How to gain more visibility to those in commercial RF?

[KC8QVO]

(hint, it had to do with how OSPF works and convergence times). If that's greek to you...can't really diagnose the cause of the issue and come up with a suitable workaround.

It isn't completely greek, so long as the terminology of "convergence" is very much the same. On some Differential GPS systems (Trimble's OmniSTAR and RTX, for example) there is a convergence period while the GPS position data is normalizing in the receiver. The higher the level of service the faster the convergence times, repeatability, and accuracy (loosely - there is a lot to it). If there are convergence issues then differential GPS cuts out.

There are lots of heavy equipment machines (Ag, Construction) that run on guidance. Most of those guidance systems are locked to where Differential GPS has to be present. If it isn't - no guidance.

Likewise, no convergence = no differential = no guidance.

 

[Project25_MASTR]

It isn't completely greek, so long as the terminology of "convergence" is very much the same. On some Differential GPS systems (Trimble's OmniSTAR and RTX, for example) there is a convergence period while the GPS position data is normalizing in the receiver. The higher the level of service the faster the convergence times, repeatability, and accuracy (loosely - there is a lot to it). If there are convergence issues then differential GPS cuts out.

There are lots of heavy equipment machines (Ag, Construction) that run on guidance. Most of those guidance systems are locked to where Differential GPS has to be present. If it isn't - no guidance.

Likewise, no convergence = no differential = no guidance.

To simplify things...convergence is the time it takes for OSPF to recalculate the Open Shortest Path First. Of course, as soon as a shorter path becomes open, it re-converges again. Anyway, this convergence time can range from 4-8 seconds for Point-to-Point links to to up to 40 seconds when you introduce Point to multi-Point or broadcast links (which are metrics in OSPF in this case).

 

[mmckenna]

Is simply a "class" enough? Or would a "certification" be better? And if so - what type?

If simply a "class" is enough - what specific subjects/applications are important?

I am sure the right work experience carries more weight, but without it what is the next best step to get somewhere where I can get experience?

My personal preference is work experience over education. I'd rather have someone who's actually done it in a real world environment than someone who just sat in a classroom and took a test. Book learning vs. street smarts. Give me the street smarts. Book learning is fine for someone who's going to sit in a well lit office from 9-5 M-F. But when the doodoo hits the fan, I want someone who knows how to get $^&* done, not someone who's going to go by the book.

But what really matters is that you are trainable. The trick is getting that capability presented to the company you want to work for. It's hard to prove in a 30 minute interview. Showing that you've taken courses and passed industry type certification tests can sort of prove that, but it won't prove experience. Showing that you have experience means you've likely learned enough to get by and have survived in the industry. That may mean you'll need more training, but that's sort of universal in the industry. If you ain't learnin', you're falling behind. All my staff has a $5,000/year individual training budget that I more or less require them to use up. And that $5K doesn't go far. That can equal a week or two of training a year.

Some managers are impressed by a list of acronyms, brand names and certifications. Especially if they match what they have themselves. Some are impressed by brand names, but those are usually the guys who don't know as much as they want you to think. Big names like Cisco and Motorola will impress those that don't really know that much about the industry. They are recognizable names that most know. But having other certifications are just as good.

In other words, you need to show that you know enough to not be a burden on the company for the next year, and that you are trainable. Not everyone is trainable. From what you've shown, I'd bet you'll pick stuff up along the way pretty easily.

On a bit of a different twist - is it worth trying to get my foot in the door somewhere in a related area of RF while working towards a GROL, networking knowledge, etc? Or (figuratively) flip burgers until I can get those?

I'd say both.
Ideally, getting your foot in the door, even if it's pulling wire for a cabling contractor, has some value. You'll likely learn no matter what you are doing.
On the other hand, flipping burgers might pay the bills until things line up.
Really depends on your situation. I'd say that your interest is in technology, so keep working in that direction.

