359 ft run??

[willager]

I have a Legacy dish with two 301 receivers. When we moved in we placed the dish on the roof next to a working DirecTv dish. All was fine until the leaves started coming out on the trees. DirecTv is still getting a good signal. But the Dish (lower angle) has lost it. The closest I can get a clear view would have about a 350 ft run. I have been told that it's too far a run for RG6, but that it could be done with RG11. My concern is the switching from odd to even, would it work at that distance. Anyone have this long a run?
Thanks
Willager

 

[SimpleSimon]

Maximum allowed with DishPro is 200'. Less with Legacy LNBFs.

Yes, lower-loss cable might help - BUT, you've hit the nail on the head - the voltage loss might kill the switching.

LMR-400 class cable has a center conductor of about 12 gauge, whereas RG-6 is around 20 (much smaller). It just might make the difference. You won't be able to find F-type connectors for cable that size, so you'll need a couple of short 'pigtails' to use as adapters. The pigtails at the Dish end need to be long enough to go through the LNBF mounting arm - there's NO way you'll stuff -400 class cable through there - it's about a half-inch in diameter.

If you get the idea that this is not trivial, you're right. Oh - and the stuff costs 50 cents a foot or more. So, for a dual LNBF you're looking at $350 plus pigtails. And once you've done all that, it still might not work (it'd be an interesting project, tho).

I think a different solution is in order. No way you can put the Dish on the north end of the house on a short mast? That often gives enough room to see over the trees.

 

[Stargazer]

Couldn't a person put an in-line amplifier every so far to help extend the length of the cable? I would have thought that by now they would have some kind of powered in-line amplifier (perhaps run off of a battery or something, better to have an option than none at all even at some expense).

 

[SimpleSimon]

Couldn't a person put an in-line amplifier every so far to help extend the length of the cable? I would have thought that by now they would have some kind of powered in-line amplifier (perhaps run off of a battery or something, better to have an option than none at all even at some expense).

Unfortunately, it can't be just a simple 900-2200MHz amplifier. It would also have to pass 22KHz, DC, and who knows what else back the other way for switch and LNBF control. Not impossible, but also non-trivial.

If such a thing exists, I would expect it to cost at least $200. I base that on the cost of WiFi amplifiers that are somewhat similar (but not able to do this job).

 

[sirlurkalot]

Don't forget the signalling is bi-directional (DC level shifting for the LNB's I think) so an inline amp probably wouldn't work.

What about cascading some DP34's at, say 1/3 and 2/3 along the cable run?

The 1st DP34 would be unconnected as far as receivers go (not sure if that would work or not?) and the second would then be cascaded off the first and connect to the receivers?

Interesting problem!!!

 

[SimpleSimon]

I was thinking about DPxx switches too - but I seem to remember that the spec is for 200' total with the switch somewhere in the middle. Don't know why that would be unless the switch is just that, and has no built-in amplification. If that's the case, it wouldn't help.

 

[A_Pac]

this could be some help

http://www.radio-ware.com/products/techinfo/coaxloss.htm

 

[SimpleSimon]

A_Pac: the table you reference doesn't go up to the necessary 2200MHz, but thanks anyway.

Try this link for a good loss calculator.

Meanwhile, RG-6 will lose about 30dB @ 2200MHz @ 200 feet - so let's use that as the maximum allowable loss.

The -400 series cable I mentioned loses 22dB @ 2200MHz @ 350 feet, so we've solved that problem by throwing piles of money at it. The 12 gauge center conductor of the -400 cable will probably solve the voltage drop problem, but we've still back to the original issue - it would still cost over $350 just to try it.

RG-11 isn't in the calculator, but using your table, it looks like there's so little difference that it just wouldn't be enough.

 

[red hazard]

Quality RG11 will work

If you use a high quality RG11 such as Belden 7731A you will have no problems. The DC resistance for 500 ft is only 2 ohms. At the maximum DishPro current of 750ma that would only be a 1.5 voltage drop. Since DishPro receivers (and powered switches) put out 19 volts, that would leave you with 17.5 working volts. DishPro will work all the way down to 10.5 volts and a 200 ft run of quality RG6 would have greater DC losses. Regarding RF losses, 350 ft of Belden 7731A RG11 would have about 14.5 db of signal attenuation at 2250Mhz. Average quality RG6 would have about the same amount of loss at 200ft; the maximum length of RG6 recommended for DishPro. Of course the cost will make gasoline look cheap!

 

[SimpleSimon]

Unh. I thought the switching voltages were 13 & 18VDC, so I don't see how it can work at 10.5VDC.

that being said, if your other numbers are correct, you're right it should work on the DC side.

As for RF - my previous post gave numbers I got from the calculator linked there.

The table referenced by A_Pac is from a merchant, and it's numbers are a lot lower than Times Microwave's trusty calculator.

