CC2538DK: CC2538DK

May try to add external attenuator on your TX.

As YK suggest, you can add attenuation. You can also do some of the same by using an antenna with poor efficiency. 

Why do you want a range less than 1 m? 

Thanks to both of you for quick reply.

I have 8 cc2538 and 8 smart rf board. Doing some research, I need to test my protocol by different topologies.

Let's say for line topology I have 7 hops and need 140 meters to run it which I do not have that space.

Another thing is that I need to connect all them to one PC to read/save messages (printfs) and do time stamping on terminal.

By saying attenuation, do you mean only adding antenna or you have other suggestions?

could you please suggest other ways of attenuation if any?

Thanks, 

If by saying attenuation you mean the AGC, kindly let me know how to find the values to set it properly and how to get it to work.

Thanks,

Does that mean you have CC2538 EMs? In this case you can place a fixed attenuation on the SMA before the antenna.(https://www.mouser.com/Connectors/RF-Interconnects/Attenuators-Interconnects/_/N-axjra?P=1yzxlfi)

Yes, it is.So there is no other ways of setting to reduce the tx power or sensitivity of rx and I have to by attenuation or antenna.

Thanks,

Have you tried to not use an antenna? 

From a setting side you are not able to set lower power than the lowest setting found in SmartRF Studio. 

Not actually.

What I exactly have is "CC2538EM (DK) Rev: 1.2.0.1".

I do not think I can simply remove the antenna or kindly guide me how to do it.

Thanks,

You are using these I guess: https://www.digikey.de/product-detail/de/texas-instruments/CC2538EMK/296-35787-ND/4027396?utm_adgroup=RF%20Evaluation%20and%20Development%20Kits&utm_source=google&utm_medium=cpc&utm_campaign=Google%20Shopping_RF%2FIF%20and%20RFID&utm_term=&productid=4027396&gclid=EAIaIQobChMInKO-_7uS5wIVCOh3Ch0OzQrZEAQYASABEgKvsPD_BwE

They you can select if you want to use an SMA (which you can solder on or the PCB antenna). With the latter it's difficult to add extra attenuation so using an SMA and external antenna would make it easier to control. 

Yes I am using that now. I have 10 of "CC2538-CC2592EM (DK) Rev: 2.0.1" too, but I think for both case the solution is attenuation as I am checking CC2538-CC2592EM, which has long range of transmission (max 250 meters) now, but cannot remove the antenna simply too. However attenuation can be placed on them simply.

Thanks,

If you only want 1 meter distance, why do you use CC2538-CC2592EM which contains CC2592 as PA/LNA to increase radio coverage?

These are the only options I have right now and trying to get them to work with minimum expenses.

Thanks,

If you use the CC2592 then you can do some changes to the code ensuring that the CC2592 rmains off (alter the setting of PA and LNA enable signals). This will attenuate the signal with ~30 dB instead of amplifying it. 

If I keep CC2592 off, won't it become like the normal CC2538?

Btw, I got these steps below from datasheet. Is that safe to do it and not to affect the circuity of the board?

The control registers CCTEST_OBSSEL0–CCTEST_OBSSEL7 can be used to override the standard GPIO behavior and
output RF Core signals (rfc_obs_sig0, rfc_obs_sig1, and rfc_obs_sig2) on the pins PC[0:7].

• Configure PC0 as output without pull-up.
• Set RFCORE_XREG_RFC_OBS_CTRL0 to 0x28 (Power Amplifier power-down signal)
• Set CCTEST_OBSSEL0 (controlling PC0) to 0x80

Thanks,

No, since you will have an attenuation through CC2592 since it's in power down. 

I assume that the register settings you refer to is the normal way to control the CC2592 but in this case you don't want to use this since you want to ensure that CC2592 is off to ensure poor range. This should be safe. 

Thank you so much TER.

I will test it and let you know the result in few days.

Dear TER,

I did the following to keep the CC2592 off (sending power down signal by LNA an PA pins). setting the TX power to 7 db too, but it does not send any signal out. I menat nothing is sent and received even when the two motes are exactly next to each other.

set_tx_power(-7);

#define RFC_OBS_CTRL0 REG(RFCORE_XREG_RFC_OBS_CTRL0)
#define RFC_OBS_CTRL1 REG(RFCORE_XREG_RFC_OBS_CTRL1)
#define OBSSEL3 REG(CCTEST_OBSSEL3)
#define OBSSEL2 REG(CCTEST_OBSSEL2)

#define EN_LNA_PIN (1 << 2)
#define PAEN_PIN (1 << 3)
#define HGM_PIN (1 << 0)

GPIO_SET_OUTPUT(GPIO_C_BASE, PAEN_PIN);
GPIO_WRITE_PIN(GPIO_C_BASE, PAEN_PIN, 1);
GPIO_SET_OUTPUT(GPIO_C_BASE, EN_LNA_PIN);
GPIO_WRITE_PIN(GPIO_C_BASE, EN_LNA_PIN, 1);

GPIO_SET_OUTPUT(GPIO_D_BASE, HGM_PIN);
GPIO_WRITE_PIN(GPIO_D_BASE, HGM_PIN, 0);

// PC3 -> PAEN
RFC_OBS_CTRL0 = 0x68;     //power down signal for PA
OBSSEL3 = 0x81;

// PC2 -> EN (LNA control)
RFC_OBS_CTRL1 = 0x6A;    //power down signal for LNA
OBSSEL2 = 0x80;

Could you please suggest?

Thanks,

Which value does PA_EN and LNA_EN have in this case?

Did you do this on both boards? If so, try with only one.

Have you tried to measure (with a spectrum) what you are actually sending? 

PA_EN and LNA_EN are (1 << 3) and (1 << 2) respectively.

I tried with one board and the result is still the same.

I do not have the spectrum to test the transceivers.

Is it possible to let me know the step by step process of doing it and use the CC2592 as attenuator.

As mentioned in the previous message value of LNA_EN and PA_EN are set as

#define EN_LNA_PIN (1 << 2)
#define PAEN_PIN (1 << 3)

then they are set as output and 1 is written in them

GPIO_SET_OUTPUT(GPIO_C_BASE, PAEN_PIN);
GPIO_WRITE_PIN(GPIO_C_BASE, PAEN_PIN, 1);
GPIO_SET_OUTPUT(GPIO_C_BASE, EN_LNA_PIN);
GPIO_WRITE_PIN(GPIO_C_BASE, EN_LNA_PIN, 1);

then the desired values to power down the CC2592 is written in RFC_OBS_CTRL0 and RFC_OBS_CTRL1 registers

RFC_OBS_CTRL0 = 0x68;

RFC_OBS_CTRL1 = 0x6A; 

if you see any mistakes kindly let me know.

What I meant by "Which value does PA_EN and LNA_EN have in this case?": Have you measured that you have the expected voltage level on these pins? 

Have you tested doing this only on the TX or RX side? If you do it on both sides it could be that the attenuation gets too large.

I would highly recommend you to get a spectrum of sorts to be able to see if you have something on the air or not.