Connect with us

Video signal amplifier.

Discussion in 'Audio' started by HANKMARS, Jul 28, 2019.

  1. hevans1944

    hevans1944 Hop - AC8NS

    4,195
    1,985
    Jun 21, 2012
    If I Google "E111272 (05-21513) 18 AWG SHIELDED" this web page of results comes up. This is less than useful. We need to know exactly what kind of wire you are trying to work with. If it is truly solid and not stranded, that is a huge problem because solid wire is not very flexible, it doesn't spool and un-spool easily. All of the example cables offered by the Google search results are for stranded-wire cables, most with an aluminum foil shield around each pair and a separate drain wire for the shield. If you can, get a pair of diagonal cutters and grab a one-foot sample of the cable you are trying to drive. If there are markings on the cable, please try to include those in the sample. Strip off about four or five inches of the outer jacket to reveal the twisted pairs underneath. Splay these out and take a close-up in-focus picture of what you have. Something similar to this:
    [​IMG]
    Notice the cable shown above consists of two twisted pairs, one black wire and one white wire in each pair, each wire being insulated, stranded 18 AWG copper, and each pair of the twisted-pair being covered with its own aluminum foil shield, with a third overall aluminum shield covering both pairs.

    Does this look anything like what you are trying to drive video down 500 feet of? If so, you are SOL. This is low-voltage, low-frequency, instrumentation cable and it is totally unsuitable for the task. I am surprised there is anything left of the horizontal sync signal after 500 feet, much less anything left of its back porch with the six cycles of 3.58 MHz chroma sync that are needed to recover the color signal. The horizontal line rate of NTSC television is approximately 15,750 lines per second, which means there is roughly 63.5μs available on each line to transmit both the horizontal sync and blanking pulse along with the video, the latter squeezed in between successive sync pulses.

    Connect an oscilloscope with at least 10 MHz bandwidth to the lipstick camera video output, and synchronize the 'scope so it triggers on a horizontal sync pulse while its linear time base is set long enough to display at least two sync pulses. You will see, between two consecutive sync pulses, the actual analog video signal that will be displayed as one intensity-modulated line (out of 525 lines) on a monitor. This is pretty much useless information presented in this form on an oscilloscope. What is useful is seeing the presence of six sinusoidal cycles of 3.58 MHz "color burst" information. These six cycles are but a sample of the continuous 3.58 MHz oscillator signal that was amplitude-modulated with in-phase and quadrature-phase components of the color portion of the video signal, and then added to the luminance (black and white) portion of the video signal to produce the composite video of a single displayed line that you see in between horizontal sync pulses.

    There is no way you can see this complicated series of events occurring in real-time on your oscilloscope display. Decoding and displaying the composite video as a color moving picture is what the television monitor does. What you can do is observe the degradation in the composite video signal as it moves away from the lipstick camera and toward the color monitor, along wires that were never intended to faithfully carry this type of signal.

    So, if you can, gather up the cable sample, strip off the outer insulation to reveal what is underneath, and take some well-lit, in-focus, close-up pictures and post them here. Once we know what the cable is, maybe someone here can offer you a way to drive the cable that will preserve the chroma information contained in those six cycles on the back porch of the horizontal sync signal component of the composite video.

    I would also investigate using fiber-optic cable instead of copper cable if your "customer" would consider starting over. Take a look at this YouTube video to see what is possible. Google AD-net technology Taiwan for more information.
     
    HANKMARS likes this.
  2. aurelZ

    aurelZ

    55
    7
    Jun 10, 2019
    " If it is truly solid and not stranded, that is a huge problem because solid wire is "
    really ..
    since when is this a problem?
    problem is in wire diameter on such a large distance
    and must be 1 mm minimum

     
  3. aurelZ

    aurelZ

    55
    7
    Jun 10, 2019
    From AD811 data :
    Flexible operation Specified for ±5 V and ±15 V
    operation ±2.3 V output swing into a 75 Ω load (VS = ±5 V)

    this looking very good ,and should work well in your case!
     
