Started By
Message

re: Which one of you made $10k last month mining Helium?

Posted on 12/23/21 at 1:08 pm to
Posted by Korkstand
Member since Nov 2003
28738 posts
Posted on 12/23/21 at 1:08 pm to
For those still following along, I have begun prototyping end devices which use the network. There are a lot of commercial products available, but they're pretty pricey and not very flexible. I'm trying to settle on a base DIY platform so that I can just swap sensors and a bit of code and deploy custom devices pretty quickly.

Here is what I have so far:

Arduino Pro Mini - $5
LoRa radio - $10
DHT22 temp/hum sensor - $4
LiPo 1000mAh battery - $4
USB battery charger - $3
Misc wire & other components - $1
Total - $27

Parts were purchased in multi packs of 3-5 units to keep unit costs reasonable, however these are Amazon retail prices and unit costs can come down considerably if bought in bulk from overseas. I still need to figure out enclosure options, but with a few more dollars for that I think I can still get my total cost per device down in the $15-25 range, depending on options.

The battery charger setup works very well. The device will run directly from a standard micro-usb cable, and when unplugged it just keeps running from the battery until it can't anymore. I'm using PlatformIO with VS Code to program, and I'm using sample code that I found and modified slightly. I have a battery life test in progress right now. It is transmitting every 10 minutes and will do so until it dies. I will do it again and transmit every 5 minutes until dead. After that I think I can do a little algebra and figure out how much battery is consumed per transmission vs. when the device is idle. Then I will know which parts of the code need to be optimized first.

Here is the data I've collected so far:



The DHT22 collects temp and humidity data. The device is in my home, so the temp is a pretty steady 20C (68F). The jump in RH from ~40% to ~50% was when I moved the device from one room to another, and I had no idea it could vary that much room to room. On the battery chart, it was a steady 3.5V for a while, which was when the device was hooked to the programmer on my laptop. Where the voltage starts rising is when I connected the usb charger and battery to the device. Once it hit 4.3V, I unplugged it and it has been running on battery power since then. After about 16 hours, it's down to 4.1V. You can see that it dropped quickly at first, but has since leveled off to a lower rate of discharge. That is typical of lithium batteries. I expect the device to run until voltage drops to around 3V. Might take a week or so.

The sensor sends a very small packet of data, about 11 bytes stores temp, humidity, and battery voltage information. The hotspot adds more information, such as the RSSI and SNR values shown above. RSSI is the Received Signal Strength Indicator, which is basically the same as a wifi signal strength icon. The value is how "loudly" my hotspot hears the device. LoRa should work down into the -120db range. The drop from -55 to -75 corresponds to when I moved it from one room to the other. A pretty big difference, but as it turns out I moved it from a location with 3 sheetrock walls in the way to a location with 3 sheetrock walls and 2 brick walls in the way. I expect the signal to improve once I start doing outdoor range testing. SNR remained pretty steady, which is good. LoRa should work down near -20db SNR.

If I continue sending one packet every 10 minutes, it will cost me roughly $0.00145 per day, or 53 cents per year. But if the packets go through my own hotspot(s), then I will earn exactly that much back through data transfer rewards, so it's essentially free. If the network continues to grow as rapidly as it is, I think very soon Helium devices will replace cellular GPS tracking devices. My MIL pays $100/year to track her dog. Crazy we will be able to do it for $1/year.
Posted by Korkstand
Member since Nov 2003
28738 posts
Posted on 1/3/22 at 5:17 pm to
quote:

I have a battery life test in progress right now. It is transmitting every 10 minutes and will do so until it dies. I will do it again and transmit every 5 minutes until dead. After that I think I can do a little algebra and figure out how much battery is consumed per transmission vs. when the device is idle. Then I will know which parts of the code need to be optimized first.
So when transmitting once every 10 minutes, it lasted about 100 hours on the 1,000mah pack. When transmitting once every 5 minutes, it still lasted right at 100 hours. The sensor measurements and data transmissions happen so quickly that they hardly burn any power, so pretty much the entire battery capacity was burned during idle. I realized it was never going into a low power sleep mode, so after adding a popular low power library to the code, the difference so far is pretty drastic:



The top curve is with the low power code. I don't know how long it will last, but obviously it is a big improvement. And I'm sure there are more things I could do to improve battery life.

I also managed to get a second hotspot set up in my neighborhood. This not only improved my mining rewards drastically due to witnessing, but it's giving me an idea of the coverage range of different setups for end devices. The hotspot is inside the home, and has the small stock antenna. My little battery powered devices are inside my home 1/4 mile away, and their transmissions are still being picked up with signal strength to spare. Pretty crazy.

first pageprev pagePage 1 of 1Next pagelast page
refresh

Back to top
logoFollow TigerDroppings for LSU Football News
Follow us on Twitter, Facebook and Instagram to get the latest updates on LSU Football and Recruiting.

FacebookTwitterInstagram