This forum has moved to forum.jeelabs.net - this read-only archive is for reference only!
Guys, can anyone explain me how to connect the RFM12B to a plain Arduino 2009?
I think i have most of the connections right, though, none of the RF12 sketches seem to work. I'm trying to receive (and eventualy send) 433Mhz signals to control KAKU alike devices.
The arduino 2009 has a 3.3V connection which i used to power the RFM12B. I'm not sure if i can connect the other "data" lines straight to the arduino 2009 pins.
Greetings,
EriSan500
You can't. You need level converters for all the signal lines to convert between 5V and 3.3V signal levels. Here's an example, but keep in mind that this approach is a bit flakey with fast I/O pulses: http://news.jeelabs.org/2009/01/13/arduino-rfm12b-µsd/
Thx, will give it a look, but will not waste to much time on this. Is there another option to send and receive 433Mhz/868Mhz signals with an arduino?
Greetings,
EriSan500
Sparkfun makes 3.3v Arduino boards, but they don't have the built-in USB port that the Duemilanove (2009) has.
Would an LM317T be overkill to step down to 3.3v? I basically have a breadboard arduino being powered by a 12v wall wart. I'm thinking I could add the LM317T just under the onboard 5v regulator to make the whole board 3.3v.
Hi
I guess you need a shield on the Arduino (ethernet?) not yet available for JeeNode?
I have connected a RF12 to a Arduino using a level-converter etc. but it is more problematic then buying a extra JeeNode and using a short pair of weirs to connect it's serial interface to the Arduino (running a soft serial port and you're code) and then using RF12demo on the JeeNode and "calling" it from the Arduino.
Unless you plan on fabricating many devices and unless you can connect the shield directly to the JeeNode (timing problems, missing pins etc.) this "double CPU" is the way to do it in my opinion.
but it is more problematic
What kind of problems can I expect to run in to?
Unfortunately I bought the components for the breadboard arduino before I found Jeenode. So I figured I'd just get an RF12 and hook 'em. I'm only using the LM317T because I can pick it up locally vs. having to wait for shipment.
The savings with the stripped down arduino+RF12 vs Jeenode are not worth all of the extra effort/problems.. so yeah, next one will be an actual jeenode.
I just attached an RFM12B to an Arduino Uno using voltage dividers to convert the signal lines down to 3.3v (10k and 15k resistors).
I havent managed to get it to talk to my JeeNode (v5) but it is displaying back ground transmissions, although they look odd compared to my JeeNode:
Current configuration:
B i2 g39 @ 868 MHz
? 0 12 12 0 0 0 0 136 0 0 248 0 12 0 0 0 0 32 8 0 0
? 16
? 8
? 12 144 0 0 0 128 0 0 0 48 0 0 0 128 192 0 0 62 0 0 0
? 0 0 0 16 0 0 192 0 0 24 32 0 16 48 12 0 0 248 0 0 0
? 0 0 0 8 128 144 0 32 0 0 0 0 0 12 4 0 12 0 0 0 0
? 0
? 16
The random noise on my JeeNode is much more uniform. Does this look normall?
Those numbers look wrong. Far too many zeros. When you do get a non 0 it's only one or two bits which are on.
What pins have you got the voltage dividers on? You don't need them on all of them, you only need them on the connections that are output from the ATmega. The inputs should be left as direct connections or you'll pull the 3.3v high pulses down even lower, and the ATmega won't see them as non 0.
So you should only have the dividers on the MOSI (Master out, slave in), CLK (clock) and the SEL (device select) lines. The MISO (Master in, Slave out) and INT (interrupt) should be left alone. The 3.3v levels from the RFM12B should be enough for the ATmega to see as 1's. I assume you have powered the RFM12B from a 3.3v source?
Yep that was exactly the problem! (I actually discovered this on my own, and was coming back to post the solution, honest!)
And makes perfect sense when you think about it.
I cant believe I actually got this working, I have never soldered anything in my life and i've built my own JeeNode and proto board in a week!
Couldn't of done it without this site and the software JCW has produced. Now I can actually finish my project.
