• State
  • Locked Locked
  • Replies 30 replies
  • Subscribers 19 subscribers
  • Views 6329 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.

Which RF would you use ?

C
Expert 2065 points
Other Parts Discussed in Thread: CC1350, CC2650, CC1310, TPL5111, TPS22860, CC110L, BQ25120

Hello, 

We are currently working on a project and we would like to get your advice about the confirmation regarding which RF to use for a project.

The project is about a wrist that is use as a Home curfew system for prisonner in custody in their appartment. 

We are thinking to use a low frequency device for the wrist item but we are worry about several things: 

1./ Range , we need it around 50m radius

2./ Battery consumption of the wrist (we need it to work about 2-3 years)

3./ Accuracy of signal 

4./ Cocnrete wall faraday cage effect

I would be happy if some Ti specialist could give their opinion about which frequency raneg they would use, which protocol and which chip as well as some advices about antenna.

Thank you in advance for your help and wishing you a great day.

Regards

  • 0
    Guru 735695 points
    I think CC2650 or CC1350 would be suitable for high performance and low power RF application.
  • 0 in reply to YiKai Chen
    Expert 2065 points
    I am concern by the range, the signal and the size of the associated antenna.
    Can you develop about your motivation on choice of the chip you are proposing ?
    Other TI specialist opinions are also welcome.
  • 0 in reply to C
    Guru 735695 points

    CC2650 can provide good link budget with low power consumption. Running BLE, you can have link budget upto 102 dbm. Running 802.15.4, it can have link budget upto 105dbm. These numbers would make 50 meters line-of-sight transmission without any problem. And the power consumption of CC2650 is the most low as I know.

  • 0 in reply to YiKai Chen
    Expert 2065 points
    Hello,
    As I mentioned we will not be in line of sight but in apartment with concrete wall. Reason why my question
  • 0 in reply to C
    Guru 31280 points

    I wouldn't choose 2.4GHz for communication in environment of "concrete wall Faraday cage effect", even for 50 meters distance. A sub 1GHz solution will have a better performance in this case. Regarding battery consumption, both the cc13xx and cc26xx family are very low power and the actual consumption will be very dependent on usage, in other words, the more time a device is in sleep mode, the better the consumption.

    I don't quite understand the meaning of "Signal Accuracy".

  • 0 in reply to C
    TI__Mastermind 18780 points
    Then I would go for the CC1310, then you will not have any range issues:) You can test the range with one of our kits to evaluate the performance.
  • 0 in reply to Igor Sherer
    Expert 2065 points
    Hello, thank you for replying. What frequency would you recommend ? Regarding signal accuracy, we have to detect the transceiver going and out a range of 50 m radius.
  • 0 in reply to C
    Guru 735695 points
    CC1310 is a good suggestion It works on subG Hz and would provide better penetration to concrete wall.
  • 0 in reply to C
    Guru 31280 points
    Range estimation based on RF signal with a precision of 1 meter tolerance (just saying) is a tough task to accomplish. I would go further and say it's practically impossible in an environment such as in your case (concrete walls medium).
    You might consider another approach, like combining a low power RF transceiver for communication and an ultrasonic transceivers for distance measuring. There are of course other solutions, like Doppler transceivers (there are 2 flavors, 10GHz and 24GHz, google it for a bit)
  • 0 in reply to Igor Sherer
    Expert 2065 points

    We actually don't have a requirement of 1 meter precisely. We can accept probably more.

    Your help and recommendation is well appreciated.

  • 0 in reply to C
    Expert 2065 points

    Hello again

    We have move forward on this project and I would appreciate again to have your feedback for a few things.

    As I explained before we have two parts on this project, one base station and one wearable.

    We are currently trying to achieve the werable part as follow:

    We have selected the CC1310 as transceiver because it is well adapted to sub 1GHz radio frequency, also it contains an AMR M3 so no need to get another mcu apart.

