image
Page 1 of 7 123 ... LastLast
Results 1 to 10 of 66

Thread: Link Margin, Transmitter Power, Theoretical Range, and Antenna Gain

  1. #1
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656

    Link Margin, Transmitter Power, Theoretical Range, and Antenna Gain

    I wanted to put together a few posts dedicated to understanding Link Margin, Transmitter Power, Theoretical Range, Antenna Gain, and Free Space Loss. There is a lot of information buried here in the FAQ section, but there really isn't a single comprehensive place to go. This is my attempt explain these terms in an easy to understand way so that those who are new to the hobby have a place to come and get the definitions and concepts in a condensed format. This will be a work in progress and the reserved posts will be updated as I have time to put stuff together.

    If you see any errors or would like to see something added to the explanation please let me know. A PM would be better as to not clutter up this thread. If you would like to tackle one of the sections please let me know, I will update the section appropriately as well as give the person credit for the work/effort that they have dedicated.
    Last edited by Derrick; 25th January 2014 at 02:31 PM.
    When nothing else out there will suit your needs... design and build it yourself.

  2. #2
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656
    Link Margin
    Link margin is a way of determining whether a specific system or setup will function under the current or desired conditions. Link margin is just an indicator of system performance and as the adage goes your results may vary. The Wikipedia page has a very good explanation of link margin so rather than reinventing the wheel:
    http://en.wikipedia.org/wiki/Link_margin

    In a wireless communication system, the link margin, measured in dB, is the difference between the receiver's sensitivity (i.e., the received power at which the receiver will stop working) and the actual received power. A 15 dB link margin means that the system could tolerate an additional 15 dB of attenuation between the transmitter and the receiver, and it would still just barely work.

    It is typical to design a system with at least a few dB of link margin, to allow for attenuation that is not modeled elsewhere. For example, a satellite communications system operating in the tens of gigahertz might require additional link margin (vs. the link budget assuming lossless propagation), in order to ensure that it still works with the extra losses due to rain fade or other external factors.

    A system with a negative Link Margin would mean the system is insufficient to transfer data, usually this means a better receiver is needed, with improved sensitivity.
    With respect to FPV and link margin there are 5 parameters that need to be determined or calculated.

    1. Receiver Sensitivity
    - Is the minimum radio frequency energy received to function (must be above signal to noise threshold), this parameter is often hard to find and must be guessed. If this value is unknown a good conservative guess is -85dB. * See Alex's explanation below of how noise floor affects receiver sensitivity.

    2. Receiver Antenna Gain
    - This is the antenna gain in decibels; this value comes directly from eh radiation plots or manufacturer of the antenna.

    3. Transmitter Power
    - This is the calculated decibel value of the radio frequency energy emitted (see Transmitter Power section)

    4. Transmitter Antenna Gain
    - This is the antenna gain in decibels; this value comes directly from eh radiation plots or manufacturer of the antenna.

    5. Free Space Loss
    - Free space loss is the loss in signal strength of radio frequency energy through free space, measured in dB. This means line of sight or no obstacles blocking the path to cause reflection or diffraction. Free Space loss is dependent on distance and the frequency of the radio wave.

    How is link margin calculated?

    LM = -RS + RxG + TxP + TxG - FSL

    Where:
    LM = Link Margin
    RS = Receiver Sensitivity (*Note: this is a negative number, a negative of a negative is a positive when plugging into the above equation)
    RxG = Receiver Antenna Gain
    TxP = Transmitter Power
    TxG = Transmitter Antenna Gain
    FSL = Free Space Loss

    I see the formula and I think I can do the math... what does it mean?

    There are three cases that you will find once you have calculated link margin.

    1. Link Margin = 0
    - This means that you are right on the edge of theoretical/ideal setup.

    2. Link Margin > 0
    - This means that you have a positive margin and you can afford some attenuation of the radio frequency energy before the system becomes inadequate... this is what you want! As a general rule of thumb a link margin of 3-6dB should be your goal. This will help account for unknown variables such as cable losses, antenna inefficiencies, transmitter power discrepancies...

