Thanks for the valuable input, I agree that there are better options than LoRa for underwater communication but I still think the transmission could be a few meters:
and strongly depends on the conductivity of the water.
You could transform “think” in to “know” in the space of a morning …
Only by buying expensive gear that won’t remotely satisfy the actual need.
This whole thread and contemplation of LoRa is a waste for the asker’s purpose.
I was trying to get it to reach a conclusion with the use of a device, a gateway and a bath tub.
That would require investing in a device and gateway that aren’t useful for the actual goal.
You could do it if you like, it makes absolutely no sense for the asker to waste their money doing so.
(Yes, one could run tests with two nodes and RadioHead… but the greater principle here is applying some common sense to recognizing grossly unsuitable ideas before sinking time and money into them)
@cslorabox thank you! well spoken.
Just for the interested reader, I’d like to share my literature research for underwater lora applications:
-
TTN forum post’: Underwater range of LoRa
-
Transmission tested in the Dishui Lake in Shanghai with relatively high conductivities of 0.24 S/m (transmission maximum was 3m): https://sci-hub.se/10.1109/APMC46564.2019.9038666
-
Blog entry with a formula to estimate the damping under water: https://smartmakers.io/en/lorawan-range-part-2-range-and-coverage-of-lorawan-in-practice/
-
From minute 14:00 on, gateway placed above water 1m @SF12: https://www.youtube.com/watch?reload=9&v=ZK-ttu5xrIE , minute 17:50 it is suggested to equip a fish with a LoRa tracker and just get the signal if it swims on the surface.
-
Internet of fish: http://folk.ntnu.no/alfredse/Forslag%20til%20prosjektoppgaver%20hoesten%202019.htm
So to sum up, there are proof of concepts up to 3m. BUT: I am still unsure what RF communication is suitable then… :).
I know this is a LoRa-forum but eventually you can still recommend me sth.
Dear TTN-fellows,
I’ve recently opened a thread about small-sized LoRa modules for under water communication:
I haven’t completely dropped the idea of using VHF and UHF bands to bridge a couple of meters in water but I’d like to emphasize here again that VHF/UHF radio modules and under water communication is not a good match.
Anyways, on my search or lower frequency modules I came across this series from hoperf:
RFM98PW
https://www.hoperf.com/modules/lora/RFM98P.html
The model number RFM98PW-169S2 supports 169MHz at 27dBm.
Has anyone of you experience with these modules and in particular what gateways can be used for those?
Thank you guys!
In theory, most gateway based on SX1302 should support it since it is usually paired with a SX1250 as RF front end which cover the band 150-960 MHz. But the antenna matching is typically targeted for 868/926MHz, so I have no idea what kind of performance you can get with off the shelf gateway …
Are you aware of wize?
Kind regards
VHF isn’t going to make enough of a difference to make this idea workable, and your costs in using atypical hardware will go through the roof.
Please stop chasing dead ends; if you need to get a signal out from under water, you’re in the regime of kilohertz to at most of few megahertz, or more likely not using RF at all. And in any case you’re talking to a very nearby pickup antenna.
If you have some other scheme for getting from the shellfish beds to a box on a pole on the shore, then you could possibly consider LoRa from there, though realistically given all the investment in gear for one small area you might as well just put a mobile data modem in it and not suffer the very extreme limitations on how much data you’d be able to send from a single LoRa device.
What are the rules for using that band in your area ?
That there are node devices available is of little advantage if there is not enough bandwidth available to run a multi-channel gateway.
169MHz might sound “low-frequency” compared to regular LoRa or BLE or so, but when it comes to under-water RF communication it’ still quite high. Attenuation in water is something like 0.0173*SQRT(f) in dB/meter. This gives around 200dB/m attenuation - no point. Submarines use frequencies in the order of 10kHz for RF communication and the above is the reason why. 
