What is your experience with lora boards?

I’m not sure if this topic already exist or not (if yes, please share the link). but i do like to know your experience with lora boards, i mean which one you like and what feature did you want your board to have.
for me i don’t have a lot to share, i just used TTGO lora32 with oled v2.1 months ago and i hope i can catch up what i have missing since then.
many boards in the market now but when it come to choose which one is better (low power consumption, high speed, long range), many ppl started with TTGO and me one of them, and they reach a point when they complain about the range, 500 m or maximum 700m, i don’t know the other board what range they can reach, hope you can share this information. and about the cpu, i do like the esp32 in TTGO, i mean i can use wifi (for OTA updates) and it’s high speed 240Mhz with its dual core. i mean it’s a good board, but what about the STM or feather (M0, 32u4) and murata or pycom, etc , what you guys have used and which one you like.

Well in practice there ought not to be much difference, in terms of distances, between any lora board. They all use the same Semtech lora device, and there is little the software can do to affect performance, as long as the power level is correct.

Possibly small differences in the effectiveness of the matching between the lora device and antenna possibly.

In terms of ‘speed’ then there is not a lot of difference in real world throughput between an allegadly 240Mhz ESP32 and an 8Mhz Arduino Pro Mini, the lora device is the limiting factor here.

The ESP32 does consume masses of battery power so is much in favour with battery manufacturers. Also good if you enjoy complexity.

I start from another place when selection hardware. Specify your requirements and then search for the best match. Buying a board at random is nice to gain some experience but won’t work if your are going forward.

Asking others about their favorite will provide some information but it is doubtful you will get a good match to your requirements.

Some things to keep in mind for criteria:

  • prototype friendly or custom pcb required
  • programming environment used and related, programming language
  • antenna connection, not for bare chips but applicable for modules
  • power requirements
  • type of power source, rechargeable or not
  • LoRaWAN stack in a module or roll your own (based on loramac, LMIC, …)
  • sensor/interfacing requirements

I can write a raving story about RM186 modules from Laird (build a commercial solution with them and use them for workshops), but I know others on the forum found them too limiting for their use case after investing a lot of effort building their solution. So my story won’t help you if our requirements are vastly different.

3 Likes

i have worked with different microcontroller like Atmega (arduino), STM, ESP32 (TTGO) and others, but the thing i like about the esp32 is like it’s a mix between stm speed + arduino easy coding plus wifi with the many GPIO and interfaces. like TTGO, Pycom heltek, , TTGO is better then heltek (at have better antenna) i have no idea about the Pycom, just many people are using it and recommand it.
for the STM board, some people are complaining about the lack of documentation and not friendly use.
for the atmega, well it’s just an arduino for me XD, in big project you may face to many problem (memory or frequency problem) which lead you only to swap to other cpu. (probelm not related to LoRa)

actually i don’t have a Good vision about the project i’m going to build, i do like to be open to all kind of projects and when i buy a board i try to buy one that allows me to do as much project as i can. this is why i’m trying to have people feedback about lora boards and have read their experience. if i have to chose a key features then i do like to have a board that consume less power, with the ability of using a solar power to recharge the battery (optionnal), and for the sensors they are all degital, i mean SPI, I2C etc, and for the language it doesn’t matter acutally as long as the board is well documented, and of course support LoRaWan (i have read some boards only works lora point to point which is shocking to buy a board only for that) , to make things easy for me, i will compaire each board to the other and i will see what is better for me,
and yes i do like to read about RM186, that would be nice

Hi,

I have used a few Whisper Nodes.Good impressions. Also it has 4MBit SPI Flash on board. It is optimised for low power consumption and it can be powered by batteries below and over 3.3 V.

Thanks for sharing, i just did a quick look for the node, and it looks decent.
some people said it needs some soldering, could you confirm this information? and what about the range?
for the last couple days i was reading about different LoRa node, to be honest with you, they are almost all the same, but what was interesting is that the RFM95 isn’t that good, TTGO is better, i know all of them are based on SX127X (new with SX126x) and whisper uses RFM95 so based what i have readh TTGO may consider better,
the thing is, i’m kinda confused about the huge number of lora board in the market.

