered, WB616 in the field, but for the so-called professional installer, he won't touch that project with a ten foot pole.

If someone wanted to power a DC RF signal amplification, processing, or switching device through the coax over that length with minimal shock hazard, he could put 60 VAC on the line like the cable company does and jury-rig something to convert that to 24VDC or whatever the SWM needs at the dish.

[AntAltMike]

...I've tested out to 1000' of RG6 on my 'test bench', from the DirecTV receivers (both H20, HR20, and older SD stuff), connected to the WB616.

Works like a champ. Amount of current flow between the two? Almost too low to measure (microamps).

You've gotta be losing at least 8dB per 100 feet at 2,150 MHz. What are you measuring for signal strength levels of your highest frequency transponders, going into and coming out of your 1000' test coax?

[Koby]

Thank you for the latest ideas. There is no AC in the vicinity, although I am friends with an electrician, so doing a run would be a possibility.

As an alternative, what about a small solar panel and battery (http://www.segogto.com/power.html)setup, similar to what you'd find with the remote power source for an electric gate opener?

Thoughts? I kind of like the James Bond aspect to this....

[veryoldschool]

Thank you for the latest ideas. There is no AC in the vicinity, although I am friends with an electrician, so doing a run would be a possibility.

As an alternative, what about a small solar panel and battery (http://www.segogto.com/power.html)setup, similar to what you'd find with the remote power source for an electric gate opener?

Thoughts? I kind of like the James Bond aspect to this....
I think you'd run out of power too many times.
Not to keep pushing "my idea", but I do think the SWM is the way to go. Add up the costs and see how it stands up to the other options.
The SWM output frequencies are closer to 1 GHz than 2 GHz and you may be able to use RG-6.
SWM $300
Amp $50
RG-6 $300?

[Hansen]

I think you'd run out of power too many times.
Not to keep pushing "my idea", but I do think the SWM is the way to go. Add up the costs and see how it stands up to the other options.
The SWM output frequencies are closer to 1 GHz than 2 GHz and you may be able to use RG-6.
SWM $300
Amp $50
RG-6 $300?



I agree with VOS. This a very good solution. Keeping it simple will avoid problems and frustrations down the road. Don't forget to add the cost of pvc conduit but that should be nominal.

[K4SMX]

I agree with VOS. This a very good solution. Keeping it simple will avoid problems and frustrations down the road. Don't forget to add the cost of pvc conduit but that should be nominal.
Agree, and the conduit required for one run of RG11 would be a lot smaller than for four runs. Major improvement! :)

[veryoldschool]

Agree, and the conduit required for one run of RG11 would be a lot smaller than for four runs. Major improvement! :)
I think the SWM amp [14 dB] would be a substitute for the need for RG-11.

[AntAltMike]

The SWM output frequencies are closer to 1 GHz than 2 GHz and you may be able to use RG-6.
Someone published the center frequencies of the nine SWM intermediate frequency channels. I think they were spaced at 102 MHz apart, beginning at 974 MHZ, and the highest was 1790 MHz.

I think the SWM amp [14 dB] would be a substitute for the need for RG-11.

He might not need the RG-11 to reduce signal loss, but his DC sufficiency will be greatly enhanced by larger, lower resistance center conductor.

The booster amp, if needed, should be placed at the residence, so that the SWM's power inserter is not supplying power to it. Whether it is needed will depend on how much more loss is incurred in the home's distribution system.

[K4SMX]

I think the SWM amp [14 dB] would be a substitute for the need for RG-11.
Even smaller conduit! :lol:

[veryoldschool]

He might not need the RG-11 to reduce signal loss, but his DC sufficiency will be greatly enhanced by larger, lower resistance center conductor.
From Doctor J: "Belden 7731a (RG-11)specs as only 1.5 ohms/1000 ft. 1694a (RG-6) at 2.8 ohms/1000 ft."
I don't see any DC issues with RG-6 at 400' and the cost of RG-11 isn't warranted, when a $50 amp would work.