The county I'm in has it's own radio shop. They handle a lot of the local public safety agencies, the 911 PSAP, and all the county radio stuff. The guy who ran the shop for years was an ex Motorola guy. He was good, knew what he was doing and was easy to work with. He needed to hire some install bay guys to do vehicle installs. One of the guys he hired was a mechanic from one of the local car dealers. He knew automotive and electrical, so was a real boost to the shop. Over time he kept taking courses, taking manufacturer training, got his GROL, etc. Now that guy that was a dealer mechanic is the shop manager.

So, it's really possible to do what you want. It's just finding the right opportunity. Might mean starting off in the install bay wiring up radios and lights, but that's not a bad thing. You'll get the opportunity to prove yourself and expand you knowledge. That'll let you move up.

 

[KC8QVO]

So, it's really possible to do what you want. It's just finding the right opportunity. Might mean starting off in the install bay wiring up radios and lights, but that's not a bad thing. You'll get the opportunity to prove yourself and expand you knowledge. That'll let you move up.

Just for sake of discussion - and my apologies if this "pokes bears" reading this forum (I would venture to say most people, just a guess, on the forum are in the two-way industry in some form) - I am not wanting to tie myself entirely, and solely, to two-way. I am not opposed to it, but I want to know and explore what all is out there.

From the opening post in this thread:

I am not sure if this is the right place to post. However, I am looking to get in contact with people in RF related industries to see what opportunities there may be out there that might be obscure. I am not necessarily interested in working specifically with two-way radios (business band radios, or any other industrial comms use) but I am not opposed to it.

There are a lot of other applications of RF systems. As some have mentioned in the thread - satellite broadcast is one example.

I suppose in stating that - what I also don't know in relation to what is out there in the industry - is how much cross-over there is. Telecommunications may be very broad terminology.

I know there are certainly two-way shops around.
There are Wireless Internet Providers around
There are tower (iron side primarily, little RF side) companies around

I will research the satellite services and see what I can find there.

Is there inter-mixing between all the industries that use RF systems, in one way or another, such that "shops" that, to me are "two-way shops", actually cater to any number of industries - satellite and internet, in my examples here? Or are there environments where these industries are, and have to be, segregated by ways of specialties in knowledge/application/information/environment?

 

[davidgcet]

you can say that you have used service monitors to tune dulexers, etc... without harping on it being for amateur bands.

I worked in commercial 2 way from 1991 until 2018 when we shut down our radio shop. I now work as a network field operations tech 2 for a cell carrier. my list of experiences and knowledge got me hired as a tech 1, even though I had more years in "rf" than even most managers. the reason, while at its root rf is rf, companies want people with direct experience as it relates to their needs. I had no experience with Lte and very little with cdma and what I did have was not on the same equipment they use. I took the job as an entry level and used my time wisely to show not only my existing knowledge but that I could and would learn what I didn't know. so in a total of 13 months I was bumped to a tech 2 which usually takes 5+ years to attain. oh and btw, I get to do very little actual rf work beyond diagnosing antenna/line issues, most stuff now is a FRU so we don't even turn pots.

as far as 2 way shops go, most techs and managers that ive met over the years(and I have met many in my travels) detest most hams. the reason is simple, so many come in with the "I know more than you" attitude and quite frankly they don't. especially modern hams, as ointed out above the test is so simple now you don't have to really know anything to pass it. we had one that worked for us that got his ticket back in the 60's, I forget what level but he had to know theory and practice as well as be proficient with CW. a friend of his got his ticket in the late 90's which was equivalent but he doesn't even know how to tune a filter. our guy built combiners for us from scratch, made the tuned shorting stubs and all at his bench, yet his "equal" could not even turn the knob and get max pass thru. along the same lines though, my stepdad got his first class back in the 60's and it took him a full day of testing and they had to work out everything on scratch paper with slide rules. I took my GROL in the early 2000's and aced the test in under 30 minutes sitting at a computer. I don't profess to have the same level of knowledge, but I do have more experience on modern systems yet older stuff id have to call him every time.

my advice would be suck it up and get your foot in the door somewhere. start low, aim high and show you have what it takes to get there. you will not step in as a manager or top tech, you simply do not have the industry experience for it yet. but it you know as much as you claim you can get there rather quickly. specialty stuff you have a better chance if you hire on at a big shop with multiple locations or at a wireless carrier, most places need more of the jack of all trades types and not necessarily the specialist that can only do rf system work.