    3. Link Margin < 0
    - This means that the setup is not adequate.


    How noise floor affects receiver sensitivity
    Receiver sensitivity is something that is definable and can be controlled to a point (you can select a receiver that has a high sensitivity) , however something that is beyond your control is the local noise floor. Alex (IBCrazy) provided this example to better explain how noise floor will thrash even the best of equipment.

    Anytime. Noise floor is one of those things that really kills the video link in short order in many case. It is the very thing that keeps us from being able to fly as far in populated areas.

    The short and dirty description of noise floor (there is a lot more to this, but this is quick and easy):

    "The level (amplitude) of undesired RF signal received by your receiver on your channel."

    Here's how it works:

    Let's say your noise floor is -74dbm and is caused by a HAM radio tower. Your receiver's sensitivity is -86dbm. While your RX can "hear" a signal at -86dbm under low noise conditions, the level of noise effectively reduces this sensitivity to -74dbm. This is a 75% reduction in range!

    Now let's add a VTX. Let's say at 1km your signal is received at -68db. This means that you have 1km (double distance = 6db) farther to go and your signal will be -74db... which is equal to that of the noise floor. Thus you will max out at 2km.

    Now let's say that HAM radio tower shuts down and the noise floor falls to -92dbm. At 2km you have -74dbm which your receiver can now detect. You go another 2km and you're at -80. At 8km, you're at -86 which is the minimum level of signal your receiver can detect. Now you're maxed out and limited by equipment, not noise floor. If your receiver had a sensitivity of -92dbm, you could do 16km in these conditions. However even with a -92dbm sensitivity, if that Ham radio tower turns on at -74dbm again, you're back to 2km range...

    -Alex
    Last edited by Derrick; 29th January 2014 at 01:24 AM.
    When nothing else out there will suit your needs... design and build it yourself.

  3. #3
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656
    Transmitter Power (TxP)

    Transmitter power is often expressed in Watts (W) or milli-Watts (mW). This is the measure of the actual amount of Radio Frequency (RF) energy that is output from the antenna of the transmitter. Looking only at the Transmitter Output Power (TOP) is misleading because a significantly larger TOP does not directly correlate to significantly increased range (see Link Margin or Theoretical Range).

    When calculating range of a transmitter/receiver system, a technique called link margin is used. Link margin standardizes parameters that affect range into values that have the same units. This allows simple addition and subtraction to be used to determine theoretical range or necessary antenna gain. The units that are commonly used in link margin calculations are decibels (dB). Transmitter power can be converted from mW to dB using the graph below or by looking up the value in a table from one of the links below.
    Name:  Tx_Power.JPG
Views: 2418
Size:  171.1 KB


    Links to some conversion tables:
    http://en.wikipedia.org/wiki/DBm
    http://www.cpcstech.com/dbm-to-watt-...nformation.htm
    http://www.gnswireless.com/dbm_chart.htm

    I often find myself on RapidTables.com as well:
    mW to dBm calculator
    dBm to mW calculator

    What does this mean and why do I care... Isn't more power better?

    As with everything in FPV the answer is "it depends", there are a lot of factors that come into play here. Generally speaking though the best explanation comes from the graph above, as you increase in transmitter power the the increase in dBm becomes smaller and smaller. By the time you reach the 500mW power output the net gain of increasing to higher power is so small that it really isn't worth the added noise to the rest of the aircraft. You get a much larger benefit in range by increasing your receiver antenna gain.
    Last edited by Derrick; 2nd April 2015 at 08:29 AM.
    When nothing else out there will suit your needs... design and build it yourself.

  4. #4
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656
    Reciever Antenna Gain (RxG)

    Section written by IBCrazy (Alex).
    Radiation plot tells this story well. Here are a few examples:

    Let's start with a low gain omni such as the Mad Mushroom. We'll focus only on Right hand pattern for now:

    Click image for larger version. 