[AntAltMike]

I think those resistance figures wer per 100 feet, not 1000 feet.

I know coax prices have gone up lately, but last time I bought RG-11 I paid less than $100 for a 1000' spool, and you can probably get a half spool on eBay for half that price.

[veryoldschool]

I think those resistance figures wer per 100 feet, not 1000 feet.

I know coax prices have gone up lately, but last time I bought RG-11 I paid less than $100 for a 1000' spool, and you can probably get a half spool on eBay for half that price.
This may not be the best price but it must have been a long time ago that you priced it: http://shop.willyselectronics.com/browse.cfm/4,9312.html

Solid copper core: Solid Cu 6.6 ohms/1k ft, so at 400' = 2.64 ohms, which still isn't a problem.

Also the SWM was designed to work with RG-59 [but not at these lengths]

[AntAltMike]

(Note: I have temporarily deleted the calculations that were in this post because I don't have immediate access to the resistance figures that I thought I could pull up, and have to hit the road before I can further verify them)

It is impossible to figure this stuff out precisely, because we don't know the internal resistance of the power source or how dynamic the load might be, but assuming that an SWM really sucks 750ma and stabilizes what is probably an unregulated power supply at 24 volts, then the impedance of the load is 32 ohms.

I don't know at what voltage threshold an SWM might act unreliably, but if you put the booster amp anywhere between the SWM and the SWM's power inserter, it will be a parallel load that will make the voltage drop even greater.

[AntAltMike]

I just called my local distrib, and he has 1000' of regular RG-11 for $175 and Direct Burial for $225, and he said those prices are higher than his catalog prices.

eBay has five auctions for direct burial RG-11. One is for a 900' roll of Belden and is at $19.99, and four others for 1000' reels range from current prices of $25 to $49.99 with one buy-it-now price of $75.

[veryoldschool]

I just called my local distrib, and he has has 1000' of regular RG-11 for $175 and Direct Burial for $225, and he said those prices are higher than his catalog prices.

eBay has five auctions for direct burial RG-11. One is for a 900' roll of Belden and is at $19.99, and four others range from current prices of $25 to $49.99 with one buy-it-now price of $75.
At those prices, it makes more sense that the SWM amp.
BTW: The SWM power inserter= 29 volts, 1.24 amps max, 36 watts max.
I don't know where you came up with your RG-6 specs, since this is what is listed here: http://www.extron.com/download/files/specs/RG6__SHR_041305.pdf
Solid Cu 6.6 ohms/1k ft, so at 400' = 2.64 ohms.
So the voltage drop through RG-6 would be 3.25 volts, leaving 25+ volts to the SWM.

[K4SMX]

Need to make sure we're talking Belden 1694a and 7731a.

Tri-State Electronics (https://www.tselectronic.com/belden/1694a.html?tse_Session=2a6a631f756e811ae63568006d9 6953f) for 1694a: 500' = $189, 1,000' = $350

[houskamp]

My head hurts from all the math :lol:
But I do remember someone (Carl6?) grabbing every piece of cable he had and still having it work.. seems he had well over the 400' mark on his SWM..

[doctor j]

The numbers I posted were per 1000 ft.
BUT retail Belden 7731a with extremely low losses is about $1.00 per FOOT.
You get what you pay for!!!!
I personally think 400 ft can be done with several of the ideas posted with more standard cable.
My favorite is the Belden 1694a which is pricey but will spec out to meet the tolerances we've been discussing.

Doctor j

[veryoldschool]

My head hurts from all the math :lol:
But I do remember someone (Carl6?) grabbing every piece of cable he had and still having it work.. seems he had well over the 400' mark on his SWM..
Tell me about it. RG6 is spec'd in feet and RG11 in kilometers :confused:

[veryoldschool]

Googling RG11 brings you back to this thread. :lol:

[AntAltMike]

Googling RG11 brings you back to this thread. :lol:

I remember Google-searching for some minutae on stacking and filtering, and discovering that I've "contributed" more commentary on these subjects than the other 5 billion people in the world, combined.

