Uno power issue (previously: Different Sensors/Sketches behaving differently)

[marceltrapman]

So, I have my temperature sensor included. Nice.

Now I am trying to do the same with a PIR (using a uno). I know it is ok, tested it with a 'MySensors-less' sketch.
Both Gateway and sensors are on my desk and the temperature sensor is still ok.

The PIR reports it presence (when I am lucky, most of the time it does not), asking for an id once and an answer is sent.

2014-04-29 16:28:31	MySensors Debug	Receiving command 255;255;4;5;
2014-04-29 16:28:31	MySensors	INTERNAL: Request ID (255:255) 
2014-04-29 16:28:31	MySensors Debug	Sending Command 255;255;4;5;3

Unfortunately a big silence is the result...
Nothing happens anymore (no output on the serial port either when the sensor is connected to my mac).

I did use a decoupling capacitor.
I also tested with other power sources and another radio.

Any hints what could cause this?

Oh, and I notice one other thing (non related because commenting it out does not help):
The motion sketch has this line in the setup that the temp sketch does not have: EEPROM.write(EEPROM_RELAY_ID_ADDRESS, 0);

[marceltrapman]

I did some further testing today and found that both nano's that I have work and one of my uno's as well.
My three other uno's don't do anything though
So, the question is what is going on...

I found this sketch that makes it possible to test the radio and I found that the cards that work report ok.
The cards/radio's that do not work report with many zero's (as described in the article) and give me the wrong model (it is a nRF24L01+), disabled CRC Length instead of 16bits and PA_MIN instead of PA_MAX.

††žModel		 = nRF24L01
CRC Length	 = Disabled
PA Power	 = PA_MIN

So there you have it. My conclusion is that the radio is underpowered. Or do I draw that conclusion too early?

Questions are:

1. Is this really a power thing?
2. Why can it be that these 3 board behave differently than the one that gives good results?
3. What can I do about this (without using external power which I would like to avoid)?

I have soldered a decoupling-capacitator on all of the radio's (did so after the first test and tested them again).
I also have voltage regulators so I can move the power to the 5v pin and use the regulator to bring the voltage down but doing so using a breadboard did not bring me anything (pins of the regulator are very short but they did 'click' into the board.

I hope somebody can help me with this!

Before you ask: yes, I did switch radio's, boards, power source so I am sure that this is really an issue of the boards and not the radio's and power source. The radio's all work with the nano's and a separate power source that gives enough power to allow a board to act as web server with two relays still does not work with only one radio on it.

[BulldogLowell]

@marceltrapman

You probably did check the voltage on the 3V3 pin, yeah?

Does it happen that the three Uno boards are the same maker and the working one is different? There could be a problem on the boards...

When you powered the radio on the breadboard, you grounded it back to the uno?

[BulldogLowell]

@marceltrapman

One more thing... did you try to Flash the EEPROM on any of the three boards that are giving you problems?

[marceltrapman]

I checked the voltage of the 3.3 volt pin on one of the three board and it was 3.30 or 3.31 volt all the time so that looks very good.
Then I checked how much ampere it delivered and that is .20A compared to something like 1.3 on a nano.

At first I thought that that was it.
But then I looked up the specs of a single chip and they say 1.9 to 3.6V and 13.5mA max (summary of power down, TX and RX).
So, it looks like all is ok unless a complete module uses more power but I can't imagine it to be very much.
Unfortunately I can't find any documentation on the ones I have ( (http://www.ebay.de/itm/360771486558?var=630173586638&ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649) or comparable.

Although I have 5v and 3.3v Pro Mini's coming my way there are some sensors that I would like to use with the uno if possible because I don't need the smaller footprint for them and because I have them lying around otherwise...

@BulldogLowell said:

One more thing... did you try to Flash the EEPROM on any of the three boards that are giving you problems?

This was one my first thoughts as well. So the answer is yes

[BulldogLowell]

@marceltrapman

Sorry, and you are only powering the single PIR sensor and the radio, right?

I was thinking if you were drawing too much off the whole Uno it may cause a problem with the radio. On that note, you changed the PIR too?

Side Note: I like the 3.3V ProMini a lot because it can accept a range of voltage in and you don't need to step it down for the radio. you are basically dedicating the on-board voltage regulator to your radio.

[marceltrapman]

@BulldogLowell

The last test was on a board without powering anything except the radio.

Reply to side note: I ordered both types so that I can decide what I want to use.

My line of thought was more that it would be 'better' to bring the 5v down to 3.3 because many sensors need 5v instead of 3.3...

[BulldogLowell]

@marceltrapman

Well I can certainly see why you are befuddled regarding the Uno's... strange.

I have been powering the ProMini with 5V (connect to to the RAW pin) so I can power the sensors at 5V and output 3.3V to the radio.

The ProMini... you have to respect that little guy for all it can do, including PWM on 6 pins. Great stuff, you just have to get used to felling like Gulliver in Lilliput when you are working on them.

[marceltrapman]

@BulldogLowell said:

I have been powering the ProMini with 5V (connect to to the RAW pin) so I can power the sensors at 5V and output 3.3V to the radio.

The ProMini... you have to respect that little guy for all it can do, including PWM on 6 pins. Great stuff, you just have to get used to felling like Gulliver in Lilliput when you are working on them.

Hmm, I was already looking forward to receiving them.
Now even more!