CC1120: Settings for 50Kbps, 100BW, -109 dBm

Hi Ilya,

It is not straight forward to measure sensitivity, especially at such low levels. Can you share details about your test setup, preferably with a photo of the setup?

Regards,
Fredrik

Hello Fredrik,

The details about their test setup(Foto attached):

1. Configuration in SmartRF - 433,42Mhz, 50 kbps, dev 25 khz, bandwidth 100 khz
2. 1 gpio set to mode 9 (see photo), then they hook the oscilloscope probe to this output. They call this mode - a carrier detector.
3. At the antenna input they feed the carrier from the generator.
4. In the presence of a powerful carrier from the generator (for example, -90 dBm) at the antenna input, they observe a 25 kHz meander on the oscilloscope.
5. Then they begin reducing the power on the generator and control the breakdown of the signal on the oscilloscope in frequency.

As result they see the breakdown for this settings at -99 dBm instead of -109 dBm as described in datasheet for high performance mode.

How to check is it really perfomance mode or they still work in low energy mode? 
https://e2e.ti.com/cfs-file/__key/communityserver-discussions-components-files/156/05.7z


BR,
Ilya

Fredrik,

Also i should add settigns for registers in SmartRF(Screenshots in doc file attached).
Register settings CC1120.docx

Good day Fredrik, Colleagues,

Could you help us to understand a correct way to measure sensitivity of the CC1120?

Customer implemented it this way:

He connects R&H SML03 signal generator to antenna input and use CC1120 pin 9 as an output to detect a carrier frequency of 25kHz.

CC1120 is configured thru RF Studio (RFStudio -  433,42MHz, 50 kbps, dev 25 kHz, bandwidth 100 kHz) . To pin 9 he connects oscilloscope.

Customer decreases an output power of the generator down to -109dBm.

If the CC1120 can't detect carrier frequency anymore there is no 25kHz at the output, so the customer consider it as the sensitivity level.

I understand this method can be incorrect. Could you explain a correct method to test the sensitivity for the CC1120 for

1.2kbps and 50kbps to reach -109dBm as it described in the datasheet.

If there is a sensitive data, please, reach me thru E-mail.

Thank you, Mikhail

Hi Ilya and Mikhail,

It's not fully clear to me how you generate the Tx data on the R&S SML03. From the picture (IMG_20190917_163521_2.jpg) it looks like you use FM modulation but I'm not able to see what deviation that is used?

Do you have a spectrum analyzer available so that you can verify that the transmitted data from SML03 is correct? You would like to see two peaks with 50KHz separation (+/- 25KHz from 433.42MHz)

One thing you could try is to use Smart RF Studio to receive the data directly. Configure as image below and see if you can receive the data:

Since there is no CRC in the transmitted data it will mark the packages as Not Ok but you should see 0xAA data being received anyway.

Best Regards,

R.M

Good day Colleagues,

Could you help to setup an external signal generator for 433MHz to test the CC1120 for a sensitivity level?

What else but frequency has to be setup correctly?

CC1120 is in mode of: 433,42Mhz, 50 kbps, dev 25 kHz, bandwidth 100 khz

Could you also help to setup registers of the CC1120 to reach maximum sensitivity for the: 433,42Mhz, 50 kbps, dev 25 kHz, bandwidth 100 khz?

Maybe to send a table with the resisters settings?

Customer would like to check if the CC1120 on their own board is setup correctly.

Thank you, Mikhail

A side note: I took a brief glance at the R&S SML datasheet and as far as I can see this generator is not capable of sending packets. Note that the sensitivity numbers in the CC1120 datasheet are measured sending 3 byte packets, fixed packet length and with a 20 % PER limit. If the customer want to get the same numbers as in the datasheet, a more advanced generator is needed. Note that they need to know the attenuation between the generator and DUT for accurate measurements. Also see http://www.ti.com/lit/an/swra603/swra603.pdf for different possible methods to measure sensitivity. 

Thank you very much!

Can we provide the registers settings for this mode to the customer?

Thank you, Mike

Hi Mike,

It's not clear to me what mode you are requesting register settings for, can you please elaborate? Register setting for 434MHz, 50Kbps, 25kHz dev and 100RxBW can be found in Smart RF Studio. 

Any feedback on the questions in my first post? 

Best Regards,

R.M

Good day,

Thank you for reply. Sorry what question did you mean? I didn't get it.

By registers settings, I meant if we can provide correct registers settings to the customer to test the sensitivity at 433,42Mhz, 50 kbps, dev 25 kHz, bandwidth 100 khz

To be sure the customer made settings a correct way and didn't miss anything to reach maximum sensitivity. So if we can provide a table or a file for RF studio will be fine.

Thank you, Mikhail

Hi Mikhail,

As I said in my previous post, you can find register settings for this Smart RF Studio.

From there you can use the Register Export to get all the required settings.

I was asking some questions on the test setup and the SML03 generator. Have you verified that it transmits correctly? 

Best Regards,

R.M

Hello Team,

We got feedback from the customer:
They got near-datasheet sensitivity of -106-109 dBm on thier end device with the settings from RF studio.
They used their end device as generator(transmitter) sending packets and PER was 20%.
Both sides was shielded into boxes, connected by an RF cable. The power level was calibrated by the generator and attenuated by attenuators.

Their custom settings give worse  results. So they trying to figure out how to compensate.
They cannot take and replace all the settings from RFStudio - They need compatibility with the previously released end devices (They cannot change the bit rate and other key parameters).

The questions:
1. By default, Rf Studio provides a sync word of 4 bytes with specific content.
The coustomer have 2 bytes of different content. How does the length and content of the sync word affect sensitivity?

2. How does the lack of a start and stop bit (as in uart) for bytes in a packet affect the sensitivity?
Rf Studio default settings turn on it. The customer have it turned off to reduce the transmission time of the packet.

3. What other methods can increase the sensitivity of the receiver?
Are there any other tricks, for example, a more competent use of AGC or ext. error correction mechanisms?

BR,
Ilya

Side question:
Could you please recommend good but not very expensive model of a generator that can work in packet mode?

BR,
Ilya