15 minutes = 900 seconds. 20000 Nodes / 900 sec = 9550
I you configure the NRF24 to the Maximum of RF24_2MBPS you can transmit - in theory - 2.000.000 MBit / sec = 250.000 Bytes / sec.
250000 Bytes/Sec / 32 Byte Payload size = 7812 payloads/sec. Thats less that the requiered 9550.
And if you must concentrate a lot of Clients on repeater nodes. you are totally lost. Forget it.
How many measurements would you like to get in which Intervall.
I think you will get in serious trouble because all the nodes are talking wireless and this is not able to handle more than one client sending at the same time.
If you want to do this wireless you have to sync all the activity fro a central Point. So you must have a "Server" which polls the Clients to get the data.
Can you please describe what are you planing in more detail?
Hi all,
the forcast is working as expected for a while now. It follows the pressure values from the device.
First diagram shows pressure and forcast. Looks like the calculation is ok.
But the weather doesnt know anything about the forecast 100% Humidity because it rains Dogs and Cats
This is no question. Just a little reminder about crystalballs
Andreas
Perhaps this makes clear what happend:
https://cdn.sparkfun.com/assets/0/7/5/5/1/51eec304ce395f104c000000.png
If you have a look at the datasheet you see:
All voltage input (AC: 90 ~ 264V)
110 should work.
This could be a good start https://github.com/solarkennedy/equail
Perhaps my following sketch can help you: https://codebender.cc/sketch:98928
I use the same Setup to meassure the Level of the pellet store. There is a lot of dust in the Air and no even Surface. So a lot of "noise".
The digitalSmooth function tries to find the relevant values. My graph looks much better with it
Just enable the debug define. You will get the output to the console.
Or use the wireshark sniffer.
I have no problem with it. fhem as controler and mysql as DB is running on the same PI.
The Gateway is not even listed in "top"
For my Point of view, every Controller which is able to use the Serial Gateway.
If you add a little edge around the holes, the water cannt flow into the electronic box.
The symlink alwayshas 777. Everyone can use it.
Please have a look at /dev/pts/2 This device should have the configured groups.
Ok?
? Are you using the code that is available on github now?
This behavior was Change by Holger a month ago. You can configure the Groups in the makefile:
The standard configuration will build the Serial Gateway with a tty name of
'/dev/ttyMySensorsGateway' and PTS group ownership of 'tty' the PTS will be group read
and write. The default install location will be /usr/local/sbin. If you want to change
that edit the variables in the head of the Makefile.
The LEDs mut be connected to VCC instead of Ground. I missed that too one my first attempt.
But this is not a big step in the desired direction?!
Hi, did you feed the NRF24L01 directly from the 3.7 V battery? The specs are alowing 3.6 V max?!? And you are taking about a maximum output of over 4 V of the battery.
That sound not healthy for the NRF.
Perhaps this is the best way to spy a location :
sniffer
Or it uses one of the FTDI Chips which will be disabled by Windows. Just another shot
I am using fhem on the same PI. connected over the pseudo serial port.
The original code on github is a bit buggy. Thats right. But the actual code after Holger has reworked it is running well.
I am using it for nearly a month now. Nothing has crashed after Holger has looked after the reason why.
Currently I am happy with the stability.
Perhaps Node8 uses Node2 as a repeater to send the packet to the gateway?
Hi,
no problem at all. The humidity scetch is running fine with a pure atmega328p-pu and a DHT11.