 

[davidgcet]

oh and btw I had a ton of classes and certs that I put on my resume and filled in on my hr form when I hired on. not one counts for a hill of beans since none had anything to do with cellular directly. they don't even list my GROL since it is not for cellular. ive honestly found that my experience and prior training is not needed at all for my current job, but it has earned me a lot of respect among coworkers and a few higher ups. so make sure you get some good general certs but definitely get some that apply to the general industry subset for which you are applying.

 

[KC8QVO]

I worked in commercial 2 way from 1991 until 2018 when we shut down our radio shop

That theme seems to come up in the ham world - a lot of businesses have gone under (both commercial two-way and ham dealers) in the last 10 years. I've heard from a ton of people, and even have seen this in interviewing for some vendors, that two-way isn't what it used to be - like back in the analog days in the 90's.

Your transition was to cellular.

What does the current industry need and what is looking like the future = voids that are getting bigger and need filled?

Networking knowledge has been defined already. Speaking to the cellular industry - is that industry on a "wave" like the two-way world was for 15-20 years? Is that "wave" nearing any crest? Is it building? Is there a wave that is in it's infancy now that will challenge the cellular industry any in 10 years?

Are there any industries, like satellite services of one form or another (not meaning DirecTV, or any other consumer cable/internet provider, but industrial/commercial services) that are in a similar wave? Are there technologies brewing that challenge those services also that are worthy of noting?

 

[prcguy]

I don't think its a good time to be a DirecTV installer since ATT took over the company and is laying off people and demanding more work for those who are left. Viasat recently launched the highest capacity satellite for Internet on Ka band (300Gbit) and there is a need for ground stations around the country and other parts of the globe. Otherwise the commercial satellite business chugs along needing occasional new blood to replace the dead techs and incompetent ones.

Are there any industries, like satellite services of one form or another (not meaning DirecTV, or any other consumer cable/internet provider, but industrial/commercial services) that are in a similar wave? Are there technologies brewing that challenge those services also that are worthy of noting?

 

[davidgcet]

any industry that doesn't keep up with evolving tech is prone to be hit in the future. 2 way definitely has taken a huge hit from cellular because we lost all the "base and two" customers, the ones that had a base station and handful of units to dispatch their guys. my area is largely rural so we had tons of farmers. the smaller ones that have survived no longer use radios, to them it is easier to call a man on his phone regardless of where he is than it is to try and raise him on a simplex vhf from 10-15 miles away.

I can also remember when paging was the golden ticket. we were tied in with a couple regional and nationwide carriers and raking in money hand over fist. had an offer to buy our company out in 98 for a few million bucks, which came out to about 3x ebidta. my family and the business partner decided it was not a good enough offer to consider seriously, as we felt it should be at least 5x. guess what, by late 00 and early 01 the bottom had absolutely dropped out from under paging and we lost nearly 75% of our sub base to cellular. how we kick ourselves still for not selling when we were at a near peak, but no one really took the threat of cell phones seriously.

I have zero experience in the sate industry so I really can't say what would be a good idea there.

if you look in to cellular make sure you learn Lte basics but rally concentrate on learning 5g. it is completely different than previous technology in the way works. of course the word is 5g will handle the future for quite some time, but so was th prediction on Long Term Evolution...…...

 

[speedway_navigator]

We sold our paging company once the cell folks figured out how to make telco to cell calls work without having to know which area the person was in and use the overdial #. My cousin saw the writing on the wall, it was a good call.

 

[KC8QVO]

A quick example, when Hurricane Harvey hit Texas, I was in Austin working on a bouncing microwave link. The link was continuously going up and down due to the unusual high winds from the hurricane and was causing three sites feed off of it to constantly go in and out of site trunking (which was problematic for two dispatch centers which used the site). Sitting there and watching the microwave radio, you could see the link come and go however until you figured out what was going on you didn't know how to fix the issue (hint, it had to do with how OSPF works and convergence times).