The data sheet for LMR-400 matches the calculator, and I trust Times Microwave.

So, we've got too many different numbers for loss factor but we DO agree it'll be big $$$ to do it.

Oh - and TMR DOES have F-connectors for their LMR-400-75.

 

[red hazard]

The LMR-400 has slightly better RF loss figures than the Belden 7731A. Decibels are a logarithmic power ratio. You can't just double the db loss figure for another 100ft of coax to get the correct loss in db. With the posts moving to DishPro characteristics I
presumed that willager was using DishPro. However since he is using legacy he still does not have a problem with the polarity switching voltages since the DC voltage drop will even be less (500ma vs 750ma) which would yield a 1 volt voltage drop which would not adversely affect the 13/18V polarity switching. As I stated earlier 200ft of RG6 has more DC resistance than 500ft of Belden 7731A RG11.

 

[SimpleSimon]

Actually, you CAN add and subtract the loss numbers - at least that's what the pros that spec'd out my WiFi hotspot system were doing - but yes, dB is a logrithmic scale - I don't know more than that.

Yeah, I forgot he was Legacy, myself. That's a weaker signal, so we might need even better cable?

I guess what's needed is for some supplier to take a 500' roll of cable and try it as a lab test. Why 500'? Just for the extra margin - plus someone might have it lying around.

 

[bcshields]

We use RG-11 for commercial installs due to the long lengths. Expensive stuff. I say upgrade to Dishpro LNBF, DPAdapter, RG-11 and run with it.

 

[red hazard]

The Calculations

Actually, you CAN add and subtract the loss numbers - at least that's what the pros that spec'd out my WiFi hotspot system were doing - but yes, dB is a logrithmic scale - I don't know more than that.

Yeah, I forgot he was Legacy, myself. That's a weaker signal, so we might need even better cable?

I guess what's needed is for some supplier to take a 500' roll of cable and try it as a lab test. Why 500'? Just for the extra margin - plus someone might have it lying around.

When I made the previous calculations there was a lot of estimating and Kentucky windage involved. Got off my A-- and brought out the calculator.
Belden RG11 7731A has a typical loss of 6.9db at 2250Mhz for 100ft. Converting the db to a factor (divide by 10 and antilog that figure) gives a loss factor of 4.898. Now multiply this figure by 3.5 to get the loss for 350ft which is a factor of 17.142. (This means the attenuation is about 17 times).
To convert back to db take the log of 17.142 (1.234) and multiply by 10 to get the loss back in db. The loss is therefore 12.34db. Hint: 3db loss is approx twice the loss or half the power. Add another 3 db (6db) and you have about 4 times the loss or 1 quarter the the signal strength before attenuation. Carol brand C5785 RG6 and Belden 7915A have 10.1db at 2250Mhz for 100ft. These are good quality coax. The El Cheapo stuff many installers use has more loss. Now if we take the loss for 165ft of this RG6 (which should work on a legacy system), we get a loss of 12.27db which is very close to the loss of 350ft of Belden 7131A RG11. The DC resistance of 350ft of Belden 7131A rg11 is 5.6 ohms as compared to 165Ft of Belden 7915A (18 Ohms) so switching polarization is not an issue either. By the way, you state that the DishPro LNBFs are better than the legacy LNBFs. Do they provide more gain or a better noise figure or both and can you post a link for us to view? Since the scenario that "willager" described is pushing the performance capability of a legacy system, he might want to consider upgrading to to a DishPro system if the LNBFs are significantly better. Thanks

 

[red hazard]

F Connectors for RG11

LMR-400 class cable has a center conductor of about 12 gauge, whereas RG-6 is around 20 (much smaller). It just might make the difference. You won't be able to find F-type connectors for cable that size, so you'll need a couple of short 'pigtails' to use as adapters. The pigtails at the Dish end need to be long enough to go through the LNBF mounting arm - there's NO way you'll stuff -400 class cable through there - it's about a half-inch in diameter.

See http://www.tselectronic.com/thomandbetts/sns.html
You can get Snap-N-Seal F connectors for RG11

 

[SimpleSimon]

Red: I also found F-connectors for the LMR-400 (Post #10).

I wonder why the TimesMicrowave calculator gives different figures than yours?

This datasheet: http://www.timesmicrowave.com/telecom/pdf/400-75.pdf
on the LMR-400 shows the loss is very near linear, and that the LMR-400 is only a little bit better than RG-11 (loss-wise - it might be a lot worse to install). The RG-11 might be a lot cheaper than 50 cents a foot - any idea?

My basis for Legacy being 'weaker' is their stated maximum distance being double for DishPro than the old stuff. Considering DP is a newer design, I would imagine that it's got both a better SNR and gain factor.

