CC2540 WITH CC2592 ISSUES

Aleksandar Milanovic50

Other Parts Discussed in Thread: CC2540, CC2592

Hello ,

I am using BLE cc2540 with range extender cc2592 .. I have configured the cc2540 with the following code "

HCI_EXT_SetTxPowerCmd(HCI_EXT_TX_POWER_4_DBM);
// Set extended power
HCI_EXT_ExtendRfRangeCmd();

I see no effect of cc2592 .. I observed the 3 digital control lines PA_EN , LNA_EN, HGM..... I see 3 pulses over 2ms on each PA_EN , LNA_EN ( not at the same time ..interleaved ) .. HGM is zero all the time .. I am sending advertizing packets at 100ms intervals. I should be seeing about 20dbm output but I do not . The output is the same as if cc2592 not in ...

Would someone please help :

A- OFFER TIMING OF THE 3 DIGITAL CONTOLL SIGNALS

B- THE BEST WAY TO TEST THE CC2592

Many thanks

Alex

over 10 years ago

0 Fredrik K over 10 years ago

TI__Guru** 106705 points

Hi Alex,

You need to test the output on a spectrum analyzer to verify the RF performance. Have you done that? Also note that the CC2592 is highly dependent on correct layout for proper operation. Which reference design(s) did you follow.

The description of the control signals sounds correct. It will advertise on three channels and it will transmit first and receive (for ACK) on all channels.

Cheers,
Fredrik

0 Aleksandar Milanovic50 over 10 years ago in reply to Fredrik K

Intellectual 360 points

Hi Fredrik , I have observed the results with spectrum analyzer.. I have one of Ti's key fobs cc2540 based ONLY and I see no difference between it and the my unit with combination of PA ( cc2592 and cc2540 ) . We did follow reference design from Ti's .. Our layout is I thing rather good .
What else can I check . We are using IC antenna 2.4GHZ from Fractus )
Alex

0 Fredrik K over 10 years ago in reply to Aleksandar Milanovic50

TI__Guru** 106705 points

I assume you measured with a coaxial cable? Or did you do a radiated measurement?

Which reference designs did you follow? We do not have a reference for CC2540 and CC2592 so I guess you used a combination of different designs. Can you post your schematic and layout?

0 Aleksandar Milanovic50 over 10 years ago in reply to Fredrik K

Intellectual 360 points

Hi Fredrik,

I have conducted radiated check and compared the results with cc2540 ONLY and our module with PA. I am not at liberty to share the the full schematic. (sorry) . But here is the layout and main part of the sch.

Fredrik,

In addition here is the snapshot of the 3 digital control lines . Top trace is PAEN bottom is LNAEN. Can you explain the amplitude decay ? Is advertizing done on all 3 channels each time ?

One more thing ... the RF_N and RF_P output of cc2540 has identical 1V signal .. Same polarity .. Should this signal be of opposite polarity ?

Alex

0 Fredrik K over 10 years ago in reply to Aleksandar Milanovic50

TI__Guru** 106705 points

Hi Alex,

I see you have a coax connector on the board, so I recommend you connect it to the spectrum analyzer to measure the exact output power.

The scope plot of the control signals indicate that the supply voltage falls when you run the PA. What is your power source? Remember that the CC2592 will consume almost 200 mA at full power.

The signals you see on RFx is bias voltage for RX. This will come right after TX as you can see from your scope traces.

The layout is not strictly following the reference design, but looks ok. I cannot see your ground vias, but I assume you have a lot of them and that both the CC2540 and the CC2592 is properly connected to a solid ground plane.

When you have fixed the power supply and connected the board to the spectrum analyzer, you can play a bit around with L2 to increase the output power.

Cheers,
Fredrik

0 Aleksandar Milanovic50 over 10 years ago in reply to Fredrik K

Intellectual 360 points

Thanks for that Fredrik ... You are correct the control lines were loaded because I was powering up the module via Programmer ONLY . I since added 3.3V additional supply and got the control lines to swing correctly . However still no anticipated gain from PA .. Any suggested values for L2
Alex

0 Fredrik K over 10 years ago in reply to Aleksandar Milanovic50

TI__Guru** 106705 points

What is the conducted output power level?

0 Aleksandar Milanovic50 over 10 years ago in reply to Fredrik K

Intellectual 360 points

We are looking to get +22dbm

0 Aleksandar Milanovic50 over 10 years ago in reply to Aleksandar Milanovic50

Intellectual 360 points

Hi Fredrik,

Can you comment on the Bias resistor setting on cc2592 . Is there a note somewhere how to set it ? I am still very puzzled that I do not get ANY gain from this device . I could understand reduced gain due to many things but gain of 1 ? Any other suggestions ?

Alex

0 CHS over 10 years ago in reply to Aleksandar Milanovic50

TI__Mastermind 18780 points

The bias resistor must be kept as in the reference design and shall not be altered.

0 Aleksandar Milanovic50 over 10 years ago in reply to CHS

Intellectual 360 points

Thanks I have that same value ... Any other suggestion why I do not see ANY gain ?
Alex

0 Fredrik K over 10 years ago in reply to Aleksandar Milanovic50

TI__Guru** 106705 points

What output power are you measuring when doing a conducted measurement on your spectrum analyzer? And what is the current consumption?

Bluetooth®︎

Bluetooth forum

CC2540 WITH CC2592 ISSUES