Name:	1280 Mushroom horizontal pattern.jpg 
Views:	150 
Size:	50.3 KB 
ID:	40169

    The gain is ~1.5dbic.

    Now lets look at the Pepperbox (13dbic):

    Click image for larger version. 

Name:	pepperbox Horizontal jpeg.jpg 
Views:	174 
Size:	80.7 KB 
ID:	40170

    At first glance the pepperbox looks like it has only slightly better coverage, but look at the scales. Now let's see the two graphs superimposed on the same scale:

    Click image for larger version. 

Name:	pepperbox superimposed jpeg.jpg 
Views:	145 
Size:	80.3 KB 
ID:	40168

    Note that the Pepperbox has 4X the range of the MM right out in front, has the same range at +/-90, and just under half the range in the back.

    That's how you read an antenna radiation plot... and why gain isn't necessarily everything.

    What about 3D patterns? We're only getting a 2D picture here.

    The Crosshair has a perfect cone radiation plot. Thus horizontal and vertical coverage are identical:

    Click image for larger version. 

Name:	1280CROSSHAIR FINAL02.jpg 
Views:	120 
Size:	82.7 KB 
ID:	40171

    The pepperbox doesn't work this way. Note the difference in horizontal VS vertical coverage:

    Click image for larger version. 

Name:	pepperbox Horizontal jpeg.jpg 
Views:	174 
Size:	80.7 KB 
ID:	40170Click image for larger version. 

Name:	pepperbox Vertic JPEG.jpg 
Views:	120 
Size:	80.2 KB 
ID:	40172

    Note that the Pepperbox beam is not a cone, but is flatter. Why? Simple: an antenna can only "occupy" a certain size volume of air. You can change the shape, but you cannot make it any larger. The only way to change it's size is to change the efficiency of the antenna. lower efficiencies = lower volume. Many times manufacturers reduce the size of an antenna by using a PCB as a dielectric. This greatly reduces the size, but also reduces the efficiency. An antenna reduced in size by 50% will also have a 50% reduction in efficiency.

    Here's why:

    As a wave propagates you can consider the air has a "voltage stress" which is measured in volts/meter. Let's say the signal strength is 10uV/meter. If your antenna has an element that is 1 meter long and has an SWR of 1.0, the receiver will see 10uV voltage from this antenna. However, let's swap that with a PCB antenna (or one that has a loading coil) to make the antenna form factor more user friendly and the length was reduced to 25cm. Swr on the new antenna is also 1.0. The voltage seen at the receiver is now only 2.5uV. because 10uV/meter *.25 meters = 2.5uV. That's 25% of a proper antenna!

    A real World example of this is the TBS patch antenna. Theoretical gain (relative gain) on the antenna is 8dbic. However, the PCB substrate reduces the size of the antenna to 50% of a full air-insulated patch. Thus it reduces the efficiency by 50% and thus it is truly only 5dbic gain. Hence TBS marks it as the realized gain: 5dbic and not the relative gain: 8dbic.

    This is why most antenna marketing is complete garbage! Antenna makers often make miracle claims. There is no miracle. Just physics.

    -Alex
    Last edited by Derrick; 11th April 2014 at 03:54 PM.
    When nothing else out there will suit your needs... design and build it yourself.

  5. #5
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656
    Free Space Loss (FSL)

    Free space loss is the loss in signal strength of radio frequency energy through free space, measured in dB. This means line of sight or no obstacles blocking the path to cause reflection or diffraction. Free Space loss is dependent on distance and the frequency of the radio wave.

    The formula used to measure free space loss is:


    Where:
    f = frequency of the radio wave in gigahertz
    d = distance between the transmitter and receiver in kilometers


    Name:  FSL.JPG
Views: 2353
Size:  159.8 KB
    Last edited by Derrick; 11th April 2014 at 03:53 PM.
    When nothing else out there will suit your needs... design and build it yourself.

