Power line channel transfer function

Hello?? Can someone help me with this problem?

-Jim

Hi!

Clarify your meaning, please. Could you describe your task circumstantially?

Regards, Igor 

Hi

      Ok, I'll try to re-explain my project. My project is about calculating the channel transfer function of the powerline channel over the frequency ranging from 50-500 kHz. I'm using the PLC development kit V3. Currently, I'm able to let the transmitter and receiver transmit and receive packets alternatively using the program come with the kit. I'd like to know how to program the PLC development kit to calculate the channel transfer function...

Hope this helps

-Jim

Hi!

1 Unfortunately I didn't understanand right what do you want to calculate. But, if you want to measure the max. data transfer distance, I would advise you to use the physical model of a communication line. It should construct the scheme of communication line from resistors, capasitors and inductors. The parameters of elements must correspond running parameters of your real communication line. The measurements of max. transfer distance can use the standard firmaware of kit V3.

2 You can use the phy_example_project from some package for kit V3.

Regards, Igor 

Hi Igor

    Thank you for your information. Can you tell me how to use the phy_example_project to determine the physical model of a communication line (R, C, L)?

Regards,

-Jim

Hi!

For the creation of physical model you can use for example the power line parameters published at 4478.Channel modeling.pdf (refer on page 4). I would recommend you pay attention on following materials also:

1 5270.On_the_Channel_Capacity_of_the_Residential.pdf 

2 0523.Power - Line As Access Medium.pdf

3 6153.Analise_of_PowerLine.pdf

4 4643.Medium Voltage Cables.pdf

When your equivalent of power line  will be ready you can begin your experiments. You should connect one modem to the begin of the equivalent and other modem is connected to some point of the equivalent. ATTENTION! You shouldn't connect to your line the some power source (AC or DC). The experiments are executed on could wire!

Further you should run on bouth PC-host the PHY-test using the PLC Quality Monitor aplication which is provided with the package of KIT. At this case you will use the standard firmware of KIT. You can choose a modulation-mode, the gain of transmitter and receiver. But  for experiments you can use the phy _example_project too. At this case the interaction with PLC Quality Monitor aplication isn't supported & the control of a signal propagation you can do with an oscillograph or a spectrum analyser only. Also you will need CCS for building phy _example_project.

Certainly the best way is experiments on real power line when you are going to use your PLC-modems. Unfortunately it is not possible allways.

At the end I would recommend you pay attention on my post G3-PLC+TMDSPLCKIT-V3. I think it is useful for you.

Regards,

Igor 

Hi Igor

   Thank you very much for your information. Fig 3-9 in 4478.Channel modeling.pdf (Attenuation vs frequency) are what I really want to have. However, if I understand correctly, based on the information that you provided, does it mean that Fig 3-9 can be obtained using a network analyser only? I saw that PLC Quality Monitor has the output SNR but I have no idea how to vary the transmit and frequency. For the phy _example_project, how can I vary the transmit frequency and monitor the attenuation without the oscilloscope? I believe there should be APIs to do this but not sure where to find it… 

Thank you in advance

-Jim

Hello Jim!

Indeed, if you will use standard firmware of KIT-V3 and PLC Quality Monitor only you can choose a modulation scheme, the gain of transmitter & receiver. With PLC Quality Monitor you can watch SNR & bit/packet error statistics. The G3-PLC - version of phy _example_project provides to tune OFDM-signal with greater opportunities (for example you can tune OFDM-spectrum very flexibly). Using the PRIME-PLC-version of phy _example_project you can choose a modulation scheme, the gain of transmitter & receiver only (you can't tune frequency). Unfortunatelly at bouth case it is impossible to use PLC Quality Monitor. What kind of PLC-package for TMDSPLCKIT-V3 you have (G3 or PRIME, what is version?)? What is materials included in you package (API-descriptions, SW-examples etc.)? Did you pay attention to my post G3-PLC + TMDSPLCKIT-V3?

Regards,

Igor

0160.7585.ti_plc_prime_phy_example.zip

Hi Igor

     Thank you very much for your quick response. My Zero Config GUI version is 2.90. I got a project phy_example_project (attached here) from Woon Soo and believe it's PRIME since I cannot tune frequency. The material that I have so far are TI_PRIME_UG.pdf and TI_plcSUITE_phy_api.pdf. I browsed through your posts and found that you've requested for some APIs for FLexOFDM which were not supported anymore...

