Where to get legit nRF24L01+ modules?

[NeverDie]

@Nca78
In that case you could get the job done using one of the ARM chips that comes with a fast sampling AD converter and, preferably, lots of RAM. You could get it on one of the cheap demo boards to make the project easier. Fortunately, the fake nrf24's are different enough from the genuine in their Tx and Rx current draws that you won't need to split hairs on your measurements to tell them apart.

I did something similar once for capturing and decoding RF signals using an Arduino Due, so I'm sure it can be done. In fact, you could probably use a Due for this if you wanted to, but no doubt there are better options available now.

[Yveaux]

@NeverDie said in Where to get legit nRF24L01+ modules?:

Fortunately, the fake nrf24's are different enough from the genuine in their Tx and Rx current draws that you won't need to split hairs on your measurements to tell them apart.

Nope. Some of them might be, but certainly not all of them...

[NeverDie]

@Yveaux said in Where to get legit nRF24L01+ modules?:

@NeverDie said in Where to get legit nRF24L01+ modules?:

Fortunately, the fake nrf24's are different enough from the genuine in their Tx and Rx current draws that you won't need to split hairs on your measurements to tell them apart.

Nope. Some of them might be, but certainly not all of them...

Which ones? Even the fakes have names and datasheets. The datasheet tells the tale. The ones I tested (see other thread for the details) were pretty noticeably different.

[Yveaux]

@NeverDie said in Where to get legit nRF24L01+ modules?:

Which ones? Even the fakes have names and datasheets. The datasheet tells the tale.

I've ordered many nameless modules (regular black ones) through Ali express the past years, which didn't come with a datasheet or any specs. A number of them are proven fakes (some even x-rayed by Nordic), but I've not seen any of them having a current consumption significantly different from original Nordic modules.
I just want to stress again that although some fake modules have a higher current consumption, this certainly isn't true for all fake modules.
A module with regular consumption still could be fake. X-ray is the only way to be 100% certain.

[NeverDie]

@Yveaux

What was it about the chips (the ones that had identical current draws to the real thing) that you sent to Nordic for x-ray that motivated you to send them?

If your requirement is 100% certainty, then you may just have to make your own modules with chips procured either direct from Nordic or from a trusted supplier, like Digikey.

[Yveaux]

@NeverDie said in Where to get legit nRF24L01+ modules?:

If your requirement is 100% certainty

That's not my requirement; I didn't start this thread :laughing:
I only have (presumably) fakes deployed around the house and they've never failed me. I won't bother getting 100% genuine ones.

[d00616]

I have received two of the red modules. This is an X-Ray image of the chip (montage)

[NeverDie]

@d00616 said in Where to get legit nRF24L01+ modules?:

I have received two of the red modules. This is an X-Ray image of the chip (montage)

Meaning?

[d00616]

@NeverDie @Yveaux said in Where to get legit nRF24L01+ modules?:

Could be that different clones of the nRF exist of which some have an increased power consumption, but currently the only way to know for sure (also for Nordic ) is to x-ray them.

I hope somebody can say if this looks like an original.

[Yveaux]

@d00616 the xray is rather vague.
Here are die shots of the original and a fake: https://zeptobars.com/en/read/Nordic-NRF24L01P-SI24R1-real-fake-copy

Maybe you can match them.

[d00616]

@Yveaux Thank you for the X-Ray images. I can't get images in that resolution.

[Yveaux]

@d00616 they aren't xray images. The die shots were obtained by eating away the chip package with acid.

[Bogus Exception]

@hek Thanks for writing. I got 4 from itead the other day (2 each of 2 types), and they are in the boxes in front of me. FOr one thing-they have a box! I am making a video for my blog comparing them to the clones (as best I can), and I'm going to test out that current statement I found online...

Thanks for the tip!

[Bogus Exception]

@tbowmo Thanks for the tip. I have some in front of me, still in box. Will post results.

[Nca78]

I received my Ebyte modules, not a warranty of authenticity but the nrf chip has a square instead of a dot like all the fakes I have been using until now. I have a few standard SMD modules and a PA LNA version, I will try to make some basic range tests soon compared to the SMD clones.

[Bogus Exception]

OK! I followed your suggestions and tried itead. These are the pics of what I got 4 of (sorry no pro rig-just a loupe & module in one hand & mouse in the other!

(all are hosted on one of my servers, not here locally)

[gohan]

@Bogus-Exception let us know how they perform

[Bogus Exception]

@gohan I'll do my best to measure meaningful metrics between these & clones. Also, I have 4 eBytes coming (2x blue, 2x red)...

