Long Range Transmission
-
Has anyone tried to use any variants of the radios to cover a very long range ~800 meters? I ran across this:
I wonder if this could be used as a bridge between sensors to cover a longer range. Or am I reinventing the wheel here?
I have a gate that is 800m away that I would like to open/close via mysensors.
-
@Ironbar said:
Has anyone tried to use any variants of the radios to cover a very long range ~800 meters? I ran across this:
I wonder if this could be used as a bridge between sensors to cover a longer range. Or am I reinventing the wheel here?
I have a gate that is 800m away that I would like to open/close via mysensors.
Yes. I use the nrf24l01+ pa/lna modules for that. But they tend to be a little problematic and need extra care to work on long range. (Such as a clean and stable power source and additional shielding because most of them aren't shielded)
I never tried the more expensive shielded nrf24l01+ pa/lna modules but it is on my list.Looking at the flutter they are using something in the 915/886MHz range to communicate over long range (xbee?) also the boards seem to be partially shielded, which is a good think.
There are pro and contra on using a lower freq. to communicate:
Pro: More stable on higher range with less energy
Contra: Less Speed (Means: should be good to transmit sensor data etc., but if you want to transmit video or pictures you are probably out of luck) -
I have not had much luck with the nrf24l01+ pa/lna modules. They only seem to transmit about 300m. I'm going to setup an outdoor test this summer to see if I can get them to transmit further.
Are you getting 1000m out of them?
I've never tried the shielded modules. I didn't even know they existed. I guess I will throw them in the test as well. I will still be highly surprised if I can get 800 meters out either one of them.
Thanks for the info!
-
I have wondered about putting a solar powered repeater in the middle...
-
I have not had much luck with the nrf24l01+ pa/lna modules. They only seem to transmit about 300m. I'm going to setup an outdoor test this summer to see if I can get them to transmit further.
Are you getting 1000m out of them?
I've never tried the shielded modules. I didn't even know they existed. I guess I will throw them in the test as well. I will still be highly surprised if I can get 800 meters out either one of them.
Thanks for the info!
@Ironbar said:
I have not had much luck with the nrf24l01+ pa/lna modules. They only seem to transmit about 300m. I'm going to setup an outdoor test this summer to see if I can get them to transmit further.
Are you getting 1000m out of them?
I've never tried the shielded modules. I didn't even know they existed. I guess I will throw them in the test as well. I will still be highly surprised if I can get 800 meters out either one of them.
Thanks for the info!
I got 1000m on clear line of sight out of them, yes. Its a matter of shielding, clean power supply and correct antenna alignment.
@Lawrence-Helm said:
I have wondered about putting a solar powered repeater in the middle...
Is doable, but you will probably need to use a rather big solar cell and battery. I'm currently building some nrf24l01+ nodes with 20 watt solar cells. I think i can give some results in about 1 month.
-
If need the higher data rate, a repeater might not solve your problem since it cuts practical data rate at least in half (probably more due to contention and increased round trip times)
@mfalkvidd said:
If need the higher data rate, a repeater might not solve your problem since it cuts practical data rate at least in half (probably more due to contention and increased round trip times)
Well.. still faster then rfm69. :D Actually i never tried whats the highest datarate i can achive on repeater mode. I do not need it yet and will probably try to switch to the esp8266 if i ever need it.
-
Hi.
Like others said, for longe range, I bet on rfm69 too. and it's mysensors compliant. I think if we would use radiohead lib, lora rfm95 is very interesting for some outdoor application. But I think using radiohead as it is, is maybe too much memory consuming for what we need but it looks well written...I don't have time to, so pure speculation..but I'm very tempted to test some lora modules in future because I will have some outdoor projects. And I have seen this very nice module recently, for gw or other thing (but not arduino compliant yet). They claim a very impressive range, and it's multi radio...things are going crazy, so good for us :) :
https://www.kickstarter.com/projects/1795343078/lopy-the-lora-wifi-and-bluetooth-iot-development-p
Sorry for this little OT...just my 2cents :) -
@scalz If I find the time I will look into making a atmega1284p based node. I know this works with the MySensors lib (I made my own arduino with this mcu), and I'm also interested in building a node with lora radio's.
The matter of protocol and radio type is a very stimulating topic in the City of Things community here in Antwerpen. The current gateways in the living lab (our city is an open lab for IoT now) have lora radio's (among others). So it would be nice to be able to use the Radiohead library.