  • adafruit with the feather M0 and feather 32u4
  • TTGO with lora32 oled and T-beam
  • laird with the RMxxx, it require it own language (hope @kersing can confirm this information since he worked on this one)
  • microchip lora module (ATSAM…) you can find it in Penguino Feather
  • B-L072Z-LRWAN1, STM32 , uses murata chip,

i think this what i have found and maybe more, and almost all of them are the same. i mean in LoRa specification not the MCU. but i’m not really sure about that, because some company promise a range of 15KM like laird but i have no clue if it’s true or not, and other promise a TX power = 20 dBm which after tesing it, it was barley reach 10 dBm like ttgo and heltec,
i watched a lot of Andreas Spiess videos and it was helpful, and one in his videos said hopeRF is better then TTGO, and TTGO is better then heltec, and the new heltec is quite good now after use the new lora chip, SX126x but documentations and libraries are not quite enough to chose that chip now for someone like me to start with, so i’m kinda confused about which one is the better, i mean i term of quality and numbers (board specification being true) and do they all support LoRaWAN (am i gonna face problem to use class A or class B or C) or am i not be able to use LMIC library (the only one i know it supports LoRaWAN).

Since most all current LoRa nodes are based on the Semtech SX127X device, the range\distance performance should be very similar.

There is often a lot of heresay and annectdotal ‘advice’ that one board performs better than another, but to get to the truth you need to do direct signal performance comparisions in a controlled environment.

that’s a way to know the performance of each board, and other way is to hear from people their opinion and what they like and didn’t like about the board they have used, this is why i didn’t want to name the topic “what is the best board” i just wanted to know other peoples experience and story about LoRa board, their favorite board and what they are using now and from that i can build a good idea about LoRa boards, i have read many conversation in many forums and watched many videos and what i have said in the previous comment was the result of those informations, more questions hhhh.

The problem is that more than half of reported difficulties trace back to a user mistake

1 Like

This module uses SmartBasic and has a certified LoRaWAN stack build in. SmartBasic limits usability in some ways but the build in LoRaWAN stack allows you to avoid all the pitfalls that come with the use of lmic, a big plus in my book especially if you are not a hardcore embedded developer.
It also allows for really fast prototyping and very low power usage designs. The module has a uFl connector to connect an antenna and has excellent range with decent antennas.

I prefer this module over esp32 based options because it has very low power use in sleep mode without the hassle of having to adapt a LoRaWAN stack to keep session variables save in non volatile memory.

2 Likes

0.4 uA for the dev board, that’s impressive, and with that kind if specification, i think i see why you like it.

Well, in a way it is impressive.

But very low sleep currents are more of a marketing issue.

If the sleep current was a ‘massive’ say 25 time greater than that at 10uA, then a set of AAAs would still have a battery life of 11 years.

So the reality is that very low sleep currents may not make a lot of difference in practice.

i get it, but promising the customer with 0.4uA and i get that 0.4uA in real practice or even close to it, won’t be a big problem as long as the battery life will be the limit, but as other board (like TTGO) saying 0.2uA in deep sleep while some people hardly get close value to that, (i’m sorry, i forget the exact value), i think it’s a problem, i hope i’m wrong in this one because i do like TTGO board but i don’t trust their specification, and you can’t just buy and test all boards to chose the right one,(wish i can do that XD ).

Which TTGO board claims a deepsleep of 0.2uA or less ?

1 Like

T-Beam and lora32 oled

Ah, using the power management to power everything off.

Having to restart from afresh everytime wastes a lot of battery power, especially with an ESP32.

A setup which allows the processor to resume from the point at which it was put to sleep will normally be more power efficient since this avoids all the startup and initialisation processes.

but, does it really reach the 0.2uA??

No idea. Don’t have one of those boards.

may i ask you the board you are using?