Comments specific to bold text, whole quote to place in proper context of where they originated.

Were you there after the hurricane = postmortem? Or were you there "for the event"?

Reason I ask is you mention high winds. That would immediately be obvious that the correlation to wind is that it was causing something to move. If what was moving was a microwave dish through which the networking was occurring and that signal level was going up and down it makes a bit of sense.

However, the following doesn't quite register:

To simplify things...convergence is the time it takes for OSPF to recalculate the Open Shortest Path First. Of course, as soon as a shorter path becomes open, it re-converges again. Anyway, this convergence time can range from 4-8 seconds for Point-to-Point links to to up to 40 seconds when you introduce Point to multi-Point or broadcast links (which are metrics in OSPF in this case).

The idea of "convergence" makes sense how you use it. However, I believe the application of that term is the overall network path - from a packet being sent to being received. The OSPF protocol is on the packet routing. In the microwave theory above - that is a physical path, not a network path, though is both. You either have that physical path or you dont (with two dishes pointed at each other a link doesn't have a choice of what path to take, as in a network path going through different routers/servers/etc). Is there something in the signal level of that link that is driving the network path? Or is it simply the physical path being there = network path is there, then when the physical path drops (from the position of the dish in the theoretical example blowing in the wind = knocked out of alignment) then settles back it re-establishes the convergence process? Then the whole issue repeats causing the link to "come and go" (see highlighted and underlined in the previous quote)?

I guess to possibly undermine the theory - if the issue was present AFTER the storms when the wind had died down then the "movement in the wind" theory doesn't stack up. Then again, if the mount of one or both dishes physically got knocked out of alignment (or the structure they were mounted to bent causing a misalignment) it would be a matter of re-aligning the dishes. Though, you should be able to see that your signal levels are way off when you check them. I would imagine that would be real easy to see remotely.

 

[Project25_MASTR]

Comments specific to bold text, whole quote to place in proper context of where they originated.

Were you there after the hurricane = postmortem? Or were you there "for the event"?

Reason I ask is you mention high winds. That would immediately be obvious that the correlation to wind is that it was causing something to move. If what was moving was a microwave dish through which the networking was occurring and that signal level was going up and down it makes a bit of sense.

However, the following doesn't quite register:



The idea of "convergence" makes sense how you use it. However, I believe the application of that term is the overall network path - from a packet being sent to being received. The OSPF protocol is on the packet routing. In the microwave theory above - that is a physical path, not a network path, though is both. You either have that physical path or you dont (with two dishes pointed at each other a link doesn't have a choice of what path to take, as in a network path going through different routers/servers/etc). Is there something in the signal level of that link that is driving the network path? Or is it simply the physical path being there = network path is there, then when the physical path drops (from the position of the dish in the theoretical example blowing in the wind = knocked out of alignment) then settles back it re-establishes the convergence process? Then the whole issue repeats causing the link to "come and go" (see highlighted and underlined in the previous quote)?

I guess to possibly undermine the theory - if the issue was present AFTER the storms when the wind had died down then the "movement in the wind" theory doesn't stack up. Then again, if the mount of one or both dishes physically got knocked out of alignment (or the structure they were mounted to bent causing a misalignment) it would be a matter of re-aligning the dishes. Though, you should be able to see that your signal levels are way off when you check them. I would imagine that would be real easy to see remotely.

Austin didn't get hit physically. There was a panic which made to a fuel shortage (I had to drive 90 miles to Waco to fill up my company van one day) and there were unusually high winds the Sunday Harvey made landfall. The actual problem I got the call for was that three of the ten sites that made up that "wing" of the system were constantly going in and out of site trunking. A P25 system goes into site trunking when it can't reach the system controller (sometimes called a zone controller) as it holds the databases regarding what ID's are valid and what talkgroups IDs can affiliate to.

