Anyone using/tried the E28-2G4M27S 2.4Ghz LoRa SX1280 27dB module?

[Larson]

@NeverDie Here is my report on the Kevin Darrah Trigboard (V8.1) as promised. Fundamentally, the power draw of this board is about as low as the TPL5110. His board ships with door monitoring firmware on it that is conducting some kind of polling every 320 mS, it appears. Between the polling the current draw is about 180 nA, per my PPKII. If one were to include the polling activity, then the average current is about 1.3 uA. A picture of this is below. If an alarm condition is achieved, the device makes an internet connection and broadcasts a message to third-party servers that will alert the user by phone/text/email. Of course, that takes time (between 2 to 3 seconds), and additional current.

If you have 30 minutes, Kevin has a start-up video that is very informative. He makes the point that this board consumes a small fraction of what smoke detectors use. While the PPKII shows that the Trigboard has an equivalent sleep current, the Trigboard delivers the monitoring function for additional, though modest, current. As before, I find value in comparing these tiny currents to the expected shelf-life burn rate that is orders of magnitude (at least one) higher.

This was my first exposure to ESP32’s. Kevin indicates that the performance of the ESP32 is better than the ESP8266. I like the idea of Bluetooth access for being able to configure devices, as Kevin shows, and for access to sensor data – especially in development & debugging. I have not reviewed his firmware that he generously has made available. His “Configurator” firmware looks to be a great example of how to build user menus to input variables. This looks like the new form of the older WiFi Manager libraries which have been very useful.

[NeverDie]

@Larson I haven't studied Kevin's solution, so maybe I'm missing the point of it, but the adafruit TPL5110 BoB is so much less expensive and draws so little current that it seems like no contest to me.

[Larson]

@NeverDie You have a point. The TPL is also simpler.

But bless the developer. Yes, I was trying to hype his product. I forgot to mention that doing this current test was really easy because this thing just worked... right out of the box (really, out of the anti-static bag.) Rarely have things work so quickly for me. Even the TPL required fumbling.

So if sleeping is the objective and timing isn't really important, then the TPL it is. But if configuration settings, Bluetooth, ADC, I2C, SPI and timing are relevant, then the Trigboard is the way to go.

[Larson]

@NeverDie said in Anyone using/tried the E28-2G4M27S 2.4Ghz LoRa SX1280 27dB module?:

https://mcudude.github.io/MiniCore/package_MCUdude_MiniCore_index.json
It's very easy to use. You can install it into the regular Arduino IDE, pick from among the MiniCore "boards" in the board manager, select the 8Mhz option and a few other obvious options, and then you're done with instalation. From that point on your code will automagically compile using MiniCore. Just to be sure, I gave it a try myself, and I'm now blinking a blue LED off of Ardino Pin 20. It works!

I leave this post as a reminder to anyone else reading this thread. The MCUDude link and installation step is VERY important. To install in the Arduino IDE add the link in File/Preferences/Additional Board Manager URLs; go to Board Manager and filter on minicore, click on Install when MiniCore comes up. Next when selecting the MiniCore board, make sure to select Clock as Internal (8 MHz), or avrdude won't be able to find your board. I went with the other board manager defaults and ... WaLa... my new barebones board is blinking on both pins 20 and 21. Overlooking, or not remembering, this May 14 post has cost me several weeks. So if you build yourself a barebones, heed this post.

[NeverDie]

@Larson said in Anyone using/tried the E28-2G4M27S 2.4Ghz LoRa SX1280 27dB module?:

@NeverDie said in Anyone using/tried the E28-2G4M27S 2.4Ghz LoRa SX1280 27dB module?:

https://mcudude.github.io/MiniCore/package_MCUdude_MiniCore_index.json
It's very easy to use. You can install it into the regular Arduino IDE, pick from among the MiniCore "boards" in the board manager, select the 8Mhz option and a few other obvious options, and then you're done with instalation. From that point on your code will automagically compile using MiniCore. Just to be sure, I gave it a try myself, and I'm now blinking a blue LED off of Ardino Pin 20. It works!

I leave this post as a reminder to anyone else reading this thread. The MCUDude link and installation step is VERY important. To install in the Arduino IDE add the link in File/Preferences/Additional Board Manager URLs; go to Board Manager and filter on minicore, click on Install when MiniCore comes up. Next when selecting the MiniCore board, make sure to select Clock as Internal (8 MHz), or avrdude won't be able to find your board. I went with the other board manager defaults and ... WaLa... my new barebones board is blinking on both pins 20 and 21. Overlooking, or not remembering, this May 14 post has cost me several weeks. So if you build yourself a barebones, heed this post.

Thanks for your post. I just now pasted it into the project's main description page on openhardware.io: https://www.openhardware.io/view/22651/Version-30-atmega328p-test-platform

[NeverDie]

@Larson
I suppose the design could be changed to include an optional diode on pin 13 for those who want to stay strictly orthodox Arduino.

One could also allow a crystal oscillator to be installed, for those who want that as well, but I happen to think running from a crystal oscillator is generally a bad idea for a battery powered application, especially when the 8Mhz resonator seems to work so well.

[Larson]

@NeverDie said in Anyone using/tried the E28-2G4M27S 2.4Ghz LoRa SX1280 27dB module?:

...but I happen to think running from a crystal oscillator is generally a bad idea for a battery powered application...

No, don't change the design. I think you have a platform with a specific low power intent - including the education of others (like me). Matter of fact, I've got several battery projects that would be FAR better by changing to the on-chip clock, and new core. And so far, we are only using the 8 MHz setting. In the MiniCore the choices go to 1 MHz.