An atmega1284p has plenty of memory and the picopower version can still go very low power. -
@GertSanders : that is very good news ;) I have some 1284p at home too, moteino mega homemade. I have already designed something for this chip +rfm69 too, but pcb are not ordered yet...I don't have lora yet. I thought too about adding the lora footprint, fun! Lowpowerlab has already a lora version ;)
I don't know if you have seen it, 1284p at ali are cheap, I have some and tested one for the moment but works well
for radiohead lib, yep it would be great, but it requires some stuff on code part, and you are right 1284p could be a good candidate! -
@tbowmo: yep, I completely agree, I think it would be a very very cool thing. I wanted to make one, but no time to test everything unfortunately..(but I have all the components!), so I am waiting after your work ;) ). So for the moment I have just designed only one thing for 1284p.
So why not sam L21 as it is picopower? In other hand, for the moment it is a little bit more expensive than aliexpress 1284p. But that is not the same power!! And if bought at Mouser, sure the difference is not big!
Sure I prefer atsam ;) I let you doing it, you are more skilled on this than me! plus I have so much things in progress, hw/sw, and trying to make a mobile app that make me some headhache! ouch I need some pause!! -
Another option could be atmel SAMD... Dev branch has support for it.
I'm considering making a node with a D20, which can go up to 256kb of flash.
@tbowmo I also considered the SAMD, it's just that I received some atmega1284p from atmel (SMD samples), so I thought of doing a board for those first.
My first design for this mcu was for the 40 pin DIP version. I use this for all sorts of demo's and tests at our codedojo. I plan to update this design, and try my hand at smd.
I received these from atmel, but I realised too late that they would be impractical to handsolder. I would love to use them.
-
@tbowmo No, I got these as samples from Atmel. They were nice enough to send me samples so I can test with them. Could not find any saml21e18 or samw25xxx on Aliexpress.
-
@tbowmo: no, I have not seen atsam at ali.
I was saying that- 1284p 2.6€ at aliexpress or 7.4€ at digikey for instance.
- atsaml21e18 6.39€ at digikey.
atsaml21 make me curious :) I hope they will lower the price.. When I see at lowpowerlab, there is moteino mega 1284p. I imagine that the next 32bit "move" could be atsaml21 if we compare the price..Something like neutrino but with atsaml21for picopower+radio footprint could be very cool! this is the idea I wanted to do but not enough time..and with a good power supply of course :) With atmel uC, we are rather sure that it will be arduino integrated faster. But there are lot of other promising things in future, so I am a little bit lost for the moment!! when you see esp32 or this LoPy...and I think for basic temp node things, 8bit is better :) but for more advanced cool things, 32bit +picopower...
@GertSanders : very nice board :)
-
About 8bit vs 32bit, it depends on price. I would take the cheapest one available.. and it seems that it could be the cortex M0+ family (samd2x / L21 etc.)
if you check mouser prices, 1 pcs. of atmega 328p-au (32pin tqfp) the price is 3.52$ for 1 pcs. While the atmel samd20, also 32pin tqfp, and 32Kb flash (so same amount as in atmega328p) is priced at 2.74$ for 1.
So in this respect the 32bit processor is winning in price. I have not looked at power consumption between the two, but as we are sleeping most of the time anyway, it might be comparable..
-
@tbowmo, @GertSanders
I wanted to add to mytinycaml board few improvements. And as it is something much more for telemetry, why not add LORA ;) Here my plan:- keep ulpnode part
- footprint for nrf24||rfm69||rfm95
I have found "standalone" libs for uno/328p but like we said, it is too limited. So I'm looking between 1284p and ATSAML. As I'm in ulpnode land, SAML is pico and SAMD some uA (seems the same thing 328vs328p). Hopefully, they are pincompatible (1 pin differs: vcc instead of io). My dilemma for the moment is:
- I'm trying to keep the same size (50x23) with all on top layer and 805(or 603 max). I want to keep this one handsolderable
- I need to add JTAG (and there is already a big MYSConnector...). Too bad to need an ICE3, I need to find a cheap one.
- I know 1284p, and arm0..not yet. but 1284p is bigger footprint. So I should stop to be afraid lol If I go for arm0, will you be ok to check my arm0 "typical application schem part" ??
Or maybe do you prefer teamwork? Problem...is I am eagle addict..