     Based on your comments, I don't think it's fair to say that we can adjust only the gain of the transmitter or receiver, MCS or tune the frequency. My organization have invested in TI product to be able to modify more communication system parameters. Our goal is to develop our own communication system protocol and it's quite disappointing to know that I can adjust very limited parameters. By the way, do you know if there's any APIs for PRIME that allows me to monitor the SNR through phy_example_project or who I should contact? There're functions PHY_rxSmRun and PHY_txSmRun where I might be able to monitor the SNR but I could not find the source code of these 2 functions so please let me know if you've looked through these functions?. Finally, where can I get the code for G3?

Thank you very much 

-Jim

Hi!

Yes, FLexOFDM is not supported anymore (I have requested plc_support@list.ti.com). This protocol had many advantages above G3 & PRIME. The main advantage is the wide range of possibilities for the formation of the OFDM-signal. But mow TI supports G3 & PRIME only. However, it relates to new processors F28069 & F28M35x, wich is used into TMDSPLCKIT-V3. For TMDSPLCKIT-V2 based on F28335 the API of FLexOFDM exists (at last year I have downloaded this API & I was executing some experiments with TMDSPLCKIT-V2). Also I was executing some experiments with TMDSPLCKIT-V3 based on F28069. At this case I used API-PRIME and API-G3 from following packages  TexasInstrumentsPrimeDevelopmentPackageV6_0_0_0.exe & TexasInstrumentsG3DevelopmentPackageV5_2_1_0.exe (unfortunately I can't upload these in the post because there is a limit 20MB; you can replay me your email or request to plc_support@list.ti.com or to Kim Wonsoo: wonsoo.kim@ti.com).So...With regard to measurements of "attenuation vs frequency" for communication line it seems the use a sweep frequency generator & spectrum analyzer is a most better way. After these measurements the task of choose  an optimal protocol for PLC-modem will be simpler.

Regards,

Igor 

Hi Igor

    Thank you very much for your information. I've sent an email to Kim Wonsoo:wonsoo.kim@ti.com to request for Prime and G3 APIs but have not heard anything from him so far. By the way, could you please tell me what kind of project you can play with phy_example_project? After sending and receiving packets with blinks of LEDs, I'm not sure what to do next...

Thank you very much

-Jim

Hello Jim!

Unfortunately it happens. TI-PLC-branch answers not always. I'd offer you my phy_example_projects for TMDSPLCKIT-V3 (its worked successfully) 7115.TMDSPLCKIT-V3.rar.

Regards,

Igor 

Hi Igor

     Thank you very much for the files. We'll try to run yours and see what happen.

-Jim

Hello Jim!

When you will work with my projects you should pay attention to setting levels of reception and transmission. They must be put to maximum (refer to PRIME_PHY_API and G3_PHY_API). This way you will achieve the maximum possible communication distance. For example for G3-project (the file main.c):

the functions of G3-API  PHY_txPreparePpdu(&PHY_tx_ppdu_s, cb_tx) and PHY_txPpdu(&PHY_tx_ppdu_s, cb_tx) use structure

PHY_tx_ppdu_t PHY_tx_ppdu_s =
{
0, //*ppdu
73, //len
32, //level of transmitter 0 - 32(max)     !!!
0, //mcs//0-ROBO 1-DBPSK 2-DQPSK 3-D8PSK
0, //tm
{0, 0}, //txgain
0, // dt
0, // rpt
0 // txTime
};

For the gain of receiver the G3-API function PHY_rxSet(PHY_RX_SET_AGC, (PHY_rxSetData_t *)&agc_gain_s) uses structure

//Receiver Gain
PHY_rxAagcSet_t agc_gain_s =
{
1,//0-auto 1-manual
7//0-7 step 6dB (42 dB max)
};

Regards,

Igor

Hello!

By the way here is a picture of our equivalent of our communication line 1346.IMG0024A.rar. The TMDSPLCKIT-V3 I tested on this equivalent. Resistors powerful as the real line is used for power DC supply of some devices and we connect 36 volts DC to create the real conditions of the data transfer. Each resistor and capacitor simulates 1km. Although inductive losses and loss of conductivity insulation are not considered. In our case they are sufficiently small as compared with the capacitive and resistive losses. 

Regards,

Igor

Hi!

Also you can visit the post http://e2e.ti.com/support/microcontrollers/c2000/f/171/t/264334.aspx. An author of this post solves a similar  task.

Regards,

Igor