  6. #6
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656
    How do I put all this together?

    Disco Tuna has done an amazing job putting together this spread sheet so rather than recreating something similar I am just going to say go here:
    http://fpvlab.com/forums/showthread....l=1#post432618

    Another member here at the lab made a neat little web application. Though this tool will tell you theoretical max range, it does not give you link margin or account for your local noise floor.
    http://www.maxmyrange.com
    Last edited by Derrick; 20th January 2015 at 08:58 AM.
    When nothing else out there will suit your needs... design and build it yourself.

  7. #7
    Co-Pilot
    Join Date
    Jun 2012
    Posts
    1,750
    The Following is Cut and paste from a similar RF budget analysis by Hugeone of TrueRC.I found it very helpful when trying to figure this out


    In this paper, I’m going to explain the RF link « budget ». Where is that budget, how it
    can be calculated and what affect it for better or worse.

    Your budget is spread over 3 things : Tx power, Rx sensitivity and antennas gains.

    The RF currency is very often the dB (decibel, for one tenth of a bel) What is a bel? A
    bel is a ratio of 10 between a quantity and a reference level . The bel is seldomly used in
    favor of the decibel, 3 dB being a ratio of 2 following the logarithmic scale of that unit.

    Example : We often set 1mW as a reference in RF power, this is our starting point or
    0dBm (dBmiliWatt) 3dBm is 2mW, 10dBm is 10mW, 20dBm is 100mW… Same can
    apply to Volt (dBV) or Amp (dBA)

    What about Dbi? We often sa that spooky unit on antenna spec sheet. « i » is for
    « isotropic » an elusive perfect antenna that spread power evenly over a perfectly round
    sphere. So, 0dBi is a perfect sphere, 2dBi is squished a little, 20dBi is totally squeezed in
    one direction, having 100 times more power in a single direction than the perfect sphere.

    We have a 500mW Vtx, that’s 27dBm, install a dipole : add 2dbi we’re at 29dB. Now on
    the Rx side we have a patch with 8dBi with a Rx sensitive to -85dBm, that’s another
    93dB added to our budget for a total of 122dB. You can see that most of the budget is
    found in the Rx and its antenna.

    Now that our budget is known we can go spend it, how? Free space loss. As the Tx is
    moved away from the Rx a loss occur in the distance that separate the two. This loss in
    dB is calculated like this :

    32.44 + 20*log(F(MHz)) + 20*log(D(km))

    Don’t let it scare you, if math is not your cup of tea, there’s plenty of online calc for this :

    http://www.microwave.gr/content/jsffield.htm

    We see MHz in the formula and it is in a similar expression than km, yes frequency play
    an equally important role. Twice the frequency, four times the free space loss or 6dB.
    Same with distance, twice as far, four times the loss or 6dB.

    At 1280MHz, the loss over a km distance is 94.6dB (that’s free space, mean no air no
    humidity, no house, no aliens) Bummer! Only one km and most of my budget is spent!
    No, remember that the dB unit is logarithmic and like said above 6dB is twice the range.
    Let’s travel another km and we have 100.6 out of 122dB spent. Still over 21dB to go,
    divide that by 6 and we got the number of times we can double our distance. 3.5 times?
    Let’s keep it at 3 times for some overhead ☺ We’re at 2km, double #one, we’re at 4km,
    double #two, we’re at 8km, double #three, we’re at 16km, end of our journey, we spent
    all the RF link!
    Many others things can tax your RF link, solid objects between Tx and Rx will eat a lot,
    multipaths and noise are some other examples.

    How can you improve your budget? You saw how the budget is divided :

    -More power? To double your range, you need an extra 6dBm from that 500mW Vtx,
    that make a monster 2W

    -More dBi? At the Tx it won’t make much sense as you need an even radiation in order
    to bank and turn your aircraft. At the Rx, it’s easy, switch to a 14dBi patch.