I think that some of the attestations posted here regarding SWMs working over long distances were of RF distances, but those SWMs were powered by separate, shorter coaxes.

[RobertE]

Tossing my $0.02 into the pot.

I'd go the four lines route from the dish to the house with the appropriate lockers/amps/switches. Having a SWM way out at the dish just bugs me for some reason. Feel like there will be problems down the road with that route.

[veryoldschool]

Tossing my $0.02 into the pot.

I'd go the four lines route from the dish to the house with the appropriate lockers/amps/switches. Having a SWM way out at the dish just bugs me for some reason. Feel like there will be problems down the road with that route.
Your two cents is worth more than that to me, but the SWM was designed to be dish mounted and soon there will be the SWM dish.

[veryoldschool]

I think that some of the attestations posted here regarding SWMs working over long distances were of RF distances, but those SWMs were powered by separate, shorter coaxes.
As part of a SWM field test group, while I may not know all, I've followed more than most about it.
"We've" excessed the recommended lengths by 200% for every distance and not had problems. The DC drop over 400' will still have enough voltage to power the SWM with good RG-6. After all it was designed to work with RG59.
You have the test equipment, get a SWM and measure it, then you'll know.

[RobertE]

Your two cents is worth more than that to me, but the SWM was designed to be dish mounted and soon there will be the SWM dish.

It's more so the distance that has me concerned along with the limited expandabilty of the SWM at the dish.

[veryoldschool]

It's more so the distance that has me concerned along with the limited expandabilty of the SWM at the dish.
Well you've "got me there". If more than eight tuners [four DVRs] would be needed or wanted in the future, then "a" SWM8 would be a poor investment, and four runs to a Spaun 616 [which has a slope compensating amp] would start making sense.

[jaguar325]

Although not as severe, I am surrounded by trees too.. there is a 10 foot perimeter cleared around my house and that's it. With the help of people on this site, I have figured out a way to find the one spot on my roof that just barely clears the tree tops. It's the highest point on the peak the furthest in the opposite direction (NE) from where the signal is coming in (SW). The line of trees I have to clear is about 25 degrees higher than the roof. If you haven't done so already, I would encourage you to see if there is a way to make it work on your roof. Also, I've got neighbors that have even mounted theirs on thick tree trunks (you don't want anything that will sway in the wind). The only down side to having just barely enough clearance is that every couple of years I have to raise my dish up about a foot to keep up with the tree tops (they're currently at about 60 feet). Mine are pines so they're in the way year-round. Towers can be a good idea if you don't have covenants that prevent you from being able to use them.

If you can't find a spot on the roof, is there some sort of powered "repeater" available? I assume apartment buildings have to figure this kind of stuff out. For what it's worth, I've probably got 125 feet of RG6 in my runs.. no issues.

Good luck.. I love wooded lots except when they stand between me and my HD.

p.s. As a last resort, there is always one other option:

[veryoldschool]

If you can't find a spot on the roof, is there some sort of powered "repeater" available? I assume apartment buildings have to figure this kind of stuff out.
That's what most of this thread is about: RF amps, good cables and DC power, all to overcome the losses due to long cable runs.

[RobertE]

Call this guy to help:

http://www.dbstalk.com/attachment.php?attachmentid=12675&stc=1&d=1204929298

[carl6]

My head hurts from all the math :lol:
But I do remember someone (Carl6?) grabbing every piece of cable he had and still having it work.. seems he had well over the 400' mark on his SWM..

Yeah, I was the one that did that. It was all new condition quad shield copper clad steel RG6 and I had about a dozen barrel connections along the way. It was just over 400 feet that I started having noticeable problems.

Carl

[houskamp]

Yeah, I was the one that did that. It was all new condition quad shield copper clad steel RG6 and I had about a dozen barrel connections along the way. It was just over 400 feet that I started having noticeable problems.

Carl
Well at least part of my mind isn't gone :grin: :lol:

[gmtussb]

I agree with jaguar; try to make it work close to your house.

