NRF24L01+ issue after connecting Logic Analyser

[ehsanito]

Hello there;

I have a weird problem. My NRF24L01+ Module work fine but as soon as I connect the logic analyzer probes to my circuit, the register values that are shown via UART change to nonsense.

any idea.

Thank you

[ogden]

NRF24L01 does not have UART serial interface but SPI. Maybe that's why? You have to connect and configure logic analyzer in SPI mode.

[ehsanito]

Maybe my poor English made you think that I think the NRF24L01+ has UART.

I'm using a MCU to read and write values to the NRF24 registers and I'm using the MCU's UART to monitor the values. When I have only the MCU and the NRF24 module connected, everything works fine but as soon as I connect the probes of my PC based logic analyzer, the values that I receive on the MCU from NRF24 module change to something that don't seem to be correct.

[ogden]

Your English is fine, no worries. It could be ground loop noise. While using PC logic analyzer - try to power MCU+NRF24 from battery (like 2xAA) or PC USB. If does not help - then it could be some circuit error or just faulty logic analyzer? It will help if you show schematics, logic analyzer used, description/photo how you connect it all together.

[MarkF]

What MCU are you using?  Specifically, what power supply voltage?

When I was testing my NRF24L01+, I used a pic16f877a at 5V.  When, I powered the NRF24L01+ with 3.3V everything worked.  However at 3V, it stopped.  I believe the problem was that at 3V the logic high level was too low to be recognized as such.

Maybe just attaching your probes presents too much of a load if you logic levels are on the edge?  If you MCU is running at 3.3V, this is probably not your problem.
 

[Doctorandus_P]

First:
I do not like the nRF24L01+ very much.
It has a very weird and illogical set of features (such as the 6 pipes) that are more of a nuisanse and hindrance to use it in a sensible way then an asset.

Then there is the "+" in the type number. Without the "+" it is a very much different chip with other (older) features.
There are also clones floating around Ali / Ebay / China, which do (often) work, but are arount 99.9% compatible.
The most used 8-pin breakout board for these chips do not have enough decoupling capacitance. Reliability and range can be improved by soldering more capacitance directly on the connector pins.

As of your Logic Analyser question:
first make sure you have proper ground connectons between all your (bread?) -boards, power supplies, and development / programming / and Logic Analyer boards.

Always connect the GND connections first between all applicable sub circuits.

15 cm long fly wires do not make a good GND connection.
15 cm long fly wires do not make a good GND connection.

If you then still experence data issues, it is time to get an oscilloscope and check the signal integrity.

Also agree with MarkF. From what I remember nRF24... works for 3V3, but not much under 3V. It does not have a very wide operating voltage range such as most uC's do.

A maybe related issue that I have encountered a few times, is that if you use a (small) single power supply for multiple boards, and you hot wire a new board in your setup, that the buffer capacitance on such a board can momentarily dip the powersupply below the operating voltage of the other boards it is connected to. Even if it's for only a ms, it is desastrous for any chip that has a flipflop in it.

[MarkF]

Doctorandus_P would you recommend a different module?

I have a bunch of nRF24L01+ modules but only tested as far as get two of them to talk to each other.
I want to have one mastar and four slaves.  The master will gather data from the slaves.
What I'm doing is to have the master connected to a PC and each slave a wireless throttle for HO Model Railroad speed control.

[TK]

I recommend using the RFM69HCW.  There is an excellent Arduino Library written by Felix at LowPowerLab, but the code can be easily migrated to other microcontroller platforms (MSP430, STM32, AVR)

EDIT: corrected name of the RF tranceiver module

[MarkF]

I recommend using the HCM69HCW.  There is an excellent Arduino Library written by Felix at LowPowerLab, but the code can be easily migrated to other microcontroller platforms (MSP430, STM32, AVR)

What is it?   Google search doesn't return any results.

Edit-  I'm using a pic18f2620

[ogden]

I do not like the nRF24L01+ very much.

Sure. It is comparably old chip, old design. People get emotional or fall in love with one chip but do not like another. That's OK. There's nothing wrong with (genuine) nRF24L01+ - if you use it for simple near range applications and use your head properly when select operating channel/frequency. I do not think OP have to change RF chip just because he can't get SPI signal integrity right. Most likely it is not chip to blame here.

The most used 8-pin breakout board for these chips do not have enough decoupling capacitance. Reliability and range can be improved by soldering more capacitance directly on the connector pins.

Good point. For testing purposes add some decoupling capacitor 0.47uF and electrolytic 47..100uF to module.

Also agree with MarkF. From what I remember nRF24... works for 3V3, but not much under 3V.

This is where I disagree. Chip (genuine one) does work at low voltages well, but it's output TX power, thus transmit range decreases dramatically. This may lead to conclusion that chip does not work.

For those who are looking for 2.4GHz chip, I would suggest Ti CC-series:
http://www.ti.com/product/CC2500

Hobbyists often use sub-1GHz transceiver:
http://www.ti.com/product/CC110L

[oPossum]

I recommend using the HCM69HCW.  There is an excellent Arduino Library written by Felix at LowPowerLab, but the code can be easily migrated to other microcontroller platforms (MSP430, STM32, AVR)

What is it?   Google search doesn't return any results.

I think the correct part number is  RFM69HCW