I am looking to purchase the Laird Sentrius™ RG186 LoRaWAN Gateway - EU868 for our school and i’m trying to work out coverage and if i’m going to need an upgrade attenna.
Does anyone have any information on coverage for this gateway or can recommend an upgraded antenna?
How long is a piece of string? 
There are many factors that affect range - gateway placement - generally higher (at least for the antenna) the better. The Ant with the system is more than adequate - indeed if you start to go down the path of using higher gain antennae then you introduce other problems (search forum) and may even need to back off the systems TX power to stay legal! Its a zero sum game.
These are great little units, that appear pretty bullet proof and that are easy to set up and deploy on TTN, I have double digits qty of them in my own fleet, they use the original SX1301 which is well battle proven. PSU is 12v but I know from own evals that they are good to much lower values, with good power brown out protection, if looking to alternate supplies. They simply self recover on power up and need little maintenance… go for it!
Depending on where you are range may be affected by e.g. RF absorbing obstables in the surroundings or terrain masking etc. I have one operating in a terrain limited area where range in one direction is only 2-3km due to hills, but in opposite direction over a gently undulating plain I get 10-15km with some hilly outcrops way to the north where the signal is again well served and at around 20-27km…your mileage may vary. Trick as stated is get ant up high and clear…a short ultra-low loss feeder cable can be used to optimise placement - keeping the GW indoors/sheltered with perhaps the Ant external and high - its not a weather proof ant but you can take steps to protect (self anealing silicon tape around the elbow and joints?) or simply buy a 3dbi external rated antenna (you may then need an SMA to N-Type connector/adapter - perhaps as that feeeder cable
There are no real differances in the ‘coverage’ or ‘range’ of the various Gateways, the base electronics are the same (from Semtech) so one is not realy better than another.
In most places in the World a basic antenna is already radiating the maximum legally permitted power.
For sure there are antennas which have more gain, so can reach further, but to stay legal you need to reduce the Gateway power, so in practice you achieve nothing.
Fully understand i didn’t really post much information, apologies.
We are UK based, We can locate it internally - i’m just unsure penetration through walls. I’m guessing with 868Mhz it would be fairly good.
Penetration through walls is generally not good at such a high frequency as 868Mhz.
As for how much the signals are impacted, it vairies enormously since the construction of walls and ceilings varies so much.
Some buildings use steel coated floor tiles, or steel plates in the concrete used to make floor and ceiling. Together with foil backed plasterboard and foil tint stuff in windows some rooms can be very good at screening RF.
So it depends. Needs to do tests to be sure.
I guess we should really start with what are you trying to achieve and what would be good coverage for you? Essentially there are two philosophies - build a network from the outside in, or from the inside out! Outside in is the classic cellco/tv broadcast model - GW’s trying to penetrate into buildings with all the limits Stuart mentions, the outdoor GW’s then de facto generally provide good outdoor coverage in the areas they serve. The inside out approach is where the business model for e.g. devices like the TTIG (and it could be said the RS186) where you put a GW in a building largely ensuring good/reasonable coverage in the building and its immediate surroundings, but then wider area/community coverage is dependent on how well the signals can penetrate out through some of the challanges of construction Stuart mentions - foil back plaster board can be a pig! The example I gave above is for a GW mounted in the attic of a businessman’s home who collaborated with me on a local deployment. When 1st deployed we realised GW wasnt high enough in the attic and the metal ring beams at the top of the house that the roof sat on caused RF shadowing some 250-1km out… knocking a nail into one of rafters high up and simply hanging the GW from that provided clearance and the results above I did note that the roof felt under the roof tiles appeard to be quite high metal/mineral content and that seems to have caused a few db’s of additional loss (when compared to deploying at one of his commercial buildings a few hundred meters away using same GW and antenna.
For you is good coverage just ensuring coverage in the school and surrounding grounds or are you looking to provide community or students coverage? Where is the school? (If concerned to declare openly on forum DM me and can take a look)
One of the target markets for the Laird GW (And indeed their sensors) is ‘cold chain’ applications - ie. monitoring fridges/freezers and open offices/kitchens etc. and potentially in the logistics/supply chain. Typically these are indoor deployments. You would be surprised how well the LoRa modulation signal can escape from the inside of a fridge/freezer, be they standalone applicances or even big commercial walk in units, despite near faraday cage like construction - with lots of metal and insulation in the way… mosts schools seem to be quite well served by a singe GW from my experience - with few areas truely dead (Maybe hidden behind a metal fuel tank at farthest distance is a problem in one case I saw), and that can often be mitigated against by a 2nd GW depoyed elsewhere on site to cover any dead spots…
If you already have the GW in hand as Stuart suggests simply deploy and test…
We are mainly looking at support sensors within schools (environment etc) but would have a few outside the building (particulate, weather etc) but these would be attached to a school building.
