nRF5 action!
-
What's the best way to make a "receive only" gateway? i.e. one that cannot transmit? Then I wouldn't need to worry about whether the gateway is turned on before powering up a MY_PASSIVE_NODE sensor mote. Is there a way to do it simply in hardware, or do I have to sabotage the gateway library code?
-
Hmmm... it looks like radiohead may be such a library: http://www.airspayce.com/mikem/arduino/RadioHead/classRH__NRF52.html
-
It's getting too baroque. What I'd really like to have is a short nRF5 library of just the bare essentials (like the MIRF library is for the RF24).
@NeverDie said in nRF5 Bluetooth action!:
It's getting too baroque. What I'd really like to have is a short nRF5 library of just the bare essentials (like the MIRF library is for the RF24).
You can take a look into the actual Nordic NRF SDK. There are the ESB library for nRF24 compatible communication.
-
@NeverDie said in nRF5 Bluetooth action!:
It's getting too baroque. What I'd really like to have is a short nRF5 library of just the bare essentials (like the MIRF library is for the RF24).
You can take a look into the actual Nordic NRF SDK. There are the ESB library for nRF24 compatible communication.
@d00616 said in nRF5 Bluetooth action!:
@NeverDie said in nRF5 Bluetooth action!:
It's getting too baroque. What I'd really like to have is a short nRF5 library of just the bare essentials (like the MIRF library is for the RF24).
You can take a look into the actual Nordic NRF SDK. There are the ESB library for nRF24 compatible communication.
The RadioHead library seems very easy to pickup and start using. Maybe it's me, but I can't say the same for the Nordic SDK.
Interestingly, it looks as though @Yveaux may (?) have written the nRF51 part of the RadioHead library.
-
I was able to program the tiny nRF51822 (earlier photograph above) by programming it as an xxaa Generic nRF51 with an RC oscillator.
Nice!
-
Here's a close-up photo:
-
FWIW, range on the tiny nRF51822 does seem compromised when compared against larger sized nRF52832 modules. Not really surprising, but I had hoped it might be a little better than it is.
@NeverDie said in nRF5 Bluetooth action!:
FWIW, range on the tiny nRF51822 does seem compromised when compared against larger sized nRF52832 modules. Not really surprising, but I had hoped it might be a little better than it is.
This is an idea, I don't know if this helps. The two pins near the antenna are GND. Try to solder an 3cm isolated wire to the pin near your PCB and route it parallel of your pcb away from the nRF51 board. The extends the GND pane size.
-
Just in case anyone uses the waveshare ble400 board.
The board was consuming 150uA when fed with 3.3v bypassing the regulator.
I have cut through one of the tracks and now it is only consuming 4-5uA and it can still be used normally with the usb lead 5v as long as i place a dupont link across 2 pins.
If anyone needs photo's let me know. -
@NeverDie said in nRF5 Bluetooth action!:
FWIW, range on the tiny nRF51822 does seem compromised when compared against larger sized nRF52832 modules. Not really surprising, but I had hoped it might be a little better than it is.
This is an idea, I don't know if this helps. The two pins near the antenna are GND. Try to solder an 3cm isolated wire to the pin near your PCB and route it parallel of your pcb away from the nRF51 board. The extends the GND pane size.
@d00616 said in nRF5 Bluetooth action!:
@NeverDie said in nRF5 Bluetooth action!:
FWIW, range on the tiny nRF51822 does seem compromised when compared against larger sized nRF52832 modules. Not really surprising, but I had hoped it might be a little better than it is.
This is an idea, I don't know if this helps. The two pins near the antenna are GND. Try to solder an 3cm isolated wire to the pin near your PCB and route it parallel of your pcb away from the nRF51 board. The extends the GND pane size.
I'll give it a try. Is your idea to make it like a dipole antenna?
-
I don't think sendSignalStrength function is implemented yet, but you should be able to get this info with:
int16_t transportGetSendingRSSI(void) int16_t transportGetReceivingRSSI(void)@scalz said in nRF5 Bluetooth action!:
I don't think sendSignalStrength function is implemented yet, but you should be able to get this info with:
int16_t transportGetSendingRSSI(void) int16_t transportGetReceivingRSSI(void)Just wondering what both of these mean?