Again, I am an astronomer and the night sky is very accurate. The attached image is for the exact coordinates that koby posted for his house. It is the Orion constellation with the satellites spread across them. This image is valid for 9:05 PM for koby's house for the next few nights. If he can get an angle over the trees to see this he has LOS.

If you hold a closed fist on your out-stretched arm up to the sky, the distance between your index finger knuckle and you small finger knuckle spans roughly 10 degrees of sky. If you do this with both of your fists together side by side you have the 20 degree span of the satellites in the sky.

[kokishin]

Hi VOS (or anyone),

I have a concern about the SWM-8 input voltage tolerance.
THe SWM-8 voltage input spec is 24VDC.
The PI output voltage spec is 29VDC.

That means the SWM-8 has to have a min input voltage spec of 24VDC+21% and that's not allowing for any deviation in the output voltage of the PI. The recommended 15ft min of RG-6 between the PI and the SWM-8 will not drop the voltage a lick (too much copper). Feels like the PI is not a good match for the SWM-8. I find it interesting that Sonora sells a PI with an output voltage of 24VDC and it's only $30. Seems like a better fit. http://www.sonoradesign.com/images/Sheet_HRPIR.pdf

My questions:
1) At STP, what is the case temp of the PI?
2) At STP, what is the case temp of the SWM-8?
3) Any reports of SWM-8 failures?

Thanks and Regards


At those prices, it makes more sense that the SWM amp.
BTW: The SWM power inserter= 29 volts, 1.24 amps max, 36 watts max.
I don't know where you came up with your RG-6 specs, since this is what is listed here: http://www.extron.com/download/files/specs/RG6__SHR_041305.pdf
Solid Cu 6.6 ohms/1k ft, so at 400' = 2.64 ohms.
So the voltage drop through RG-6 would be 3.25 volts, leaving 25+ volts to the SWM.

[AntAltMike]

Good to see Sonora is using a switching supply, since the dual 15 volt .600 mA wallwarts that powered their big, stacking "Hot Boxes" were junk (and had no-load voltages of about 30 VDC). I bought a dozen laptop switching supplies on eBay for around $20 each to replace them.

Keep in mind that Sonora has been, up to this point, a commercial component supplier, and they surely anticipate that almost all of what they sell will be for MDU systems where the powering cable distances are short. Assuming that the "PI" power supply is also regulated, it may be better suited for residential use because of the voltage drop incurred in remotely powered residential installations. Is it a switching supply?

When JVC was briefly producing commercial components, their 2x16 rack mouinted multiswitch put out over 16 volts on the satellite "odds" input port, and when any customer called to say they were getting the wrong polarity, they would tell them to put in a 100 foot loop of wire, which then screwed up the signal strength differential between the odds and the evens. JVI made the best looking commercial DBS headend equipment. They had larger headed pins and jumper wires for powering (.157 v .100), which was irrelevant since the smaller ones that everyone else uses exceed the needs of these devices by such a large amount that they never fail, and they had the nicest baked finish on their metal cases. But their components failure rates were the highest, and since they exited the business, they have not been missed.

BTW, I just clicked the Extron link with the RG-6 specs and see that the resistance of the center conductor and shield are separate. The shield resistance is 2.4 ohms per 1000 feet, making the total "loop" resistance 9 ohms per 1000 feet.

[carl6]

My questions:
1) At STP, what is the case temp of the PI?
2) At STP, what is the case temp of the SWM-8?
3) Any reports of SWM-8 failures?

Thanks and Regards

I have been running an SWM8 for some months now, one of the first SWM8 testers. It is working just fine.

The SWM8 and the PI are both mildly warm (not hot) to the touch. Previous Terk BMS58 multiswitches ran somewhat hotter than the SWM. Sorry, but I don't have a method of accurately measuring the surface temp.

Carl

[veryoldschool]

Feels like the PI is not a good match for the SWM-8.
When I first heard about the required 15' between the SWM & PI, I suspected this was for matching. I asked the DirecTV engineer about this directly during a chat and he acknowledged this. It has nothing to do with DC voltage. Some have even used less [6'] without a problem.
The PI is labeled at 29 volts, but "my meter" [cheap & very old] measures 25+ volts.