  4. hevans1944

    hevans1944 Hop - AC8NS

    4,195
    1,985
    Jun 21, 2012
    [/QUOTE]
    Not a problem when you selectively quote fragments of my original text:
    This was not a comment about wire size, it was a comment about my doubts that 500 feet of two pairs of solid, shielded, 18 AWG twisted-pairs, could be easily wrapped around a six-inch diameter spool and easily fed out and retrieved as needed. In other words, I was trying to politely ask the OP to verify what kind of cable he had to work with.

    Well, if that were the problem, 1mm is the same as 18 AWG. It doesn't make any difference whether it is stranded or solid as both solid 18 AWG and stranded 18 AWG have the same effective diameter. The only difference is stranded is more flexible than solid. Using solid wire in an application where the wire must be repetitively un-wound and then re-wound on a spool is an invitation for disaster because copper wire will "work harden" and eventually break under those conditions.

    It appears to me that your statement that the "wire diameter ... must be 1mm minimum" is a pure rectal extraction, unless you simply translated the original 18 AWG that the OP specified into its metric equivalent, which is 1.02mm. How clever of you to mix metric and Imperial measurements that mean the same thing.

    It does look good, as others before you have pointed out, if it is driving 75Ω cable properly matched on each end.

    That means a 75Ω source impedance feeding a 75Ω transmission line terminated at the far end with a 75Ω load. Even if you get all the impedances matched, there is still the question of signal attenuation as a function of frequency and cable length. We have no idea what that is since the OP has not provided any relevant specifications for the cable he has.

    If the OP was capable of successfully building such a do-it-yourself project based on the AD811, there would still be the problem that the customer cable is probably not anywhere close to 75Ω impedance, probably was not even manufactured as "impedance controlled" cable, and given how the cable loses the color sync information (and probably the chroma component of the composite video too), probably has too much attenuation and not enough bandwidth to propagate color-encoded NTSC video the requisite distance. Wire gauge has very little to nothing to do with that: it's all about geometry and the materials of construction determining cable impedance and signal attenuation.

    But, at less than five bux each plus shipping and handling, and being experienced in this sort of thing, I would buy a pair of AD811 amplifiers and give it a go. Nothing ventured, nothing gained. And the customer already has plenty of cable to test it out on. Experimentation is what this forum is all about. The LM747 experiment has simply demonstrated, as Thomas Edison was fond of saying, just one way that doesn't work. Why not try again with a different amplifier to see if there is possibly another way that does work? As we know now, Tom Edison was guaranteed to eventually succeed, because filamentary tungsten light bulbs can now be found everywhere that they have not been replaced with "mo betta" technology... LEDs for example. My crystal ball is a little hazy on this one though.
     
    HANKMARS likes this.
  5. WHONOES

    WHONOES

    651
    130
    May 20, 2017
    HANKMARS likes this.
  6. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    Why I said it was solid wire, I am not sure. Definitely stranded and not twisted pair. Foil shield with stranded nickel plated 18ga copper ground wire. DSCF2274 (2).JPG DSCF2272 (2).JPG DSCF2274 (2).JPG DSCF2272 (2).JPG
     
  7. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    Not a problem when you selectively quote fragments of my original text:

    This was not a comment about wire size, it was a comment about my doubts that 500 feet of two pairs of solid, shielded, 18 AWG twisted-pairs, could be easily wrapped around a six-inch diameter spool and easily fed out and retrieved as needed. In other words, I was trying to politely ask the OP to verify what kind of cable he had to work with.


    Well, if that were the problem, 1mm is the same as 18 AWG. It doesn't make any difference whether it is stranded or solid as both solid 18 AWG and stranded 18 AWG have the same effective diameter. The only difference is stranded is more flexible than solid. Using solid wire in an application where the wire must be repetitively un-wound and then re-wound on a spool is an invitation for disaster because copper wire will "work harden" and eventually break under those conditions.

    It appears to me that your statement that the "wire diameter ... must be 1mm minimum" is a pure rectal extraction, unless you simply translated the original 18 AWG that the OP specified into its metric equivalent, which is 1.02mm. How clever of you to mix metric and Imperial measurements that mean the same thing.


    It does look good, as others before you have pointed out, if it is driving 75Ω cable properly matched on each end.