    -More sensible Rx? This is also a good one, switch to a -90dBm Rx and you nearly
    double your range. However, Rx sensitivity is rarely given as spec, let alone honest
    rating of the sensitivity.
    Most of the money I made this year I spent on FPV,the rest I just wasted.

  8. #8
    Co-Pilot DiscoTuna's Avatar
    Join Date
    Jul 2013
    Posts
    1,893
    Thank you very much Derrick for taking the time to do this. This is already an excellent high-quality thread. I hope you can continue to edit the initial posts (to keep the cream at the top) as questions and requests for clarification come through.

    Obviously it is early days in your posts and your plans may be to include this, but I would like to see some kind of case study posts where you take a couple of given systems and apply all of the knowledge you are describing above and describe those systems using the above terms, and show calculations to work out some of the specifics for the system so that it brings it all too life and cements my understanding of the points you are making - (making the move from information to understanding in learning design terms).

    This is the kind of thing that makes FPVlabs a great place to come to.

  9. #9
    I see you... Derrick's Avatar
    Join Date
    Dec 2011
    Location
    Vernon, NY
    Posts
    6,656
    Quote Originally Posted by DiscoTuna View Post
    Thank you very much Derrick for taking the time to do this. This is already an excellent high-quality thread. I hope you can continue to edit the initial posts (to keep the cream at the top) as questions and requests for clarification come through.

    Obviously it is early days in your posts and your plans may be to include this, but I would like to see some kind of case study posts where you take a couple of given systems and apply all of the knowledge you are describing above and describe those systems using the above terms, and show calculations to work out some of the specifics for the system so that it brings it all too life and cements my understanding of the points you are making - (making the move from information to understanding in learning design terms).

    This is the kind of thing that makes FPVlabs a great place to come to.
    Thank you for the kind words... I have been toying with the idea of creatign this thread/series of posts for awhile now. I hope it helps.

    Yes, do an example or two. I need to get all of the posts populated first, and then I will do an example... I actually have an Excel Spread sheet that I have been using for a few weeks now to "play" with different setups, to compare and contrast theoretical/ideal ranges. Once I get it a little more polished I will post it for others to use.

    Quote Originally Posted by sparkyincali View Post
    The Following is Cut and paste from a similar RF budget analysis by Hugeone of TrueRC.I found it very helpful when trying to figure this out
    .
    Thank you for this, it is great to add a real world example. This is where I am going... I am just going to take a little longer to get there. I want to develop the concepts a little more and define each of the terms.
    Last edited by Derrick; 25th January 2014 at 10:28 PM.
    When nothing else out there will suit your needs... design and build it yourself.

  10. #10
    ImmersionRC Support Guru swillhide's Avatar
    Join Date
    Aug 2012
    Location
    Mountain House, Ca
    Posts
    8,443
    very good info Derrick! great for me to link to some "More power" guys I know.
    AMA # 1015480 HAM Callsign # Procrastinating..
    Futaba 9c w/Ezuhf /Skywalker 1880 or anything else that I can Lawn Dart.
    Retired: Disco(Sold), Skywalker 1900(Dirt Nap), Qav540g/Nazav2/Iosd mk2

Page 1 of 7 123 ... LastLast

Similar Threads

  1. Theoretical range with TBS recommended gear
    By margoye in forum IFR - Video Link Discussion
    Replies: 4
    Last Post: 4th January 2014, 01:15 PM
  2. Replies: 11
    Last Post: 6th October 2013, 01:46 PM
  3. Replies: 10
    Last Post: 6th August 2013, 09:44 PM
  4. What does antenna gain mean? (article)
    By dfallon in forum L-Com
    Replies: 0
    Last Post: 3rd April 2013, 09:01 AM
  5. Replies: 2
    Last Post: 3rd April 2012, 06:37 AM

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •