Meshtastic

TTG Lora32 V2.1 ESP32 Power Amplifier 4W

I posted the following on reddit.com/r/Lora/ ; hoping this forum can help out also:

Post:

I am experimenting with boosting the transmission power of a TTG Lora32 V2.1 ESP32 (for specs: TTGO LoRa32 V2.1 ESP32 Bluetooth WIFI Wireless Module 433/868/915Mhz 0.96 inch Bluetooth wifi wireless module SMA IP5306|Circuits| - AliExpress).

I found the following 915MHZ RF 4W power amplifier:

This would certainly push transmission, but would it be capable to transceive also? If not, does anyone know of an alternative 915MHZ power amplifier that can transmit and receive? Thank you

Edit 1

Would adding a diplexer make the amplifier capable of receiving and transmitting? Also, I know adding a PA is illegal; this is strictly research.

Edit 2

I added a schematic; would this work? I am trying to a compatible diplexer.

3 Likes

I got what you described working, with the exact same amplifier, as a proof of concept.

Code:

It uses this RF switch:

https://www.analog.com/en/products/hmc544a.html

The amplifier needs to have an attenuator to bring the TX signal from the LORA radio down to below 1dbm or it gets some nasty harmonics and signal reflections, as seen on a spectrum analyzer. I used a -20db attenuator for that.

The amplifier also needs to have a power mosfet to switch it on and off. It gets insanely hot when not transmitting. The code (from above) will need to be modified to give it a few ms to power up when it’s first turned on. It also uses a logic inverter, configured to match the truth table of the RF switch, to drive the RF switch from one GPIO rather than two. It also allows for faster switching than what can come out of the very slow Arduino digitalwrite function.

The code from above will not be merged into the main meshtastic code. It’s a bit of a hack. I could see it going into a power amplifier plugin.

2 Likes

3 Likes

I greatly appreciate your shared research. The feedback I was receiving on reddit didn’t seem hopeful. I love the picture also, but could you possibly share a schematic? I think I have an idea of how you have the parts arranged, but all the cables have me confused.

BTW, I found several suppliers for the RF switch, but had trouble finding a -20db attenuator; mind sharing a link of where you found your attenuator?

One last question…when you turn off the amp, will the ESP32 still function at its regular capacity via meshtastic?

Thank you again sir!

Sharing a schematic or diagram would mean having to make one. I threw that together in an evening. I’d like to have it (eventually) manufactured, but got distracted with other things. It’s very close to the block diagram you posted on Reddit with the attenuator coming in between the switch and the amplifier. The other wires are for power and control.

I made the attenuator myself using the Pi configuration, you’re welcome to use my design.

https://oshpark.com/shared_projects/GhuHmt0u

You can also get them premade for like $20 or $30 on Amazon.

The power amp is only used to transmit, so the RF switch would route the received signal path around the amplifier when not in use.

1 Like

Thank you for helping. I was looking at antennas as well to further improve performance. Would these antennas drastically improve performance & be worth the extra money for farther/reliable range?

  1. Fiber Glass Antenna | LoRa High Gain Antenna | Long Range Frequency Antenna – RAKwireless Store
  2. Fiber Glass Antenna | LoRa High Gain Antenna | Long Range Frequency Antenna – RAKwireless Store
  3. 8dBi Fiber Glass Antenna | Supports 900-930MHz – RAKwireless Store

I would directly connect them to the ESP32 with a N to SMA adapter (non-cable).

No experience with those antennas, but I have the 15dbi antennas from here:

They look good on the VNA but I have not had a chance to measure the actual gain.

Thanks for the share!

I have the RAK 5.8dBi which I haven’t trialed in the wild yet. It also tested very well on the NanoVNA so I would expect it to perform well.

Eliminating a cable and using only an adapter as you plan will help out too of course.

By look good, do you mean that they appear to be resonant and have a low VSWR at 915 Mhz?

At least, it’s measuring that there are no impedance mismatches at those frequencies more so than any of my other antennas.

I took the chance and purchased a pairf (got the covid relief check, so less than $20.00 won’t upset the household budget and spouse).

I’ll make some measurements, comparing them against the half wave “rubber ducky” verticals that Heltec furnishes with thier ESP32 modules using an SDR and RTLpower to see how they look before doing a field test. and post the results.

Using a 5 element yagi on one end didn’t make much of an improvement (at SF12 and minimum data packet settings), so I’m not expecting miracles.