• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 32 subscribers
  • Views 759 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

THVD8000: Questions for THVD8000 application

Jacky Wang(QD)
Genius 11010 points
Part Number: THVD8000
Other Parts Discussed in Thread: THVD8010, THS6222

Hi Team,

Our customer is evaluating THVD8000 to upgrade their existing Homebus solution in commercial A/C applications, and they want to check for below questions, can you help to answer?

1. They want to achieve 19200bps data rate for over 1000m cable distance, can we support it with THVD8000?

2. With above conditions, can we support up to 256 nodes?

3. Can you help to share one reference design with full duplex communication with THVD8000?

Thanks.

  • 0
    TI__Mastermind 48057 points

    Hi Jacky,

    Thanks for reaching out!

    1. At 1km @ 19.2kbps (minimum OOK modulation for +/- 2% duty cycle distortion is 192KHz - closest available modulation frequencies are 187.5KHz (which could cause more duty cycle distortion ) and then it goes to 500KHz - using the THVD8010 you can choose 300KHz - so the THVD8010 may be the best place). However at that high of a modulation frequency acts similarly as a data rate 2x the mod frequency. So using the THVD8000 at 500KHz is like 1Mbps and THVD8010 at 300KHz is like 600Kbps. Using the below chart we can get an approximate estimate:

    So looking at the Red section - it is going to be close at the 300Khz due to the length of the cable (this also is assuming a cable with 120 Ohm characteristic impedance)

    So the answer is "maybe" but the attenuation may  become problematic near 1000m due to AC losses becoming higher than the driver can overcome. To avoid this issue the  implementation of a line driver on the output of the THVD8000 or THVD8010 can be used - please see app note on implementing a line drive (it says High voltage but it also applies to low impedance systems - which this could potentially be considered one) App Note: https://www.ti.com/lit/an/slla590/slla590.pdf?ts=1661441160857

    2.  Depends on how it is implemented. It cannot be done the way that is suggested in the data sheet because the number of transceivers will make the inductors impossible to size correctly as with 256 transceivers there is no more "impedance" budge left to add any other connection to an "ground". Essentially the loading impedance becomes too small for the THVD8000/THVD8010 to drive. However - if you take a look at that application note from the first point - that application use case is for High voltage and/or low impedance systems - so it is possible but it may be a bit more complicated at first glance. 

    3. Please see application note from point 1- it includes a full duplex communication with line driver attached as a schematic -all the tests were done with the individual parts EVMs connected together  - since this device is half - duplex each node requires 2 THVD80x0 devices. 

    Please let me know if you have any further questions.

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    TI__Genius 11010 points

    Hi Parker,

    Thanks for the comments, regarding to THVD80x0+THS6222 solution, can you help to share the source reference design files for schematic and PCB? We can share them with customer for their further design. BTW, if we want to reduce the cost in further, can we just use only 1pcs THS6222 in master node or slave node side to meet above customer's design requirements?

    Thanks.

  • 0 in reply to Jacky Wang(QD)
    TI__Mastermind 48057 points

    Hi Jacky,

    I have added a few slide decks on the process behind this application as well as a schematic file.

    One thing to note is that we didn't build a specific board when testing  the application with the THS6222 + THVD8000 - we used already existing EVMs (detailed in attachments) and the schematic I provided is just the separate EVM schematics placed onto 1 page. Everything here can be shared - none of its confidential (just as a FYI); its just supporting material for the application note attached in my first reply. 

    THVD8000_line driver_AC power.pptx

    THVD8000_line driver_solar.pptx

    THVD8000+line driver AC board setup (006).pptx

    Schem1_THVD8000_PLC.pdf

    Schem1_THVD8000_PLC.SchDoc

    Every node that communicates with other nodes on the line requires a THS6222. If there are nodes that only receive data and never transmit you wouldn't need the THS6222 there (but the protection and HV interface circuits are still necessary in this case). So it depends on the direct use case where you can remove the drivers. As a note - this is one of the simpler implementations of power line communication (very little firmware and both the THVD8000 and THS6222 ) - and the vast majority, if not all, will require some sort of line driver + data modulation (in this covered by the THS6222 and the THVD8000). The one thing that can lessen the cost of the system is to include only 1 THVD8000 per node instead of the recommended 2 THVD8000's (figure 1.2 in the application note I sent earlier in the thread). This method makes layout harder as its not as easy to mitigate sources of EMC issues and signal reflections which could complicate the design process - but it is possible and will make the solution have 1 less THVD8000. We have only tested down to 1 Ohm loads so if the 256 nodes in parallel have an effective impedance of >= 1 Ohm it should be okay  to work. 

    Please let me know if you have any further questions!

    Best,

    Parker Dodson

  • 0 in reply to Parker Dodson
    TI__Genius 11010 points

    Hi Parker,

    Thanks for the detailed reply. I'm still having one question about the THS6222 driver connection, it seems that it is suitable for AC power line system, but how to connect it into DC power line system? Can you help to suggest how to modify below circuit to adopt DC bus system?

    Thanks.

  • 0 in reply to Jacky Wang(QD)
    TI__Mastermind 48057 points

    Hi Jacky,

    The above circuit can work for both AC and DC. The circuit you show above is how the data is connected to the shared line - the power signal doesn't need come back to this node through the data interface. (Power will go somewhere else - in the above system power is being sunk into a rectifier + DC/DC converter - if its just DC than it would just be the DC/DC converter). The capacitor and transformer make a high pass filter to allow data but block DC/ Low frequency AC power from the line. The capacitor used in the reference is 1.5kV to handle the large amount energy it will blocking. The AC_Mains_Line would be the positive terminal of the DC source where the AC_Mains_Neutral would be the return path to the DC power source. Pl

    The above design only shows how to connect the data node to the shared power/data line in order to transmit/recieve data over the coupling but block power signals. The design is pretty independent from how it is setup as long as the HV interface provides good coupling for data and protection from high voltage systems. 

    Please let me know if you have any other questions!

    Best,

    Parker Dodson