[Bogus Exception]

So far I've noticed on all versions of the legit nRF24 chips that the 'square' is actually 4 vertical lines, covering a square area, at the 'top left' of the chip.

Just tried over the past 2 hours to replace the clone in my MyController gateway with either the itead or ebyte legit modules, and neither one of them would communicate with the other nodes. I put the clone back in, and it worked. Weird.

Not like it matters much, but here is the sketch (all comments removed) using a BME280 and a light sensor module, on a Nano, USB powered:

#define MY_DEBUG
#define MY_RADIO_NRF24
#define MY_RF24_PA_LEVEL RF24_PA_LOW
#define MY_GATEWAY_SERIAL
#if F_CPU == 8000000L
#define MY_BAUD_RATE 38400
#endif
#include <Adafruit_Sensor.h>
#include <Adafruit_BME280.h>
#define BME_SCK 13
#define BME_MISO 12
#define BME_MOSI 11
#define BME_CS 6
#define SEALEVELPRESSURE_HPA (1013.25)
const float ALTITUDE = 18;
Adafruit_BME280 bme(BME_CS); // hardware SPI
#define LIGHT_CHILD 0
#define BARO_CHILD 1
#define TEMP_CHILD 2
#define HUM_CHILD 3
boolean metric;
uint8_t decimals;
#include <MySensors.h>
MyMessage lightMsg(LIGHT_CHILD, V_LIGHT_LEVEL); // 0
MyMessage pressureMsg(BARO_CHILD, V_PRESSURE); // 1
MyMessage tempMsg(TEMP_CHILD, V_TEMP); // 2
MyMessage humMsg(HUM_CHILD, V_HUM); // 3
#define LIGHT_SENSOR_ANALOG_PIN 0
int lastLightLevel;
unsigned long SLEEP_TIME = 60000; // Sleep time between reads (in milliseconds)
void setup()
{
  #ifdef MY_DEBUG
    Serial.begin(115200);
  #endif
  decimals = 2;
  metric = getControllerConfig().isMetric; 
  bool status;
  status = bme.begin();
  if (!status) {
    #ifdef MY_DEBUG
      Serial.println("Could not find a valid BME280 sensor, check wiring!");
    #endif
    while (1);
  }
}
void presentation()
{
  sendSketchInfo("Master Nano", "1.0");
  present(LIGHT_CHILD, S_LIGHT_LEVEL);  // 0
  present(BARO_CHILD, S_BARO);          // 1
  present(TEMP_CHILD, S_TEMP);          // 2
  present(HUM_CHILD, S_HUM);            // 3
}
void loop()
{
  int16_t lightLevel = (1023-analogRead(LIGHT_SENSOR_ANALOG_PIN))/10.23;
  if (lightLevel != lastLightLevel) {
    #ifdef MY_DEBUG
      Serial.print(F("Sending lightMsg: "));
      Serial.println(String(lightLevel));
    #endif
    send(lightMsg.set(lightLevel));
    lastLightLevel = lightLevel;
  }
  float p_Mb = bme.readPressure() / 100.0F;
  float p_Hg = p_Mb / 33.86;
  #ifdef MY_DEBUG
    Serial.print(F("Sending pressureMsg: "));
    Serial.println(String(p_Hg));
  #endif
  send(pressureMsg.set(p_Hg, decimals));
  float t_C = bme.readTemperature();
  float t_F = (t_C * (9/5)) + 32;
  #ifdef MY_DEBUG
    Serial.print(F("Sending tempMsg: "));
    Serial.println(String(t_F));
  #endif
  send(tempMsg.set(t_F, decimals));
  float h_Pct = bme.readHumidity();
  #ifdef MY_DEBUG
    Serial.print(F("Sending humMsg: "));
    Serial.println(String(h_Pct));
  #endif
  send(humMsg.set(h_Pct, decimals));  
  #ifdef MY_DEBUG
    Serial.print(F("Sleeping for: "));
    Serial.println(String(SLEEP_TIME));
  #endif
  sleep(SLEEP_TIME);
}

It works with the clone... I am using a regulator board for the nRF24-the one it plugs into & breaks out the 8 pins...

[Yveaux]

@Bogus-Exception said in Where to get legit nRF24L01+ modules?:

So far I've noticed on all versions of the legit nRF24 chips that the 'square' is actually 4 vertical lines, covering a square area, at the 'top left' of the chip.

You cannot rely on the print of the nRF to distinguish fake from genuine. Nordic is fabless, so they use different factories to produce their chips, each with its own specific marking.