OSPF is known as a dynamic routing protocol. As such, it is commonly employed as a redundant routing protocol. In other words, multiple paths to the same destination. As an example say you have a router, plug it in to another router then that one into another and so on until you get to your last router...which you plug into the first (to form a ring). Each link from one router to the next has a cost. Say you want to send info to a router two hops away. The shortest path...is two hops away but the long path is also available. The router uses the open shortest path first. The protocol is constantly sending status between each link so if a link goes down...it reroutes but the change doesn't propagate across the entire system immediately...thus the time to convergence. Now if that downed link comes back up...well we need to re-propagate that change because it's the shortest path.

Routing is Layer 3...the connection between one microwave radio and the other is Layer 2. That being said, it wasn't a misalignment issue. The issues was one of the dishes used only had 1.5 degrees of beamwidth so when they caught the wind and began shaking causing the beam to sit there and oscillate on and off the target site however many miles away. The solution was simple...put a stiff arm on the dish to hold it steady...it just wasn't initially engineered that way. I would've personally used +24 dBi dishes instead of the +42 dBi dishes that were used as it would have provided more (quite literal) wiggle room given how hot the signal was for the 10 mile shot (there was some 60 dB of spare link budget) but I didn't design it nor did I install it.

 

[KC8QVO]

The issues was one of the dishes used only had 1.5 degrees of beamwidth so when they caught the wind and began shaking causing the beam to sit there and oscillate on and off the target site however many miles away. The solution was simple...put a stiff arm on the dish to hold it steady...it just wasn't initially engineered that way. I would've personally used +24 dBi dishes instead of the +42 dBi dishes that were used as it would have provided more (quite literal) wiggle room given how hot the signal was for the 10 mile shot (there was some 60 dB of spare link budget) but I didn't design it nor did I install it.

Thanks for the explanation. That was a really interesting read.

Was all you looked at the "symptoms" of that particular network dropping the 3 locations to site trunking? If you didn't have intimate knowledge of that particular network from the get-go where do you start gaining that understanding from which to follow the rest of the troubleshooting? I imagine the 3 sites of that system that had the problem were the ones routed through that wobbling dish and I assume those are the 3 that were noted in the initial complaint. Or, were the 10 sites that you mentioned the total number served by that microwave link and only 3 of the 10 had a problem where as the other 7 did not (though they were routed through the same link)? Or is the 10 figure the total network with 3 being the number in the "wing"?

Going back to the signal strength being obvious - I used a program a while back for setting up a satellite internet service where the radio module on the dish was able to ping the satellite service. The idea was, when in the test/set up mode, that you move the dish slowly by hand to peak the signal then fine tune with the mount screw adjustments. I presume the type of "radio" used in the microwave link would, likewise, have a test mode, either active (that you have to put the units in to do the test), or passive (where you can take readings without interrupting service), that would show you the signal level in some fashion.

A lot of the systems I've worked on use RSSI, so I am going to throw that out there as one level to watch. Although, every RSSI spec I've ever seen varies wildly per the device/manufacturer. How RSSI is calculated/displayed from device to device is never uniform except to specific devices. For example - a cellular modem's cellular RSSI isn't going to be the same scale of measurement of strength as a repeater receiver's RSSI scale, in other words.

If you saw a block diagram of the network where the 3 sites feeding off that microwave link had the issue and the dishes were suspect then wouldn't it be obvious to check the signal reading (RSSI levels) on the dishes? Or is there a different process done with those systems that helps you align them?

 

[Project25_MASTR]

The total network size was roughly 70 something sites at that time. The 10 site wing was a sub-section of the network. Essentially owned by a rural COG and interconnected completely via microwave (no leased circuits). That particular site where had a leg with two sites behind it (zero redundancy for those two other sites) so when the site I was working on's network connection went down, it drug the other two with it.

I had a map with the link paths on it, was familiar with the IP scheme (standard Motorola scheme for the most part) and a good chunk of what I didn't know, I could map out using network tools (traceroute and nmap). What I didn't have were the as-built documents regarding the links (such as RSSI and settings at time of construction). Those had to be tracked down as we were unsure who had them...did Cambium have it (since Canopy split from Motorola into Cambium), did Motorola have it, did the NOC have it, took a day or two. The temporary work around was to simply unplug the microwave radio from the router and force the site to go the long path (no redundancy).