In the MDU setup, the SWM is powered through the #3 legacy port instead of the FTM #1, so it may have a different requirement.

[kokishin]

Impedance matching?

If so, isn't SWM1 and PI out both 75ohms?

How does adding 15' of RG6 which is also 75ohms going to help matching?



When I first heard about the required 15' between the SWM & PI, I suspected this was for matching. I asked the DirecTV engineer about this directly during a chat and he acknowledged this. It has nothing to do with DC voltage. Some have even used less [6'] without a problem.
The PI is labeled at 29 volts, but "my meter" [cheap & very old] measures 25+ volts.

In the MDU setup, the SWM is powered through the #3 legacy port instead of the FTM #1, so it may have a different requirement.

[carl6]

Impedance matching?

If so, isn't SWM in and PI out both 75ohms?

How does adding 15' of RG6 which is also 75ohms going to help matching?

Yes, both are 75 ohms.

I can't offer an explanation, except that I was also told the same thing as VOS by a respected DirecTV source. That is their specific recommendation. Many have varied from that and not had problems.

The best I can suggest is, if you use a shorter length and do experience some type of problem, putting in a 15 foot length as a test probably would be a good first step. It may or may not make a difference.

Carl

[kokishin]

I just want to understand the technical reason behind this. Sometimes "corrective actions" are implemented without knowing the root cause. IOW, someone may have connected a SWM to a PI with 1' of RG6 and had problems. Then they tried 15' and it worked. The word got around and it became the defacto recommendation. But no one really knows why.

Yes, both are 75 ohms.

I can't offer an explanation, except that I was also told the same thing as VOS by a respected DirecTV source. That is their specific recommendation. Many have varied from that and not had problems.

The best I can suggest is, if you use a shorter length and do experience some type of problem, putting in a 15 foot length as a test probably would be a good first step. It may or may not make a difference.

Carl

[AntAltMike]

I just want to understand the technical reason behind this. Sometimes "corrective actions" are implemented without knowing the root cause. IOW, someone may have connected a SWM to a PI with 1' of RG6 and had problems. Then they tried 15' and it worked. The word got around and it became the defacto recommendation. But no one really knows why.

Amen to that. Birdview used to recommend that back-to-back 75 ohm to 300 ohm baluns be placed between their receivers and the stand alone Videocipher descramblers to "improve the impedance match" without anyone really having known why it worked, and when I was in the coin-op games business, you can't imagine the number of Mickey Mouse remedies we did to attempt to reduce the incidence of intermittent failures in the 1980s. And once we became convinced that the remedies did reduce the failures, we implemented them religiously forevermore, even if some hard or soft circuitboard changes were worked into later revisions that probably totally eliminated the earlier problems. I can't begin to count how many 24 pin IC sockets I changed, replacing the ones that contacted the pins on one side with the "better" ones that touched it on two sides, just because that remedy worked on one production run of one popular game that actually had PROMs with pins that tended to tarnish and so did slightly benefit from the redundant contact. Or, you could have just replace the PROMs with ones with better pins.

[veryoldschool]

I just want to understand the technical reason behind this. Sometimes "corrective actions" are implemented without knowing the root cause. IOW, someone may have connected a SWM to a PI with 1' of RG6 and had problems. Then they tried 15' and it worked. The word got around and it became the defacto recommendation. But no one really knows why.
The SWM output is 75 ohm and has some gain [amp] which seems to not like the input impedance of the PI, where the addition of some "line length" will move the match around the smith chart enough to not give the amp any problems.

[doctor j]

Confirmed this AM at kokishin's request that the standard PI for SWM-8 (29 v) connected directly to Legacy Port 3 with only coax /splitters out of SWM-1 & SWM-2 works just fine.

No reboot needed. No channels lost. Sat signals unchanged.

I don't have a clue why NAS spec'ed this SWM-8 at 24 v, and provided the standard residential PI as 29 v.