I have build it on a breadboard for you to see if it is working:
As you see, the sensor starts and sends temperature and humidity value:
sensor started, id 99
send: 99-99-0-0 s=255,c=0,t=17,pt=0,l=5,st=ok:1.4.1
send: 99-99-0-0 s=255,c=3,t=6,pt=1,l=1,st=ok:0
read: 0-0-99 s=255,c=3,t=6,pt=0,l=2:M
send: 99-99-0-0 s=255,c=3,t=11,pt=0,l=8,st=ok:Humidity
send: 99-99-0-0 s=255,c=3,t=12,pt=0,l=3,st=ok:1.0
send: 99-99-0-0 s=0,c=0,t=7,pt=0,l=5,st=ok:1.4.1
read: 99-1-0 s=0,c=0,t=7,pt=0,l=5:1.4.1
send: 99-99-0-0 s=1,c=0,t=6,pt=0,l=5,st=ok:1.4.1
read: 99-1-0 s=1,c=0,t=6,pt=0,l=5:1.4.1
send: 99-99-0-0 s=255,c=3,t=0,pt=1,l=1,st=ok:62
read: 99-1-0 s=255,c=3,t=0,pt=1,l=1:62
Normally I am using a DHT22 on my selfmade board.
Andreas
You dont Need to Strip the voltage Regulator from the board. Just use the marked solder pad to feed the board with 3.3. This works fine for me.
And dont forget to enable SPI in the PI config tool.
Out of my mind:
Change in bcm2835.h from
define BCM2835_PERI_BASE 0x20000000
to
define BCM2835_PERI_BASE 0x3F000000
and
sudo make clean all install
Change in PiGatewaySerial.cpp from
gw = new MyGateway(RPI_V2_GPIO_P1_22, BCM2835_SPI_CS0, BCM2835_SPI_SPEED_8MHZ, 60);
to
gw = new MyGateway(RPI_BPLUS_GPIO_J8_15, RPI_BPLUS_GPIO_J8_24, BCM2835_SPI_SPEED_8MHZ, 60);
thats depends on the GPIO Pins you use. These are from the sample.
hope that was all. Please come back if it doesnt work.
Just on more comment on your code.
The pseudo TTY Port you create in PiGatewaySerial with openpty creates and port with the Default rights RWX for the User and W for the tty group.
That leaves it unusable for other users than root. Even members of the tty Group are limited to write. Very usefull for this Point of usecase
I have added a "chmod(ttyname(pty_slave), S_IRWXU | S_IRWXG | S_IRWXO);"
That gives the rest of the world "enough" rights to work with the pseudo tty port.
So my Installation with PiGatewaySerial and fhem connected over the tty port ist running fine for the last day with the above modification.
Andreas
The cause for the stop is the lenght of the sketchname "Pressure Sensor". It is stored in the same memory area where the childNodeTable holds its data, So the
"if (childNodeTable[childId] != route) {" in MySensor::addChildRoute brings up an Segmentation fault.
For my point of view the following is wrong in MySensor,h:
287 #ifdef DEBUG
288 char convBuf[MAX_PAYLOAD];
289 #endif
it should be
287 #ifdef DEBUG
288 char convBuf[MAX_PAYLOAD*2+1];
289 #endif
because in MyGateway.h it is defined the same
Line 58: char convBuf[MAX_PAYLOAD*2+1];
if you look in the Arduino MySensor,.h ist is defined the same as in MyGateway,h
That would explain the memory overrun and that it work for some people because the dont define debug!
I have changed this lines and it became much more stable. But after only one day of testing I would not leave the Patient out off Hospital.
Just my two cent with very limited cpp knowledge.
Andreas
Hi all,
I try to use the PiGatewaySerial on a new raspberyy pi 2. After some modifiactions on the code concering the differnet addressing and gpio Pins it starts up and logs incoming datagrams on the console.
That works fine until a Client was switched on and sends his capabilities.
here ist the Output before the PiGatewaySerial stoppes responding:
0;0;3;0;9;read: 10-10-0 s=255,c=0,t=17,pt=0,l=5:1.4.1
0;0;3;0;9;read: 10-10-0 s=255,c=3,t=6,pt=1,l=1:0
0;0;3;0;9;send: 0-0-10-10 s=255,c=3,t=6,pt=0,l=2,st=ok:M
0;0;3;0;9;read: 10-10-0 s=255,c=3,t=11,pt=0,l=15:Pressure Sensor
0;0;3;0;9;read: 10-10-0 s=255,c=3,t=12,pt=0,l=3:1.0
The control I would like to is fhem on the same PI connecting over the pseudo Serial port.
Any idea what happens and how to fix?
Andreas