BTW: cheers tankslappa, i also used your ATmega pin diagram to wire up my arduino and RFM12B.
Great news, well done. Glad to be of service with the diagram if nothing else!
Personally I think I would have swapped the regulator out on the Uno board and made it 3.3v. JC has made too many inexpensive, but very cool I/O boards designed to run at 3.3v!
Now you're a qualified solderer you'll be ordering the JeeNode SMD next then? :o)
(Only joking, they're a tricky little beast, I've been soldering over 20 years and they still cause me some headaches - mainly from squinting!).
Hi, I am attempting to connect an RFM12B module to my existing Arduino and will be using voltage dividers on the MOSI, CLK and SEL pins. The following page provided a great start for me: http://news.jeelabs.org/2009/01/13/arduino-rfm12b-µsd/
Now, this might be a silly question (I am new to electronics) but is there any reason why a 3.3V regulator should be used on the VDD pin of the RFM12B rather than another (cheaper) voltage divider?
Current...
The current draw of the clock, data and select lines is negligible, so the RFM12B doesn't really change the voltage created by the divider.
With the power connection however, the current demand will vary, which will change the balance of the resistor divider and cause the voltage to change. This could cause the RFM12B to crash/lock. A proper regulator will keep the voltage constant.
I'll try to draw an ASCII picture...
5v source ---[ R1 ]---+---[ R2 ]--+- GND
| |
+--RFM12B---+
Without the RFM12B in circuit, to divide the 5v down to 3.3v you just need R1 to be just under half the value of R2. You want the highest values of R1 and R2 you can get away with to keep the current demand on the 5v source to a minimum.
As soon as you add in the RFM12B, some current will go through the radio module. With the data/clock/select lines this current is tiny, so doesn't really change anything. With power however, it has the effect of reducing the resistance of the R2 side of the divider, this will cause the voltage drop across R1 to increase, and across R2 to decrease, and the RFM12B suddenly doesn't have 3.3v to run from, it will have dropped. How much will depend on the resistances chosen and the current draw of the RFM12B at that moment.
Thanks tankslappa - yep, the RFM12B current varies between a few µA and a few dozen mA. No way to do that with just resistors.
Thanks guys, your explanations make perfect sense. Besides, I'm really just nit-picking when you consider that a 3.3V regulator costs less than $1 anyway.
Now another thing to think about - I am planning on building my own arduino (based on breadboard designs) that will incorporate an RFM12B, a sensor that can operate between 3V and 5.5V and an LCD module that requires 5V. Yes, I'm sure I could find a 3.3V LCD module, but I already have a bunch of 5V ones and I'm cheap.
So the question is do I power the ATMEGA328P off 3.3V and pull the LCD up to 5V (using a secondary 5V line and a basic transistor amplifier), or power the ATMEGA328P off 5V and pull the RFM12B down to 3.3V (using the method described above)?
I do actually use a 5v 2x16 LCD panel on one of my JeeNodes. I power it with 4 x NiMH batteries. That gives 4.8v, which is enough to keep the power rail of the LCD happy. The JeeNode of course regulates this down to 3.3v to feed the RFM12B.
The LCD seems happy enough with the 3.3v logic levels being fed to it by the JeeNode.
So that that's probably the easiest route. Extra components required, errr, zero :-)
Give it a try, you can't break the LCD by feeding it too low a voltage!
Argh, I hadn't thought about the logic pins of the LCD requiring less than the power rail! I re-checked the datasheet and there it was, "Supply Voltage for Logic: 2.7V - 5.5V"
Thanks tankslappa, you've been a really great help in getting me started!!
Hi, I will try to make this my last question on this thread, but I was hoping someone could review my design for the power section of my custom arduino, providing 3.3V to the ATMEGA and 5V to power the LCD. Just checking I haven't missed a crucial diode/capacitor somewhere...
power source -+--[ 5V VR ]--+- GND
| | |
|--||--+--||--+
10uF | 10uF
|
+---+---[ 3.3V VR ]--- GND
| |
to LCD power to ATMEGA