Back when I was... on PIC's... I bought 32 KHz crystals to minimize power. I just didn't know enough to use them and I still don't. But I'm working that direction and you have helped A BUNCH including you're idea of friends sitting around the table exchanging bits and bytes in a MISO/MOSI/SCK kind of way. I'm gathering that if you have control of the clock, you could have it all.

Besides you, Kevin Darrah, and Felix @ LowPowerLabs have also made great contributions to the low-power idea. I just haven't dedicated enough energy, yet, to their efforts.

16 MHz is impractically fast when so many designs just need low power.

I wanna go back to school,
DogWithA_Bone

[Larson]

Here are the pin labels for the V001 barebones board for mounting on headers. The document I prepared for this wouldn't upload, but had nice lefthand/righthand orientations. This made it easier than having to open the CAD file everytime I needed to make connections.

But here is what I did in MS Word:
Turn on grid set spacing to 0.1"
Turn on ruler
Copy the below text
Align to the right or left as you wish
Font Calibri 5.5
Verify spacing (14 pins = 1.4")
Print

I also made columns and multiple copies of the below with both alignments. I have two boards built. Maybe you have more. Hope it helps.

GND
D5
D7
D9
D10/SS
D11/MOSI
D6
D8
A0
A1
VCC
GND
A5/SCL
A4/SDA

VCC
D4
D3
D13/SCK
D12/MISO
D2/INTO
A5/SCL
A4/SDA
A3
A2
VCC
GND
A5/SCL
A4/SDA

DTR
RX TX
TX RX
VCC
X
GND

[8/6/22 edit: reversed TX and RX as noted above - just incase someone has use for it]

[NeverDie]

@Larson

[Larson]

@NeverDie Yea, I think I knew you upversioned the silkscreen. I think I remember you commenting on it. Looks pretty. BUT in my effort to add to the community, since I typed it for my self for my V001 boards, I thought I'd share it incase there is anybody like me with V001.

I like the V003 upgrades, specially that reset bypass via as discussed with @alphaHotel . Maybe I'll try it. I've got so much into my little V001's (yours actually) that I just can't give-em up. When I get the boards rigged up with radios, GPS (on the USART bus) and SD Card reader (on the SPI bus with a different CS), I'll shoot and send a picture in the spirit of a party!

[Larson]

@Larson Okay, No picture party yet. But, I did get a notification from Digikey that the Atmega328P-AN is back in stock for $3 each.

[mfalkvidd]

@NeverDie today I re-read the fcc rules and I may have mislead you back in https://forum.mysensors.org/post/111853

It seems like the fcc rules say that if you use more than 500kHz bandwidth, you don't need to use frequency hopping, and there is no dwell time requirement.

(2) Systems using digital modulation techniques may operate in the 902-928 MHz, 2400-2483.5 MHz, and 5725-5850 MHz bands. The minimum 6 dB bandwidth shall be at least 500 kHz.

I don't understand why fcc encourages applications to hog a big portion of the spectrum but it looks like you can go willy-nilly if you just use enough bandwidth.

[SMH17]

2.4GHz LoRa modules offer a good data rate compared to lower frequency ones that are very limited on this aspect, but they suffer path loss much more and wide use of 2.4GHz increase interference problem a lot.
I'm interested to know how they perform in typical urban use scenarios.
Has someone tested E28-2G4M27S in urban environment?
I have found only test result of people in very good condition as line of sight with no buildings or other obstacle interposed or at very short range (from a room to another in same building).

[SMH17]

@NeverDie I have noticed that in the latest version of your adapter you have lowered the capacitor value used for the first stage filtering from 100uF to 10uF, leaving the second stage one to 0.1uF. Did you have empirical benefit in tests with this configuration? 0.1uF it seems to me at first glance too small to accommodate current draw fluctuation of this module. Would be appreciated if you explain your design choice. Thank you.

[NeverDie]

@SMH17 said in Anyone using/tried the E28-2G4M27S 2.4Ghz LoRa SX1280 27dB module?:

@NeverDie I have noticed that in the latest version of your adapter you have lowered the capacitor value used for the first stage filtering from 100uF to 10uF, leaving the second stage one to 0.1uF. Did you have empirical benefit in tests with this configuration? 0.1uF it seems to me at first glance too small to accommodate current draw fluctuation of this module. Would be appreciated if you explain your design choice. Thank you.

The 10uF isn't for filtering. It's just to prevent voltage droop at the radio when the radio starts to pull current. So, yes, testing reveals that it seems adequate for that purpose. I originally slotted in 100uF because I wasn't sure, and overkill is better than underkill. It's possible that 10uF may also be overkill....

[haxn2]

@NeverDie @Larson Have you had success with the Ra-01SH? I'm having trouble using high SF values (11, 12) with low bandwidth (<250 kHz). I wanted to know if you had the same experience and if you have other recommendations for sx126 modules if I just want really long ranges.

[NeverDie]

@haxn2 I don't recall there being any problems of the type you describe. You didn't say what kind of "trouble" you were having, but, if anything, the high coding factor and narrow bandwidth should improve range, unless there is interference in the narrower band. Have you tried changing the frequency? Are you sure you're using a suitable antenna? Exactly what kind of range are you trying to achieve?

[mfalkvidd]

@haxn2 https://github.com/jgromes/RadioLib/issues/388 might be useful

[NeverDie]

^^^This. Good point. Maybe do some testing with RadioLib to see whether or not you experience the same problem.