HELP! Microwave link dish questions

[Equinox]

I hate to bother you guys but I was hoping someone with more knowledge than me could give me some answers. I have an old 2.1 ghz Starpoint microwave link that shoots ~30 miles point to point. We had to replace an OLD tower with a new one, and everything went fine when the tower crew made the move. With minimal sweeping, RX levels were better than before. That lasted two days, when the crew taking the old tower down whacked the feed horn of the dish. We have tried bending it back but we are getting no joy. The dish in question is an Anixter-Mark 10 ft grid. We happen to have a 6 ft grid that is not being used and was in good working order when it went offline. The question is, will the 6 ft dish work in place of the 10 ft, or will there be too much of a drop-off in gain? Or will a better option be to try to locate a new feedhorn for the 10 footer (may be tricky)? We aren't shooting anything back from the location in question, just receiving. Desperation setting in, any help will be appreciated greatly.

[wa6jbd]

Going to a 6' dish, from a 10' dish will cause the received signal to drop by about 4.5 db, give or take. Unless the fade margin is so close to the edge to start with, it might not make much of a difference. If you're concerned about path availability, the path may no longer meet the requirements with the smaller dish. But it WILL establish the link.

[N4DES]

Your beamwidth will probably change as well and a material change like this usually requires a re-coordination via ComSearch (or one of the other coordination companies).

As to performance, I'm sure it will work just fine and you won't see any real differences.

[Jim202]

As others have said, changing the dish out to a smaller one might get you back into service marginally. However,
it will cause you untold grief. Long term solution is to replace the damaged dish with a new one. Not knowing
just how close to fade margin the path and original dish were designed, there is no way of giving you any
feedback on how the path will work with the smaller dish. Sure you could get a short time waver on
using the smaller dish. But that won't help you with bad weather and thermal inversions taking out
the path.

Bottom line is I hope the tower crew has good insurance. This poor quality work on their part will cost
someone like the owner of the tower crew a big bundle. Not only will he have to shell out for the new
10 foot dish, he will have to go out and install it for free also. This is his mess up and he is responsible
for making good on it.

Big problem here is trying to get a replacement dish in a short time frame.

Jim

I hate to bother you guys but I was hoping someone with more knowledge than me could give me some answers. I have an old 2.1 ghz Starpoint microwave link that shoots ~30 miles point to point. We had to replace an OLD tower with a new one, and everything went fine when the tower crew made the move. With minimal sweeping, RX levels were better than before. That lasted two days, when the crew taking the old tower down whacked the feed horn of the dish. We have tried bending it back but we are getting no joy. The dish in question is an Anixter-Mark 10 ft grid. We happen to have a 6 ft grid that is not being used and was in good working order when it went offline. The question is, will the 6 ft dish work in place of the 10 ft, or will there be too much of a drop-off in gain? Or will a better option be to try to locate a new feedhorn for the 10 footer (may be tricky)? We aren't shooting anything back from the location in question, just receiving. Desperation setting in, any help will be appreciated greatly.

[mruwave]

All the guys above pretty much nailed it. Looking at the Andrew (now Commscope) catalog, a 10' gridpak at 2.1 GHz. will have a midband gain of 32.8 dbi with a beamwidth of 3.7 degrees. A 6' solid 'Standard' dish will have 29.4dbi gain with a 5.5 degree beamwidth. A 6' solid 'High Performance' antenna will have identical gain and beamwidth specs as the 'Standard', but will give you 9 db better front/back isolation. I ALWAYS recommend users purchase 'Hyper' antennas because of congested bands and limited channels, but depending on the service you're licensed for, you may not have this issue.
Bottom line, make the contractor replace the antenna (Andrew Catalog 38 still lists the 10' gridpak [KP10F-19]) and don't accept substandard performance. If your path reliability was in the 98% or higher range, the temporary 3.4 db loss will probably keep you working until you can get a replacement for the 10'.

[Equinox]

Thanks for all the replies. Come to find out, it was going to take much longer than I would have liked to get a new feedhorn. But as luck would have it, the tower crew had a source for an 8' grid locally. We put it up today and it is working fine. I'm probably still going to get the feedhorn for the 10' but at least we are back on the air. Again, thanks for all the help.

[/\/\y 2 cents]

Hello.

Use the Internet or a LAN or WAN and you wont need expensive microwave or towers or any of that cold war stuff.

Google "PackeTUBE" to learn more about the future.

Steve

[wa6jbd]

Use the Internet or a LAN or WAN and you wont need expensive microwave or towers or any of that cold war stuff.

Google "PackeTUBE" to learn more about the future.

I hope you're kidding. With internet service availabilities on the order of 99% or less, and individual packet loss of around 3% (97% availability?), I would hesitate to do anything less than a licensed, engineered microwave system for critical circuits. Link availability of 99.9999%, and longhaul availability of 99.999% is easily achieved with real radios on protected frequencies. I guess it's all a matter of priorities. You can get it done right, or you can get it done cheap.

