• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 54 subscribers
  • Views 717 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Original question:

CC1110-CC1111: CC1110 output amplitude settings

CC1110-CC1111: CC1110 output amplitude settings

Niels Harthoj
Intellectual 310 points
Part Number: CC1110-CC1111
Other Parts Discussed in Thread: CC1101, CC1190

In attached document I have collected some questions regarding sensitivity vs. FCC-compliant settings for bandwidth and modulation.

Devices is CC1110-CC1190 combination compared to CC1101.

Questions to CC1101, CC1110+CC1190 and DN006
Sensitivity for 38,4kHz baud and 1% package loss is considered for following numbers:
For CC1101 following sensitivity is listed in the data sheet:
Sensitivity:
915MHz, GFSK, 20 bytes packages, 20kHz deviation, 100kHz Rx bandwidth: -104dBm
In Design Note 006 following is listed:
Preferred settings, Table 4:
915MHz, 2-FSK, 20kHz deviation,100kHz Rx bandwidth: -104dBm
For FCC compliance (Table 3) following is recommended:
915MHz, 2-FSK, 177kHz deviation, 540kHz Rx bandwidth: -100dBm
In CC1110 data sheet (page 15) following is listed for sensitivity:
915MHz, 2-FSK, 20 bytes packages, (20kHz deviation?), 100kHz Rx bandwidth: -100dBm
(deviation is estimated from SmartRF Studio preferred settings).
Questions are now:
1) For CC1101: GFSK and 2-FSK modulation seems to have same sensitivity (data sheet and
DN006) - is this true?
2) Comparing DN006 (CC1101) and CC1110 it looks like CC1101 is 4dB more sensitive than
CC110 for same settings. Is this expected?
3) Which sensitivity should we expect for CC1110-CC1190 with the DN006-settings for FCCcompliance?
4) In my current setup I see an improvement of 9dB in sensitivity when CC1190 HGM is
changed from 0 to 1 and other settings (from DN006 FCC compliance) are unchanged. Is this
to be expected - noise figure etc. considered?
5) When using 2-FSK instead of GFSK, deviation and Rx-filter unchanged, the CC1110 device is
approx. 2 dB less sensitive. Is this reasonable from expectations, see 1)?

Regard

Niels GFSK_2FSK.pdf

  • 0
    TI__Guru**** 317180 points
    1) As I read the app note it's a typo in table 3, 4, and 5. The app note state that the preferred settings were used as a starting point and GFSK is used in these.
    2) As far as I can see the delta is 2 dB for 38.4 kbps. Some difference is expected due to transceiver vs SOC.
    3) ~6 dB improvement with CC1190 compared to without.
    4) The expected delta could be found from www.ti.com/.../swra356.pdf
    5) Yes
  • 0 in reply to TER
    Intellectual 310 points

    Hi

    Thank for the reply. Some further comments to the answers:

    1) OK, but table 3, Modulation format for FCC compliance is still 2-FSK? Will an updated DN006 be available?

    2) ?, but that is also what I see during test

    3) OK, we see this number in other documents, but not really see where it comes from. LNA-gain, CC1190, is specified to 11,6dB dependent on input level. Some loss during antenna filter, SAW and other circuits should be expected and is implementation dependent?

    4) swra356 (AN094) deals with narrowband datarate/modulation/Rx bandwidth. We must use 177kHz deviation, 541kHz Rx-bandwidth and maybe some other filter settings (settings from 250kHz preferred..)

    5) Ok, that's what we see

    Regards

    Niels

  • 0 in reply to Niels Harthoj
    TI__Guru**** 317180 points
    1) I suspect that table 3,4,5 should have GFSK and not FSK. But the engineer that wrote the app note is on vacation and will be back in 3 weeks so I'm not able to double check. But all settings for CC1101 is GFSK based so I'm not seeing the reason why FSK should have been used here.
    2) I was comparing the CC1101 and CC1110 datasheets for 38.4 kbps.
    3) Gain alone does not translate to sensitivity. It's mainly the the Noise Factor and the noise factor that sets the sensitivity improvement. (www.ti.com/.../slaa652.pdf and others)
  • 0 in reply to TER
    Intellectual 310 points
    As per answer, see 1) above, I am waiting for an engineer to return from vacation and then answer the question.
  • 0 in reply to Niels Harthoj
    TI__Mastermind 39860 points

    For FCC 15.247 measurements 2-FSK was used in order to get the 6 dB BW above 500 kHz. DN006 uses 2-FSK; data sheet uses GFSK. 

    With reference to http://www.ti.com/lit/an/swra361a/swra361a.pdf  (Using the CC1190 Front End with CC1101 under FCC 15.247) there is a 5 dB improvement in sensitivity when using CC1190 with CC1101 @38.4 kbps. 

    Assuming CC1101 has a noise figure of approx. 10 dB. CC1190 has a noise figure of 2.9 dB. If CC1190 had infinite gain (which it does not) the improvement in sensitivity would be 10 – 2.9 = 7.1 dB. Assuming CC1190 gain of 11.6 dB the improvement in sensitivity would be approx. 6 dB. Accounting for the SAW filter the improvement in sensitivity will be closer to 5 dB.

  • 0 in reply to Sverre
    Intellectual 310 points

    Some few questions: where does noise figure of 10 for CC1101 come from, and which gain for CC1101 do you use to reduce gain (datasheet?) sensitivity improvement from 7,1dB to approx. 6dB?