    That means a 75Ω source impedance feeding a 75Ω transmission line terminated at the far end with a 75Ω load. Even if you get all the impedances matched, there is still the question of signal attenuation as a function of frequency and cable length. We have no idea what that is since the OP has not provided any relevant specifications for the cable he has.

    If the OP was capable of successfully building such a do-it-yourself project based on the AD811, there would still be the problem that the customer cable is probably not anywhere close to 75Ω impedance, probably was not even manufactured as "impedance controlled" cable, and given how the cable loses the color sync information (and probably the chroma component of the composite video too), probably has too much attenuation and not enough bandwidth to propagate color-encoded NTSC video the requisite distance. Wire gauge has very little to nothing to do with that: it's all about geometry and the materials of construction determining cable impedance and signal attenuation.

    But, at less than five bux each plus shipping and handling, and being experienced in this sort of thing, I would buy a pair of AD811 amplifiers and give it a go. Nothing ventured, nothing gained. And the customer already has plenty of cable to test it out on. Experimentation is what this forum is all about. The LM747 experiment has simply demonstrated, as Thomas Edison was fond of saying, just one way that doesn't work. Why not try again with a different amplifier to see if there is possibly another way that does work? As we know now, Tom Edison was guaranteed to eventually succeed, because filamentary tungsten light bulbs can now be found everywhere that they have not been replaced with "mo betta" technology... LEDs for example. My crystal ball is a little hazy on this one though.[/QUOTE]
    Edison liked to import some of his findings from other peoples experiments.
     
  8. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    Edison liked to import some of his findings from other peoples experiments.[/QUOTE]
    I am not quite done with the LM747. The reason, I believe, that I was losing the low end of my signal was because I neglected to use a voltage divider to pseudo provide a dual supply and therefore using my minus full rail value as signal ground.
     
  9. hevans1944

    hevans1944 Hop - AC8NS

    4,195
    1,985
    Jun 21, 2012
    Well, go ahead and try that, and please let us here on the Mothership know how it turned out. From my experience, messing with composite video is a very tricky business, what with dc restoration of the blanking level and knowing how to pass a 4MHz bandwidth video signal without distortion, and preserving the color-burst information that resides on the back porch of the horizontal sync signal. IMHO any serious work absopositively requires an oscilloscope.

    BTW, it looks like the cable you have is ordinary shielded instrumentation cable, four 18 AWG insulated wires inside a foil shield with a so-called drain wire for the shield and a plastic cover over all. Belden 9418 appears to match this description (see image below).
    [​IMG]
     
    HANKMARS likes this.
  10. aurelZ

    aurelZ

    55
    7
    Jun 10, 2019
    Sorry h1944
    when you right you right !
    OK!
    So he need to rool and unrool caable every day ?
    That really sucks,...
     
  11. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    I am very anxious to get "restarted" on this project. Parts, even just the nuts and bolts resister kit, expire
    Not a problem when you selectively quote fragments of my original text:

    This was not a comment about wire size, it was a comment about my doubts that 500 feet of two pairs of solid, shielded, 18 AWG twisted-pairs, could be easily wrapped around a six-inch diameter spool and easily fed out and retrieved as needed. In other words, I was trying to politely ask the OP to verify what kind of cable he had to work with.


    Well, if that were the problem, 1mm is the same as 18 AWG. It doesn't make any difference whether it is stranded or solid as both solid 18 AWG and stranded 18 AWG have the same effective diameter. The only difference is stranded is more flexible than solid. Using solid wire in an application where the wire must be repetitively un-wound and then re-wound on a spool is an invitation for disaster because copper wire will "work harden" and eventually break under those conditions.

    It appears to me that your statement that the "wire diameter ... must be 1mm minimum" is a pure rectal extraction, unless you simply translated the original 18 AWG that the OP specified into its metric equivalent, which is 1.02mm. How clever of you to mix metric and Imperial measurements that mean the same thing.


    It does look good, as others before you have pointed out, if it is driving 75Ω cable properly matched on each end.

    That means a 75Ω source impedance feeding a 75Ω transmission line terminated at the far end with a 75Ω load. Even if you get all the impedances matched, there is still the question of signal attenuation as a function of frequency and cable length. We have no idea what that is since the OP has not provided any relevant specifications for the cable he has.