 

[speedway_navigator]

The solution was simple...put a stiff arm on the dish to hold it steady...it just wasn't initially engineered that way.

During a talk at IWCE last year one guy who had a microwave dish in PR during Maria said his dishes were cat 4 rated but the tower crew used cat 1 or 2 rated mounting brackets so the dishes either were blown out of alignment or damaged, can't remember which.

 

[speedway_navigator]

(there was some 60 dB of spare link budget) but I didn't design it nor did I install it.

A certain PE we know quiet well had to admit he once burned a receiver by having too much gain in the system. He was new to microwave and figured more power was better.

 

[mmckenna]

During a talk at IWCE last year one guy who had a microwave dish in PR during Maria said his dishes were cat 4 rated but the tower crew used cat 1 or 2 rated mounting brackets so the dishes either were blown out of alignment or damaged, can't remember which.

I recall that talk. I think the brackets bent and the path was no longer aligned.

A very long time ago I had a link where the feedhorn broke off one of the dishes, so just an open waveguide pointed at the other dish a few miles away. Link still worked well enough. Sometimes over building works.

 

[KC8QVO]

That being said, it wasn't a misalignment issue. The issues was one of the dishes used only had 1.5 degrees of beamwidth so when they caught the wind and began shaking causing the beam to sit there and oscillate on and off the target site however many miles away. The solution was simple...put a stiff arm on the dish to hold it steady...it just wasn't initially engineered that way. I would've personally used +24 dBi dishes instead of the +42 dBi dishes that were used as it would have provided more (quite literal) wiggle room given how hot the signal was for the 10 mile shot (there was some 60 dB of spare link budget) but I didn't design it nor did I install it.

The temporary work around was to simply unplug the microwave radio from the router and force the site to go the long path (no redundancy)

So the microwave link wasn't the only path. That's interesting. The attempt of putting in redundancy to keep things from "going down" was the very source of the "going down". When one of the two paths was having intermittent issues the switching between caused convergence issues which in turn caused site trunking to bounce up and down. Yea, there's a lot of "if's" in there:
- If the wind wasn't blowing
- If the dish had been secured properly to start with
- If a wider beamwidth dish were used
- If the dish was in a different spot
- If the microwave link was some other type of link/service

And what was that saying between Ham radio and commercial/public safety?

On the public safety side, failure really isn't an option. On the amateur side, not a big deal, someone will eventually drive out to the repeater site and fix it. Eventually. That doesn't fly with a police, fire or EMS system. It isn't supposed to break, and if it does, it gets fixed now. Middle of the night, in the rain, with the power out, and you haven't slept in 36 hours. I've been through things like that, and it's not fun. But it does teach you to do things right the first time and mitigate risks. That's something that's hard to teach those that haven't been through it. Amateurs might talk about "when all else fails" and "emergency communications", but most don't really understand what that means.

If those that designed that microwave link had the mindset (bringing back the earlier point of there being a "different way of thinking" on the commercial side again here) of "failure isn't an option" and they knew they were building in hurricane country - who's head is on the chopping block for that one when the system DOES go down?

These are theoretical questions, by the way (IE - no need to answer them, they just illustrate how I thought through the conversation that's happened here).

If the inadequate installation of the wiggling microwave dish was a "cost-saving measure" - on who's calculator? Was it the customer's and they were given the lower grade installation as an option which they chose? Or was it the installers trying to squeeze out profit margin over the cost of the total job? I assume the installers are likely those that "designed" the system.

Usually projects done in the public sector go out for bids to companies that can do them. Lowest bidder wins. That is, unless there is a specific link between a government agency and a qualified/designated government contractor that is who is required to do certain work for certain agencies.

Interesting conversation. The point I am highlighting here is you can have things not set up well even on the commercial side. The "thinking" may be "different", in general. However, that does no preclude any installation and design issues from occurring even in the commercial world.