    We are also using a fiber optic transmitter and a fiber optic receiver which are continuously on, we have decided to use Everlight module for this.

    1./ We have checked the reference design which are provided on TI website and they all use non rechargeable batteries, they use cell coin CR 2032. We want to use Lipo or Li ion battery so the device can be recharged with USB.

    We are a bit confused about the parts we should use and how potentially this change from non rechargeable to rechargeable battery woudl affect the cricuit.  We have understood mostly what the nano power system do but we are unsure if this nano power system and the DC DC converter already on reference design would be compatible with the Lipo battery being interfaced via the BQXX device.

    We are unsure which BQ device we should use. There are so many, that we don't know which one we should use.

    Can you please advice which BQ we should use and also confirm wether we can/have to keep the TPL5111 and TPS22860 into the power managemen circuit ?

    2./ Also, we are wondering wether or not we should connect the optic fiber transmitter and receiver on the GPIO of the CC1310 or rather on the sensor port of it.

    Can you please recommend ?

    3./ We plan to have a power switch/key to turn on turn off the device, we are unsure wether we have doubt where this power switch should be connected

    Can you please recommend ?

    Your help and time is very highly appreciated

    The block diagram we have established is as below so far.

  • 0 in reply to C
    TI__Guru**** 317180 points
    - I would recommend you to post question 1 in the forum that covers the devices in question. This forum covers RF devices and we don't have the overview in the power domain.
    - The sensor domain is primarily for analog signals or infrequent digital inputs. Since you want to transfer a datastream I suspect that using a GPIO would be best.
    - Load switch: If you want very low power consumption you need to turn on/ off the supply to CC1310, eq VDDS.
  • 0 in reply to TER
    Expert 2065 points
    Hello,
    Thank you for your reply.
    Well the optic fiber is used as tamper system in here. We just send light and the sensor look if some ligh is here or not. If not, then the MCU will send a signal to CC1310 for transmission and on the other side the receiver will do something.

    I have also another question: one of our guy started the receiver design with CC1101L while we are considering here CC1310, does those two chip can communicate properly you think, or do we take risk in trying to make two different chip to talk to each other?

    Also, what kind of mode and what kind of coding / module woudl you recommend ?

    Regards
  • 0 in reply to C
    TI__Guru**** 317180 points
    As a tamper system you can probably set up the GDO pin as a interrupt source, eq when the light turns off you get an interrupt.

    As long as CC110L and CC1310 uses the same datarate, deviation and modulation format they can communicate without any problems. The CC1310 can use the same packet format as CC110L.

    Not sure if I understand your mode/ coding question. Could you elaborate?
  • 0 in reply to TER
    Expert 2065 points

    I think you partially answered my questons with the datarate, deviation and modulation. Now, we need to implement it and this is another story (but we will eventually success in it with a bit some perseverance and patience)

    In order to achieve our poject can you please point a few ressources you recommend to digest properly about radio frequences, modulation, deviation, datarate, antenna and pcb design oriented to RF consideration so that we can accelerate our learning curve ?

  • 0 in reply to TER
    Expert 2065 points

    I have another question:
    1./ i am currently drawing the reference design and I see (using several reference design related to CC1310 that decoupling capacitor blcok for VDDR and VDDS have different value from reference design to reference design.
    How do we set the number of capacitor to add up in parallel for each of those block ?

    2./ i see on the reference design CC1310 EM-7PA-4751 that the block between the CC1310 and the antenna is is different -it use a skyworks components  and a discrete PA block) than on the other other schematic CC13xxEM-7XD-7793 and CC13xxEM-5XD-7793_4L

    Why this difference ?

    3./ I see on the schematic that some capacitor and resistor are marked "DNM".

    What is the meaning of DNM ?

  • 0 in reply to C
    TI__Guru**** 317180 points
    1) Use the reference design that is closest to your needs. Eq if you intend to use a 4x4, use the reference design for 4x4.