    If the OP was capable of successfully building such a do-it-yourself project based on the AD811, there would still be the problem that the customer cable is probably not anywhere close to 75Ω impedance, probably was not even manufactured as "impedance controlled" cable, and given how the cable loses the color sync information (and probably the chroma component of the composite video too), probably has too much attenuation and not enough bandwidth to propagate color-encoded NTSC video the requisite distance. Wire gauge has very little to nothing to do with that: it's all about geometry and the materials of construction determining cable impedance and signal attenuation.

    But, at less than five bux each plus shipping and handling, and being experienced in this sort of thing, I would buy a pair of AD811 amplifiers and give it a go. Nothing ventured, nothing gained. And the customer already has plenty of cable to test it out on. Experimentation is what this forum is all about. The LM747 experiment has simply demonstrated, as Thomas Edison was fond of saying, just one way that doesn't work. Why not try again with a different amplifier to see if there is possibly another way that does work? As we know now, Tom Edison was guaranteed to eventually succeed, because filamentary tungsten light bulbs can now be found everywhere that they have not been replaced with "mo betta" technology... LEDs for example. My crystal ball is a little hazy on this one though.[/QUOTE]
     
  12. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    Just a note of criticism, I find the format very non-intuitive as far as reading and replying to comments in a thread. However, having this communication ability is an absolute helpful device. I just typed 200 words and lost most of them and scattered the rest. The point was that it may be a week until new parts arrive and I will keep accurate notes and post them here in this thread or a new one if this one happens to expire. BTW The AD811s I ordered were $10.65 each. I went with the 8pin dip for ease in bread boarding. Obsolete part I believe. Digikey had less than 400 left. Although thousands available in surface mount packages.
     
    hevans1944 likes this.
  13. hevans1944

    hevans1944 Hop - AC8NS

    4,195
    1,985
    Jun 21, 2012
    Not so much worried about me or you being right, more worried about the people seeking help here, that they get it right. You know from your decades of electronics experience that there is always more than one way to solve a problem. Sometimes "good" solutions become the "gold standard" and are widely copied. Sometimes not-so-good, but workable, solutions pop up and enjoy a brief period of propagation either by word of mouth or short articles published here and there such as this forum.

    Hopefully, the bad "solutions" will just die and disappear, but I have no faith that will happen. Just based on the few years I have participated on this forum, I see the same bad ideas repeated over and over and eventually they appear here as a "I'm a newbie and don't know sh!t about electronics, but please explain me why my circuit won't work. This is my first time trying to build anything, so please be gentle." posted with no schematic, no pictures, no nothing and not a word about what they are trying to DO!. They often post with bad English. Misspelling I can tolerate, but bad grammar and poor sentence construction is a sure sign the poster has poor communication skills. Why should I bother?

    Yeah, it sucks like a Hoover. I got the impression early on that the OP's "customer" needs to un-reel the camera cable, like maybe to take videos of a high-school ball game, and then reel the camera cable back up to take home. That might help to explain the zero budget approach and the wishful thinking about LM747 op-amps.
     
  14. hevans1944

    hevans1944 Hop - AC8NS

    4,195
    1,985
    Jun 21, 2012
    I agree the online editor here can be a little hard to cope with. It is especially intolerant of misplaced quote and un-quote bracket pairs, which can occur if you just clip and paste text. I usually just hold down the shift key and move the cursor over the text I want to quote, which will highlight that text. When I release the shift key the text will remain highlighted and a little "reply" bubble will appear to the right and below it. Click on the bubble and the highlighted text will be copied as a reply, complete with quote and un-quote brackets, to wherever your cursor last was in your reply window. Sometimes the bubble doesn't appear, but if you start at the end of the text you want to copy, hold down the shift key, then move the cursor towards the beginning before releasing the shift key, the reply bubble usually appears.