[/\/\y 2 cents]

Hello??!? Earth to the telecom folks (thats what you are right if you really get down to the heart of it? Motorola is plain old telecom gear even if you have a fetish for it, same rules apply so you should think like a telecom person and not just a 2-way person!)

You can use BGAN which is like microwave (Directional RF antenna) but pointed to the sky, packetswitched, and available to everywhere but the most remote places on earth. It does not have the upfront cost or overhead associated with building out and running terrestrial microwave. Also it is more hardened from failure unless we have a Regan era star wars missle scenario, in which case everybody is S.O.L like when try to use the ATM to pump gas to get out if town and your bank cant see your checking.

Typical Microwave transports in TDM which is very high bandwidth (24 64k DS0's)..The its all a radio carrier frequency SAT vs. Directional microwave just a better mode to use IP vs. TDM. You could make a terrestrial IP system like canopy (but licensed) with ethernet instead of TDM, but the RF side is the same and you will need much infrastructure because fo the spatial nature of radio waves.

Sounds like you may be thinking about how you have seen it done you whole life rather than how people will be doing it in the future and your twilight years. Backhaul is backhaul and there are many ways to do it, the question becomes do you want to do it progressively and improve or do you want to do it the old way and stat quo. BTW BGAN will not be cheap or inexpensive and neither is the hardware although it's a better deal and lasts longer than microwave. Just ask your cellular carrier.

Steve

[Astro Spectra]

Sorry to disagree but I’m with WA6JBD on this. It is hard to beat a properly engineered and licensed microwave link.

Inmarsat BGAN is good if you’re funded by USAID and wandering around Africa but the $ charges are huge, the latency is high and every month your money is going to an overseas corporation that won’t even guarantee service (service is on a ‘best effort’ basis). BGAN is really not suitable for more than itinerant use and, as with all satellite solutions, you have two sun outages a year to work around. In VSAT systems with tight link budgets you need to also worry about heavy precipitation.

Finally wait till you try and use a PTT based LMR system over a satellite hop…

Today’s modern microwave from respectable vendors (NEC, Alcatel/Lucent, Ericsson, 4RF, and Harris/Stratex) offers both IP packet and TDM PDH operation.

If the budget is tight, the path length relatively short and not over a built up area, then you could consider 5.8GHz unlicensed. Products like the PTP 500 from Motorola are rather good (actually designed in the UK by Orthogon). They are not actually a lot cheaper than higher frequency microwave but you don’t have to do the FCC thing. Only problem is most unlicensed is IP only as standard while in the real world we still need 4W E&M, T1, etc.

There is no way you can rely on cellular or ISP infrastructure for public safety or critical infrastructure applications. Great adjuncts, sure, but not to trust lives on.

[/\/\y 2 cents]

"Sorry to disagree but I’m with WA6JBD on this. It is hard to beat a properly engineered and licensed microwave link."

Microwave has performed but not in earthquakes, hurricanes, tornado, etc. Yes microwave can hold too but not as well.

Latency is very low. Right now many public safety and heavy industrial rely on BGAN and they don't even know it because the last mile is delivered over legacy copper to a legacy TDM or E+M physical interface jack. If you would like to see what latency is like using BGAN via a 800MHz ASTRO smartzone system I invite you to talk to me over it from you PC or Smartphone to a DIU3000 or ASTRO Spectra+ which is sitting on the back of a BGAN Terminal. Latency is somewhere hovering about 120ms which cannot be perceived by any of us mere mortals, especially with push to talk, release to listen. Also have EMA SIP telephone sharing bandwidth and there is perceived latency or echo.

It all comes down to cost vs. performance. BGAN is becoming less expensive. Sure, it's not our fathers connection, but it has many advantaged over building a land based backhaul to couple to a Land Mobile radio system.

Both methods are valid.

[Astro Spectra]

Sounds great but how can the latency be less than at least the earth-space-earth time of 240ms or 480ms for a both way ping?

Inmarsat themselves quote 800ms to 1000ms if network issues are factored see:

http://business.singtel.com/upload_hub/ ... eneric.pdf

I'm not saying it can't be used, just that when I've used LMR over VSAT in Africa the turnaround times made PTT operation to VSAT linked units not on the same repeater a very stilted affair.

[/\/\y 2 cents]

Hi,

The units we used were not form Inmarsat. They were from tracstar.

http://www.tracstar.net/

All I can say is that they had the same QoS as a managed circuit from the bell and really worked great and were not all that expensive. Latency was not calculated round trip but was totally transparent.

Steve

[tvsjr]

Enough. OP's questions have been asked and answered. Time to stop flogging the deceased equine.