I assumed the transportGetReceivingRSSI(void) was the strength of signal from my Gateway ?
But what is the transportGetSendingRSSI(void)? -
@scalz said in nRF5 Bluetooth action!:
I don't think sendSignalStrength function is implemented yet, but you should be able to get this info with:
int16_t transportGetSendingRSSI(void) int16_t transportGetReceivingRSSI(void)Just wondering what both of these mean?
I assumed the transportGetReceivingRSSI(void) was the strength of signal from my Gateway ?
But what is the transportGetSendingRSSI(void)?@rmtucker said in nRF5 Bluetooth action!:
But what is the transportGetSendingRSSI(void)?
This is the RSSI of the receiver. With nRF5 it's part of the ACK payload.
-
Out of curiosity i stuck a cake tin over the top of the node and it just carried on transmitting,so i put it behind 4 walls and 1 floor down then behind the fuse box and consumer unit and eventually got it to drop to -68dB for received rssi but send rssi stayed at -45dB so it seems to be booming out and in.😉
-
Results: I don't think it made the link worse, but it's not obvious that it made the link better either. Range seems about the same.
Even more surprising: prior to adding this piece of wire, I didn't notice much improvement when I went from 2Mbps at Tx 0db to 250kbps at Tx 4db either. I had really thought it would be a tangible improvement in range, but if there was any improvement (and I'm not sure that there was), it seemed like only a modest amount.
Conclusions/recommendations/suggestions/comments?
-
Results: I don't think it made the link worse, but it's not obvious that it made the link better either. Range seems about the same.
Even more surprising: prior to adding this piece of wire, I didn't notice much improvement when I went from 2Mbps at Tx 0db to 250kbps at Tx 4db either. I had really thought it would be a tangible improvement in range, but if there was any improvement (and I'm not sure that there was), it seemed like only a modest amount.
Conclusions/recommendations/suggestions/comments?
@NeverDie I don't know how much work is needed to make this work with nrf5, but @Yveaux has created a range tester that might be useful. https://github.com/Yveaux/MySensorsRangeTest
It does use MySensors though, so it might provide more overhead than the bare-bone functionality you are looking for.
-
@NeverDie is your wire soldered to gnd?? if so, i would have soldered it to the antenna transmission line, as a monopole, with taking care of disabling the pcb antenna. i guess you're trying sort of dipole, but one branch is meandered/"coiled" so not sure if that would improve a lot like you noticed..
-
@NeverDie is your wire soldered to gnd?? if so, i would have soldered it to the antenna transmission line, as a monopole, with taking care of disabling the pcb antenna. i guess you're trying sort of dipole, but one branch is meandered/"coiled" so not sure if that would improve a lot like you noticed..
@scalz said in nRF5 Bluetooth action!:
@NeverDie is your wire soldered to gnd?? if so, i would have soldered it to the antenna transmission line, as a monopole, with taking care of disabling the pcb antenna. i guess you're trying sort of dipole, but one branch is meandered/"coiled" so not sure if that would improve a lot like you noticed..
In this instance, I don't see a way to attach to the actual antenna. On some boards I see a little hole where a wire can be attached. On this one, I guess maybe it could be done by carefully scraping off the solder mask and then soldering to the trace.... Its a gamble though: t would be all too easy to scrape off the trace in the process.
Anyhow, enlarging the footprint of the board kinda defeats the purpose of its small size. I think maybe it just is what it is, and the relatively poor performance explains its relatively low price.
Maybe what would rescue it is an adequate ground plane on whatever PCB it attaches to. For instance, I'm thinking it might be a nice match for a "Chirp" soil moisture sensor, which maybe (I'd have to look) has a long--though narrow--ground plane. Making the Chirp wireless would be a nice upgrade. :) It has an attiny MCU, which (unless someone knows differently) isn't enough to control, say, an SMD nRF24l01. I suppose it could be redesigned to use an atmega328p (which would be preferable), but you can already buy cheap pre-made attiny Chirps from China, so there's an argument for leveraging that instead by attaching maybe this cheap wireless module to it.
