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.

CC1120-CC1190EM915RD: Where to buy???

Part Number: CC1120-CC1190EM915RD

I have the CC1120DK Dev kit from digikey. Where do I get the CC1120+CC1190 combo?

  • For CC1120 + CC1190 combo the only kit we have for sale is www.ti.com/.../boostxl-cc1120-90
  • Is that the board used in the more than 100km range youtube video?

  • No, the board used in the video is not for sale but the underlying schematic is the same for the boosterpack I linked to.
  • Ok. I want to evaluate the CC1120+CC1190 as a standalone radio, not with Sigfox, as simply as possible for range vs power, error rate, etc. using the setup in the video. What is the simplest way to do this?
  • This kit has Sigfox in the name since it can be used for Sigfox. But you are free to use it without any specific software since the boosterpack doesn't have a MCU and can be used together with SmartRF Studio for easy testing.
  • Are there component values available to change the matching on the SigFox boosterpack down to 400MHz?

  • CC1190 doesn't support the 400 MHz band.
  • Sorry for the flawed question. Its irritating to make a mistake like that! My focus is on the CC1120/1125/1310/etc. and I forgot the CC1190 doesn't work at 400MHz.

    For indoor radio links with the maximum range, my research is pointing me toward a 400MHz solution that can operate under FCC part 90 and part 15 (with the power reduced). I'm thinking 400MHz because I've read through some very good reports that confirm the longer wavelengths perform better through construction materials, and because an efficient antenna isn't hard to make at that frequency.

    I'm looking for a good demo board for a 400MHz 27dBm+ solution.
  • Is an evaluation board available using a CC1120/1125/1310/etc with a front-end LNA/PA at 400MHz  and 27dBm+?

    Thanks

  • We only provide demo boards for the 868/ 915 MHz band since we want to promote CC1190. For 400 MHz you have to see if you find a evaluation board from Skywork etc and connect this together with a CC1120/ C1310 board.
  • Am I right in thinking a 400MHz radio link at 20kbps provides more range inside a building than 900MHz?

  • In theory, yes, that should be the case. But due to the physics the antennas for 433 MHz has to be larger and require larger ground plane to be as efficient as on 868 MHz.

    => With the same space/ volume available for an antenna the answer is not given since what you gain going from 868 MHz to 433 MHz in the ability of going through walls etc is at least partially lost in poorer antenna.
  • An initial test of the CC1120 420MHz to 470MHz with:

    • the supplied antennas
    • PER test
    • CC1120 stand alone
    • Easy mode
    • 434MHz

    showed a signal strength of -11dBm with the transceivers next to each other on the bench. 

    2 questions.

    1. Is that signal strength about right for the demo?
    How much improvement should I expect to see in a working system with a better ground plane, better antenna, etc? I know this isn't a fair question, but I need some idea of what is do-able with this chip at +14dBm.
    1. How much can the signal strength be increased before over-stepping the FCC limitations under part 15?
    2. Part 90?

    I am working toward a head-to-head comparison of a TI narrow band solution and LoRa and want the test to be as fair as possible.

    Thanks

  • Measuring radiated with the units close to each other doesn't give much information since the RSSI saturate.

    It depends if you plan to follow 15.247 or 15.249. If you use wide band modulation or frequency hopping you can send more than 14 dBm.

    Part 90 require fairly low symbol rate and deviation to fit into the mask. What are you planning to use?
  • I'm still in the early planning stages, but I for FSK I'm looking at 20kbps and a link budget of 130dBm+, in the 400-470MHz range. This system will be working inside multi-floor buildings (such as an apartment building) with concrete and sheetrock/metal stud construction. I need a range of a 100+ feet from one side of the building to the other, through at least 1 floor.
  • 20 kbps is high for part 90. Take a look at how the mask look like and the signal bandwidth.
  • That's what I'm thinking. At a bandwidth of 12.5kHz, I could get 25kbps with 4-FSK, but that bandwidth is hardly ultra narrowband. Such a wide bandwith with 4-FSK, even with a transmit power of 27dBm out of the antenna (under part 90), an FSK system just doesn't appear to have the link budget compared to LoRa.

    Which puts LoRa in the better position for a building-wide radio network that can provide an average of, say, 250bps to each of 100 radios, with a peak of 1kbps, and a link budget in excess of 145dB.

    Its looking like that isn't possible with FSK.

  • Not sure how you are comparing.
    I assume that you need the same throughput regardless of the radio you want to use. LoRa has typically fairly low throughput according to my understanding so how are you comparing?
  • I'm still learning, thus my questions here.

    My understanding is that a LoRa gateway can establish multiple simultaneous links to multiple radios, even on the same frequency, by changing the spreading factor. My understanding of chirp spread spectrum modulation is that two radios can share the same frequency, but with different spreading factors, and both simultaneously communicate.

    The "chirp" is actually very clever implementation of a mathematical transform of a pulse. Because the Fourier transform of a pulse (as in Dirac delta) has an infinite spectrum, a chirp similarly is a sweep of frequencies. The receiver essentially scoops up all the noise, runs it through a reverse transform, and the pulse is recovered out of the noise. Spreading the energy of a pulse out over a large band keeps everything beneath the FCC limits.