Issues are never a good thing to any system because they prevent normal operation, in some capacity (or, worst case - prevent even some level of normalcy to operation to begin with).

 

[mmckenna]

If the inadequate installation of the wiggling microwave dish was a "cost-saving measure" - on who's calculator? Was it the customer's and they were given the lower grade installation as an option which they chose? Or was it the installers trying to squeeze out profit margin over the cost of the total job? I assume the installers are likely those that "designed" the system.

Project manager.

Ideally the PM should be watching all this. Parts/materials should be all specified.
-If the wrong part is spec'd, then it's on the engineer.
-If the wrong part is installed, then it's on the installer.

The PM should be watching the installer. Someone should be checking their work.
If the engineer screwed up, then they are on the hook for it. Most have licenses/insurance, etc.

Doesn't fix the short term issue, though.

 

[Project25_MASTR]

So the microwave link wasn't the only path. That's interesting. The attempt of putting in redundancy to keep things from "going down" was the very source of the "going down". When one of the two paths was having intermittent issues the switching between caused convergence issues which in turn caused site trunking to bounce up and down. Yea, there's a lot of "if's" in there:
- If the wind wasn't blowing
- If the dish had been secured properly to start with
- If a wider beamwidth dish were used
- If the dish was in a different spot
- If the microwave link was some other type of link/service

And what was that saying between Ham radio and commercial/public safety?



If those that designed that microwave link had the mindset (bringing back the earlier point of there being a "different way of thinking" on the commercial side again here) of "failure isn't an option" and they knew they were building in hurricane country - who's head is on the chopping block for that one when the system DOES go down?

These are theoretical questions, by the way (IE - no need to answer them, they just illustrate how I thought through the conversation that's happened here).

If the inadequate installation of the wiggling microwave dish was a "cost-saving measure" - on who's calculator? Was it the customer's and they were given the lower grade installation as an option which they chose? Or was it the installers trying to squeeze out profit margin over the cost of the total job? I assume the installers are likely those that "designed" the system.

Usually projects done in the public sector go out for bids to companies that can do them. Lowest bidder wins. That is, unless there is a specific link between a government agency and a qualified/designated government contractor that is who is required to do certain work for certain agencies.

Interesting conversation. The point I am highlighting here is you can have things not set up well even on the commercial side. The "thinking" may be "different", in general. However, that does no preclude any installation and design issues from occurring even in the commercial world.

Issues are never a good thing to any system because they prevent normal operation, in some capacity (or, worst case - prevent even some level of normalcy to operation to begin with).

There are different levels of down and different levels of redundancy. P25 has three levels of operation for example...wide area (situation normal), site trunking (wide area down), and failsoft (last ditch conventional attempt). Site trunking can be fixed using patches with conventional resources. Case in point, Brazoria county during Hurricane Harvey where the links back to the zone controller failed and left the county's simulcast system in site trunking (simulcast/voting still operational). Simple patch between the system and now independent simulcast cell resolved the issue.

Now being 150 miles from the coast, Austin isn't exactly hurricane territory. In fact, when Harvey made it's first landfall, it was a Cat 4 hurricane but when it made it's second landfall 10 miles away (after crossing Aransas Bay) it had already downgraded to Cat 3 which is typically within 125 mph wind ratings most towers are currently built to. It should be noted wind ratings are for structure/component survivability and not link maintainability. Then again Harvey completely rewrote the 100 year flood plane standards in the Houston area as it dumped more rain than was ever seen with some of the most influential hurricanes in Texas history, notably Hurricane Ike in 2008, the Galveston Hurricane of 1915 and the Great Hurricane of 1900. You can only build to a standard that has been recorded...can't predict the future.

There's a joke we have, if you are ever stranded on an island you should bury a piece of fiber...a backhoe will show up shortly to cut it. There's only so much redundancy that you can put into a plan and it's difficult to predict it all as you tend to redesign things based off of past failures and adapting to constant changes is difficult. For example, since that issue, 3 sites have been added to that wing and the system as a whole has doubled in size (becoming the largest P25 system in Texas in terms of coverage area) as it was merged with another regional system.