Erik Lins
https://www.chip45.com
2 Likes
@cslorabox as you say, I’ll probably end ‘chasing dead ends’ and see for a solution to pull out the antenna from the water. Anyways, we’ll also do some testing together with a university and I hope to get back with some quantitative (while most likely not surprising) results.
@chip45 that’s a nice selection of products you have there!
@LoRaTracker I’m based in Germany, the band can be used from what I know: https://www.elektronikpraxis.vogel.de/module-fuer-das-neue-169-mhz-ism-band-a-112416/
@jfmateos I didn’t know, thank you, I’m looking into it right now.
@
There are similar looking allocations in the UK, although I dont read German.
That frequency could be used for point to point LoRa. But thats not the point I was making, remember this is the TTN forum.
When I checked the detail in IRE2030, it did not seem practical to set up a proper TTN node\gateway arrangement, not enough bandwidth available and duty cycle allowed is mostly 0.1% or 0.001%
How much bandwidth\dutycycle is available in Germany, assuming you do read German that is ?
I’m wondering how LoRa actually can work at 169MHz. The channel bandwidth here is 25kHz only and LoRa usually requires 125mHz minimum…
Erik Lins
https://www.chip45.com
I assume you are confusing that with 125Khz bandwidth - and that is to confuse LoRa wirh LoRaWAN. LoRaWAN ia an instantiation of a network standard, protocol, architecture etc. that builds off the LoRa RF physical layer implementation and they are not the same thing.
@ecosoph 169Mhz is commonly used IIRC in Smart Metering applications and similar low power, low data rate applications - esp in S.Europe. 169Mhz is a band available in many area of the world. It has an advantage over the higher UHF bands (863/868/905/915/926) etc in terms of range and also for into building (from external infrastructure) penetration and ability to penetrate into in ground metering boxes and water meter pits & e.g. water distribution infrastructure monitoring points.
For the reasons (of physics) above it is better than 868/915Mhz bands for small amount of water penetration…but not by much, as pointed out above.
LoRa devices - either running LoRa modulation or classic/legacy modes like FSK/(g)FSK are used in many applications in that band and at volume by such meter manufacturers. If you look at the device data sheets for many of the LoRa (node focussed) Si devices they include suppport for such frequency bands. As noted, however, at such lower frequencies the available bandwidth and channels are similarly & significantly reduced thereby further limiting the max data rates available to end applications. A fact that I believe was a consideration when the LoRa Alliance establisted LoRaWAN as a standard targeting (initially) the higher UHF bands. Also note the RF front end devices (SX125x) that support the LoRa ‘Gateway’ baseband processing silicon (SX13xx) also have versions with options for such low bands -) 
LoRa & LoRaWAN have found multiple uses at the water/air boundary - where devices may submerge a few cm and/or get splashed, esp where they act as surface data comms devices for sensors tethered deeper down or closely attached below the water line (Swimming pool water temp - and even water quality monitoring devices, back-up water vests/jackets, water sport trackers, etc.), however, deeper water penetration applications (Meters+) are not an option unless devices ‘surface’ periodically to send any captured telemetry.
1 Like
Some of the LoRa chips can operate at VHF frequencies and can be programmed for narrower bandwidths.
However, this forum is about the implementation of LoRaWAN specific to TTN.
The bandwidth you can use (reliably) is related to a percentage of the operating frequency.
It’s more a regulatory issue
It is true that an SX127x won’t let you set 250 or 500 KHz on low band - but it will let you use 125 KHz, which may not be legal in a given place.
And at a high frequency and narrow bandwidth you could have frequency accuracy issues, though with good oscillators it is possible.
@cslorabox @LoRaTracker @kersing @Jeff-UK and @chip45
Thank you for contributing to my unusual question, I have a pretty clear idea now of the under water limitations and you pointed me to alternative solutions like 169MHz point-to-point communication.
So from my side, thank you, I’ll not bother you with more questions on LoRa underwater.