 

[red hazard]

Trustworthy Site

Unh. I thought the switching voltages were 13 & 18VDC, so I don't see how it can work at 10.5VDC.

that being said, if your other numbers are correct, you're right it should work on the DC side.

As for RF - my previous post gave numbers I got from the calculator linked there.

The table referenced by A_Pac is from a merchant, and it's numbers are a lot lower than Times Microwave's trusty calculator.

The data sheet for LMR-400 matches the calculator, and I trust Times Microwave.So, we've got too many different numbers for loss factor but we DO agree it'll be big $$$ to do it.

Oh - and TMR DOES have F-connectors for their LMR-400-75.

Well that link has technical errors under losses for meters. Apparently a marketing weenie thought that since there are 3.28 feet per meter that they could simply multiply the losses in db for 100ft by 3.28 to get the losses for 100 meters (6.5db x 3.28 = 21.3db??) Wrong!! This represents 14.8db additional loss (21.3db-6.5db) which would be 30.2 times for a length of cable that is only 3.28 times longer. That don't pass the common sense test!! The correct answer for 100 meters at 2500Mhz would be 11.66db which represents a factor of 14.66 times. The actual additional loss for 100 meters as compared to 100ft is 5.16db (11.66db-6.5db). 5.16db loss is equivalent to a factor of 3.28 times. FYI 6.5db loss is a factor of 4.47 times. Somone who exclusively uses the metric system would never buy that product looking at those loss figures.

 

[SimpleSimon]

Even with a possible typo as you mentioned, I still fail to see why the manufacturers (Belden, Andrew, Times) all seem quote loss per 100', AND the 100m numbers seem to be linear extensions (ie x3.28) of that.

Sure seems to me that your calculations must be off.

 

[red hazard]

Belden 7731A RG11 59 cents a Foot

Red: I also found F-connectors for the LMR-400 (Post #10).

I wonder why the TimesMicrowave calculator gives different figures than yours?

This datasheet: http://www.timesmicrowave.com/telecom/pdf/400-75.pdf
on the LMR-400 shows the loss is very near linear, and that the LMR-400 is only a little bit better than RG-11 (loss-wise - it might be a lot worse to install). The RG-11 might be a lot cheaper than 50 cents a foot - any idea?

My basis for Legacy being 'weaker' is their stated maximum distance being double for DishPro than the old stuff. Considering DP is a newer design, I would imagine that it's got both a better SNR and gain factor.

The cheapest place I found that sells Belden 7731A by the foot is http://www.westlake-electronic.com/ which charges 59 cents per foot. Other cheaper RG11 is swept to only 1000Mhz and has higher losses so I can't recommend it for willager's application.

As you probably surmised from my previous post, decibels are not linear. They use decibels because when you are calculating losses and gains on satellite budgets, long haul circuits, etc. it's a lot easier to just add for gains, and subtract for losses. Technically losses should be stated as a negative number. A "-3db" indicates a loss of about one half. A "3db" indicates a gain of 2 times. But for some reason manufactures prefer to state losses in positive numbers and I do too to keep from confusing folks. Decibels are calculated by taking the log of the loss or gain ratio and then multiplying by 10. For example. if you have an amp that doubles the power, the ratio would be 2 to 1 or simply 2. The log of 2 is .30 and then times 10 gives you 3db. If I cascade a second amplifier that has the same gain, then I have an additional factor of 2 times or 3db more gain. For total gain in decibels I merely add another 3db for a total of 6db gain. Now if I put the signal into coax that attenuates by 4db (-4db), I merely subtract that figure and I'm down to 2db total gain. As you can see 6db represents about 4 times the gain. 9db would indicate about an 8 time gain. -9db would be a loss of about one eighth (actually .125892541 according to my calculator). If you want to multiply a figure in db, you must first convert it back to the power ratio then perform the multiplication and then convert that figure back to db. I've seen folks make the same mistake that the TMR site did many times. They think db is a linear unit and it is not.

I'm not sure if its safe to conclude that DP LNBFs are significantly better than the legacy LNBFs. It's too bad that Echostar doesn't publish the figures. One thing I don't like about DP is that they up convert one transponder polarization up by 700Mhz. I roughly estimate that these higher frequencies (1650Mhz - 2150Mhz) sustain an additional 2 db loss on good RG6 than the legacy frequencies (950Mhz - 1450Mhz). Food for thought!

 

[SimpleSimon]

I fully understand that dB is not linear - but as you say, you CAN simply add and subtract the numbers. Therefore, a 350' run is going to 3.5 times the loss of a 100' run.

I saw that Andrews AND Belden make the same "mistake" that you say TMR does.

Maybe there's 2dB more loss, but the install specs say you can run it twice as far. To me that means more gain, and probably a considerably better SNR.

BUT WE'RE WAY OFF-TOPIC.

The guy just wants to know if he can do it. We agree that we almost certainly can - but he's gonna need RG-11 or better to get it done.