    That sometimes happens and I don't understand why. A workaround is to compose your comments in a text editor and copy them to the reply window. I use Microsoft WordPad with Times New Roman 12 pt type font. It seems to pair well with the default font and size used here. If I am feeling especially wordy, I will use the Microsoft Office 2010 version of Word, but that takes longer to load and has more features than I will ever use in a forum. Microsoft bloatware. I also use the Microsoft Symbol Library. Most symbols embedded in Word or WordPad text import just fine into this forum's online editor. Those that do not import seem to just be ignored, so I need to check my imported text to make sure. You should always click on the "More Options..." button before posting. Then click on the "Preview" button to see how your text will appear after it is posted. This is also a good place to check out any links you have embedded to make sure they actually work. Beware of copying links from other posts, especially older posts, because the links may no longer be valid. The new website owners kept the subscript and superscript features I had previously asked Ian to install (which he did), but they are now hidden between the "Insert" symbol and the "Drafts" symbol on the tool bar.

    <sigh> It seems the entire world of electronics as we know it is moving to surface mount packages. I looked at some of the "professional" packages available for doing SMD work and re-work and decided they were too expensive for my current hobbyist activities. However, I did discover that I can "drag solder" some SMD into place using a very fine Weller soldering iron tip and some solder paste or solder flux. Takes a little finesse, a steady hand, and (for me) a pair of stereo head-worn magnifying glasses. I would not want to have to solder up anything complicated, but I am considering modifying a toaster oven to perform mass re-flow soldering of components placed onto solder paste applied through a mask to a circuit board. Sounds complicated, especially the part where the mask is silk-screened onto the circuit board. I know nothing about silk screen printing. Offshore pick-and-place services begin to look attractive now, but where did my hobby go?

    The AD811s may not work with your cable for the reasons I and others have described earlier. I found no evidence that this part is obsolete or is at the end of production, but you only need two of them anyway. Please download, read and understand this datasheet on the Analog Devices AD811 before beginning construction. Good luck!
     
    HANKMARS likes this.
  15. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    Personally, I consider the hour spent building an amp circuit with the LM747 to be time well spent. I can inject my signal and load it down and document its failings, just for fun, let's say. Back in the day, everything was overbuilt. Not so much today. The specs on components manufactured by reputable companies are guaranteed. "Under these conditions, this device will deliver this result." Switches don't seem to follow suit on this but that is a debate for another day. Succeed or not with the LM747, I am sure that it will be interesting. Remember, Thomas Alva was dead certain that low voltage DC was the way to transmit power because of the danger of electrocution by high voltage AC. Opps.
     
  16. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    I referenced only digikey company and only the 8 pin DIP package concerning the obsolete part category.
     
  17. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    I think
    lt appears that with most opamps, as the freq approaches its limit, gain capability becomes quite limited.
     
  18. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    I finally found specs on color camera. See attached file. CAMERA SPECS 1.1.JPG
     
    73's de Edd likes this.
  19. HANKMARS

    HANKMARS

    26
    2
    Jul 28, 2019
    OK. My parts are in and I will leave to go retrieve them in a short time. However, I do not wish to leave my faithful brain trust in an unwanted lackadaisical state that would leave them feeling uneasy. Therefore, I will post another unrelated quandary here for contemplation. You may want to recommend a more suitable subject title for me also. Maybe off-topic category, but it actually is an electronics related question. I desire to know the potential energy of a metric ton, suspended 3 meters above Earth surface. The result will be expressed in kilowatt hours. I was given a formula which I have not verified but does seem fitting at first glance. Mass(1 metric ton expressed in kilograms [2000]) times gravitational constant of Earth (32 feet per second per second [that's around 9.09 meter, you will have to convert]). times distance in meters (3). This will give you joules of energy. Divide joules by conversion factor, 3,600,000, and the resultant will be kilowatt hours. Does this seem correct?
     
  20. hevans1944

    hevans1944 Hop - AC8NS

    4,195
    1,985
    Jun 21, 2012
    Yes. Or you can go to this website and just plug and chug, remembering to divide joules by 3,600,000 to convert the answer (in joules) to kilowatt-hours. Answer is 58,800 J /3,600,000 J/kWh or about 0.0163 kWh.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Similar Threads
There are no similar threads yet.
Loading...
Electronics Point Logo
Continue to site
Quote of the day

-