Of interest I did voltmeter checks at PI output-29.2 v: 6 ft RG-6 - 29.2 v ; 20 ft RG-6 - 29.2 v and for grins a 300ft section of standard cableco ? RG-6 left by cable as temp line to house also an amazing 29.1-29.2 v!

FWIW

Doctor j

[veryoldschool]

Of interest I did voltmeter checks at PI output-29.2 v: 6 ft RG-6 - 29.2 v ; 20 ft RG-6 - 29.2 v and for grins a 300ft section of standard cableco ? RG-6 left by cable as temp line to house also an amazing 29.1-29.2 v!

FWIW

Doctor j
I must ask [to put a nail in this for good]: were you driving the SWM while you made these measurements [ie tapping off to make the measurement]?

[houskamp]

Confirmed this AM at kokishin's request that the standard PI for SWM-8 (29 v) connected directly to Legacy Port 3 with only coax /splitters out of SWM-1 & SWM-2 works just fine.

No reboot needed. No channels lost. Sat signals unchanged.

I don't have a clue why NAS spec'ed this SWM-8 at 24 v, and provided the standard residential PI as 29 v.

Of interest I did voltmeter checks at PI output-29.2 v: 6 ft RG-6 - 29.2 v ; 20 ft RG-6 - 29.2 v and for grins a 300ft section of standard cableco ? RG-6 left by cable as temp line to house also an amazing 29.1-29.2 v!

FWIW

Doctor j
I'd be a little nervous running it that way.. What I would suspect is that the legacy port power capability is designed to be hooked close to the power supply and have a regulated 24v (no line drops to worry about).. the FTM1 line is problably connected to a internal regulator (to drop it to 24v) so that it will be more tolerant of line drop (thus needing higher voltage).. I would be afraid of running it 5-6v over.. It may work but for how long??

[doctor j]

I must ask [to put a nail in this for good]: were you driving the SWM while you made these measurements [ie tapping off to make the measurement]?


No just measuring voltage at end of wire.
Be glad to try another way.
Educate me as to what I'm missing.
I was just surprised to see essentially no voltage drop thru 300+ ft of wire.
This was not "my" wire just Cableco standard issue.
the resistance of the wire itself seems negligible which I thought would have been more of a concern.

Doctor j

[doctor j]

No Voltage on Legacy 3 with PI connected to SWM-1

Doctor j

[veryoldschool]

No just measuring voltage at end of wire.
Be glad to try another way.
Educate me as to what I'm missing.
I was just surprised to see essentially no voltage drop thru 300+ ft of wire.
This was not "my" wire just Cableco standard issue.
the resistance of the wire itself seems negligible which I thought would have been more of a concern.

Doctor j
The voltage drop is because of the current draw [Ohm's law = a 1 volt drop, across a one ohm resistor, with one amp of current flow].
So without "load" what you measured was the standing voltage [wave] and why you didn't see a voltage drop.

[AntAltMike]

No just measuring voltage at end of wire.
Be glad to try another way.
Educate me as to what I'm missing.
I was just surprised to see essentially no voltage drop thru 300+ ft of wire.
This was not "my" wire just Cableco standard issue.
the resistance of the wire itself seems negligible which I thought would have been more of a concern.

Doctor j
What's missing is, the "load". Assuming a perfect (regulated) voltage sourrce, the voltage drop at the load end of the wire results from the ratio of the resistance of the load. If you have a 24 ohm source, and 10 ohms of wire resistance and, say, 50 ohms of load resistance, then the the voltage measured at the load will be 24V times 50/60. But if you measure an open wire, the resistance of the meter is practically infinite, so the no-load voltage will remain the same.

Further complicating the estimate is that the cheap, unregulated "wallwart" power supplies are far from being perfect voltage sources, and while the voltage output at the source end of a switching regulator powered circuit stays the same when loaded, regardless of the load, the output voltage of a wallwart drops precipitously. With an unregulated source, you have to know the internal resistance of the source, which will never be furnished to you with any precision with this grade of equipmet.