    So, in an FSK system each radio has to share the same channel using a multiple access scheme.

    LoRa WAN also provides a multiple access scheme. And for those radios close enough to the gateway, Lora WAN works out the chirp spread-spectrum settings (spreading factor, channel bandwidth and frequency) to get as much throughput as possible. The total bandwidth of course has to be shared among those radios on the same LoRa settings.

    But, for those radios further away where the link budget is lower, LoRa WAN works out the settings for those radios at a different spreading factor, data rate, etc. and thus of course a lower effective bandwidth. And for the radios furthest away, still more aggressive settings with a still lower bandwidth- as low as 1kbps. But for those radios furthest away, even if 2 have to be put into a group with just 2kbps because of the LoRa settings, its OK.

    So, the radios are effectively put into different groups that can communicate at the same time, and it works out that the larger groups also have access to a larger shared bandwidth, while the smaller groups share less.

    The kicker is that a LoRa gateway can listen on multiple channels and spreading factors simultaneously, communicating with all the radios, each with access to a larger bandwidth.

    Whereas a radio in an FSK network has to share one channel with all the other radios. An advantage an FSK system would have over LoRa is that each radio is still able to communicate with the other radios in range, where in a LoRa network the nodes can only communicate with the gateway.

    In an FSK system, to increase the link budget you can increase transmit power and decrease bandwidth at the cost of effective data rate. In an LoRa system, to increase the link budget you can also increase transmit power and decrease the bandwidth, but also increase the spreading factor and even change channels, effectively *increasing* the effective system datarate.

    What am I missing?
  • Lance,

    Chirp has been used in radar systems for very long time, but it is in general never been adopted for communication systems. If it had been such a great idea for communication, then it would have adopted long time ago for communication too (check with NASA). The main problem is the inherent spectrum efficiency, this is particularly bad for ISM band where the concept is that everybody shares a bit of a "free to use "frequency band. FCC has practically banned LoAR in part 90 by requiring a certain spectral efficiency. In Europe (ETSI En 300 220) both bandwidth and also duty cycle restrictions will to some extent also protect agains LoRA "pollution" - or at least the worst version of it (500kHz bandwidth and very low data rates). Transmitting several carriers on the same with different coding sounds like a good idea, but the truth is that you hardly get and "protection" such that the two carriers end up disturbing each other. In a narrow-band solution, you can divide the solution into different frequency channels and you will be protected by the selectivity of the receiver (50-60 dB protection). By the way, i am not sure how you got the information that LoRA is using chirp, in their public information, they do not mention this as far as i know. This is a classical "lock in vendor" tactic - they want to hide the fact that there is only one provider of RF silicon (but several MCU vendors that product SIP solutions with the same RF transceiver).
  • [Again, bear in mind I am still in research mode here...]

    I get that, to a certain extent, there is a diminishing return to using any spread-spectrum technology, especially one like LoRa that uses a continuous swath of bandwidth at a low data rate. If the technology is adopted widely it will begin to be less effective.

    My problem is that right now it is a great solution to a difficult problem- I need a long range (as in multi-floor, building-wide) intra-building communication solution with maybe 100 nodes in a star topology at a data rate on the order of 1000bps for each node. LoRa appears to be the only technology with the numbers.

    Ultra narrow band *could* work but it isn't there yet, that I can see, for one main reason- It's long range capabilities are hobbled by its attendant low data rate. (Even at that, does ultra narrow band compete with LoRa at similar low data rates?)

    In a star configuration, using ultra narrow band means by definition it is a multi-access technology, and therefor the available bandwidth is divided by the number of nodes. Worse, in any busy multi-access network collisions will also take a further toll on throughput. A general rule of thumb is a 50% toll.

    If ultra narrow band can compete with LoRa on range, then what would help is a TI solution where a gateway chip could scan a number of narrow band channels for activity, then either stop on that frequency to receive, or task another receiver to handle receiving the data. This would require a longer preamble for the gateway receiver to prepare, but a small price to pay for much longer range.

    In fact, like LoRaWAN, it would help to have a network layer to handle adjusting transmission speed and transmit power so that nodes nearer the gateway use a wider bandwidth and reduced power while the most distant nodes run narrow band at high power, while at the same time working out the multiple access/collision avoidance problems (such as with time slots).

    The bottom line for me is this- early in my research I saw (like most do) that LoRa has an Achilles heel that will most certainly bring it down- its own success. You can't take up large swaths of bandwidth without there being some sort of problem later. The dilemma is, will LoRa congestion ever actually become a problem? Maybe yes, maybe no. In the mean time do I risk being beaten by the competition who will use LoRa?

    TI is a major player in the IOT market where bandwidth + range is king. All the research I've read points to the fact that inside buildings, 400MHz will always beat 900MHz for wall attenuation. Yet I can't get my hands on a 27dBm 400MHz demo. TI even has a 100km solution that I can't evaluate because it isn't available.

    I would love to get my hands on a narrow-band solution that can compete with LoRa. But it doesn't appear to exist.

    Question- LoRa trades bandwidth for range. Is there a modulation scheme that trades time for range? That works even below the noise floor?