The whole aim around this is the feed the data into Zabbix and then display via Grafana. It is a part of us looking to educate children around sustainability and the impact our decisions have on it.
I’ll PM you our largest school.
Thank you for the replies.
Just had a quick look at the details on the PM so will come back with more there, for the Forum, however, I would say this is likely a no brainer in terms if a single well placed GW giving coverage to the school buildings and surroundings, with <100m approx n-s spread and approx 250m e-w building spread and with overall site approx 100-120m n-s and possibly 500m e-w no point is a long way away from a centrally located GW, and it would be a bit of cruel fate to find some point on campus for what I assume is a 60’s/70’s build with later additions and enhancements (nice fleet of solar panels 
) where there is a dead zone. From there and depending on if the main rail line to the south is in a rail canyon (good!) or a raised embankment (bad!) you will likely see good community coverage - likely 1.5-2km+ (perhaps more depending on placement) to the raised ground to the north/north east and likely 5-8km generally south/east/west Shame the school sits low relative to the northerly row of housing and access road.The only way to be sure is to test as we have said, so to that end if interested I have a configured for TTN RS186 sat doing nothing in my home office (pending early Q1’23 deployment) that I would be happy to lend to you to trial (possibly with a basic node or two - Things Uno or more likely a Things Node) to see what coverage you will get in areas of interest. I am due up your way before Xmas to collect a family member from one of the regional Uni’s and could drop it by (may have to be a Sunday!) and to collect when they return to Uni mid Jan …'fraid it means you might be working over Xmas/NY! 
If interested we can co-ordinate over the PM thread… more later.
All I would ask is that if successful and goes ahead that you post some summary details to the forum as benefit to other community users after the fact… 
I looked at the trust site and see the list of schools - perhaps we can discuss some of those in the New Year… I note also the potential new school call out and as I used to live in that village I would be happy to help there for certain as a legacy and to benefit the local community
You do know it’s a community, no secrets in saying where you are thinking of placing a gateway - maybe there is help at hand?
Indeed but for some new users and especially with something like a littlies school a little discretion or caution isnt unjustified to begin with… no doubt opening up as and when needed Have seen this many times in the sector…
I would rather not publicise our schools at this point, but of course i understand the community nature of this project and so will do further down the line.
@Jeff-UK I have just placed an order for a RG186 GW and a temp sensor so will be able to do full testing. Happy to post some summary details though. We are hoping to work with a partner on hooking up the water meters (we have had a few underground leaks that were hard to detect and therefor costly for schools. Insurance will cover the fix, but not the water lost. One of them is around £6k!) as i haven’t the knowledge on how to interface with them.
I will be using Zabbix for recording the data (we use this already) and to establish a baseline, we then have a SMS gateway hooked up to for sending SMS alerts. Grafana will be used to create dashboards for children/staff to look at.
The biggest area is around the education, i think it becomes powerful when children have a local context to learn from and i can’t do that without local data. We are in the process of writing a few grant proposals to see if they will part fund. Certainly exciting and fingers crossed we are accepted.
If you learn the basics of DIY devices you can double or triple the value of any grants. But do a mapping exercise before hand, never good to give a device to somewhere that doesn’t have gateway coverage. On the other hand, some schools may be open to hosting a fill-in gateway.
I dont think its the knowledge that is a problem, but the time and capacity of diy sensors.
Can you explain what do you mean by that ?
I presume diy devices is meaning dev boards etc as ppposed to off the shelf
I would presume that too, but what do you mean by ‘time and capacity’
Some dev boards are used in off-the-shelf devices and some off-the-shelf devices are just dev boards, some dev boards are off-the-shelf and some off-the-shelf devices aren’t available ‘off the shelf’. YMMV
You can take something like an Adafruit Feather M0 with RFM95 and have it doing what you need in an hour, or less. You can take the very versatile Dragino LT-22222 and spend a day figuring out the AT commands to get it to do what you need. Many devices work out the box in-line with the world view of the company that sell them. A very rare few actually get their creations tested by real human beings. Even less of them resolve the arising issues.
So much easier to define what you need and then look for a solution. The single biggest road-block on this forum is people not saying what they actually want to do, thereby allowing the mass-mind to make recommendations based on real-world experience.