    2) This is a reference design for the China with a discrete PA. See www.ti.com/.../swra490.pdf

    3) DNM: Do Not Mount
  • 0 in reply to TER
    Expert 2065 points

    Hello, 

    Thank you for your answer, still, let's take the EM-7XD-7793, if you look at R12, it is DNM, then what do you replace it with (just a normal connection ?). If R12 is not mounted then the antenna can not be connected to the chip.

    I plan to use 7x7 chip, however I see that CC1310 didn't implemente yet for 433Mhz, in Asia it seems to be the ISM available.

  • 0 in reply to C
    TI__Guru**** 317180 points
    The default on the reference designs is to use the SMA connector.

    See e2e.ti.com/.../472245 for 433 MHz.
  • 0 in reply to TER
    Expert 2065 points

    0734120110_RF_COAX_CONNECTORS.pdfHi, 

    Thank you for your answer. 

    I see in this thread there is a first version done but we don't know if this design is validated or not as they say another version were about to be completed in december then later in feb 23 someone ask if this one is completed but the people from TI answer the last version is from November.

    Can you confirm this release: CC13xxEM-7XD-4251_1_1_0.zip  is the stable release one ?

    Regards

    EDIT1: Another question: 

    Those design are based on using either helix antenna or SMA connector. 

    In case we use SMA  connector, I assume we can place the antenna in any position ? Do we need to create  shielding on the top of cicruitry to protect the circuitry from the antenna radiation and interference ?

    EDIT2: I am willing to replace this big SMA connector by a small microcoaxial plug such as this in the link below. Is it going to be a problem you think ? if yes, which smaller plug would you recommend? 

    http://www.molex.com/molex/products/datasheet.jsp?part=active/0734120110_RF_COAX_CONNECTORS.xml&channel=Products&Lang=en-US

  • 0 in reply to C
    TI__Guru**** 317180 points
    The 1.1.0 revision is a stable one and it's the one used in the lab at the moment.

    You can place the antenna in any position, just keep the distance to the ground plane. No shield required.

    No issues using the Molex part, we have started using a similar part from Murata on the Launchpads. Before deciding on the part, check that you are able to find a cable compatible with the connector with wanted performance. Also check if the part is spec'ed for the required number of insertions.
  • 0 in reply to TER
    Expert 2065 points

    I would like to submit my schematic for review. I have assembled the schematic you propose and the the schematic of the TIDA 00712.

    Before moving forward to prototype and potentially starting to trash some pcb, I would appreciate to have someone from TI to review the assembly of those two schematic has been done properly.

    How can we process ?

  • 0 in reply to C
    TI__Guru**** 317180 points
    Send me a fried request and I can take a quick look at it offline.
  • 0 in reply to TER
    Expert 2065 points
    I have sent you the friend request.
  • 0 in reply to C
    Expert 2065 points

    I have sent you the file via PM

  • 0 in reply to C
    TI__Guru**** 317180 points
    I sent a comment to the schematic on Friday as a PM.
  • 0 in reply to TER
    Expert 2065 points
    Sorry, I didn't see it earlier. I got it, and I have replied.
    I am mostly concerned about the power management system I have used which is the BQ25120 coming from the other schematic
    Also, i am about to layout the PCB now, do you have any design guide rules for this ? Track width, distance related to the antenna strip etc... ?
  • 0 in reply to C
    TI__Guru**** 317180 points
    For layout, please follow our reference designs (www.ti.com/.../technicaldocuments)
  • 0 in reply to TER
    Expert 2065 points

    Hello

    In the note etc... there are some design who are integrating a Skyworks component seems to be to get a better signale.

    Should we also use this on our design ?

  • 0 in reply to C
    TI__Guru**** 317180 points
    It depends where and how you want to use the design. Dependent on the region you are allowed to use different output power. The use of a FEM (Front End Module) is a system decision.