Announcement

Collapse
No announcement yet.

HSync & VSync -> CSync Conversion

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • HSync & VSync -> CSync Conversion

    Hi Everyone,

    I am working on adding a CarPC to my 2003 4Runner, and I want to use my OEM Navigation Screen.

    Many people have gone the route of converting the image to a NTSC/"TV" image, and then feeding it to these kinds of screens.

    I have decided (in the interest of a better-quality picture) to use the RGB signal directly. My navigation system uses a CSYNC signal (Combined Sync). This CSYNC signal is non-standard, and seems to have it's logical 1 at +1V (as opposed to +5V). So, if you're familiar with video signals, the CSYNC signal is +1V for most of its cycle, and then drops to 0V (Ground) for about 10% of its cycle.

    Obviously this signal is nonstandard, as TTL operates with logical 1 at +5V.

    But, a lot of people reference Tomi Engdahl's circuits, so I decided to go with his circuits.

    His RGBS Conversion Page: http://www.hut.fi/Misc/Electronics/c.../vga2rgbs.html

    To me, it sounds like I need to use his "Analogue composite sync signal converter" circuit. He describes his circuit as having 1 Vpp and 75 Ohm impedance. I have all the required parts for it...but what I just don't understand is no matter what, the CSync output is always 0V. Makes no sense to me. I've hooked it up to my computer, and tried manually changing the inputs to the circuit. No change, always 0V.

    Am I doing something wrong? I'm under the impression that this circuit *should* oscillate between 1V and 0V (thus 1 Vpp), but I get no readings whatsoever on the multimeter.

    Anyway, let's put that circuit aside for the moment. It's obvious that I need something that oscillates between +1V and 0V with 75 Ohm impedance (I actually measured the impedance and this is correct).

    Is there another way I can go from +5V to +1V? I know I can use resistor methods, but this doesn't seem like the correct solution. Although, maybe it is! I really don't know.

    Here are some of the voltage & current readings I took:

    Navigation ECU:
    1.852V Open-Circuit Voltage
    24.11mA Short-Circuit Current
    (About 75 Ohm Impedance)

    OEM Navigation Screen:
    about 1.54V Open-Circuit Voltage
    0.12mA Short-Circuit Current

    When the two are connected:
    0.977V Average Voltage
    12.28mA Current When Connected

    KEEP IN MIND THAT I TOOK THOSE READINGS WITH A MULTIMETER, AND THUS THEY ARE AN AVERAGE

    Anyway, my basic question is: What do I need to do to get the correct SYNC output to my screen? Does anyone have any ideas or experience in this area? Can anyone look at the circuit and tell me what could be going on with the CSync output that I am getting with +0V?

    I know that the XOR part of the circuit does work correctly; I am getting a signal out of the XOR chip, but the resistor/transistor part kills it.

    Lastly, does anyone know what the purpose of the capacitor/resistor signal going into the XOR gate is? These gates seem redundant to me, but I am probably wrong. I know that a typical purpose of inserting a capacitor like that could be to prevent instantaneous changes in voltage, but I am no expert

    Any help would be greatly appreciated!

  • #2
    Did you manage to get the Sync frequencies correct already? (15 Khz issue)
    Needstorage
    --------------------------

    Who the hell is general failure?!?
    And what the f@#k is he doing reading my drive C:\?

    Comment


    • #3
      Well, I'm not sure yet. I think I can do it, but the thing is, I don't want to connect my computer to the screen until I am 100% sure that I have the correct signal coming out of the sync converter.

      If the sync converter is outputting around 5V, and the screen is looking for around 1V, that could certainly do some damage. So, getting the sync converter to work is top priority.

      Comment


      • #4
        Originally posted by Mr Castulucci
        Lastly, does anyone know what the purpose of the capacitor/resistor signal going into the XOR gate is? These gates seem redundant to me, but I am probably wrong. I know that a typical purpose of inserting a capacitor like that could be to prevent instantaneous changes in voltage, but I am no expert
        I'm not an expert either but here goes my guess. Assume the capacitors are initially uncharged. If HSYNC goes high, the XOR gate will see a HIGH on pin 1, and LOW on Pin 2 and output a HIGH. After the capacitor charges, both inputs will be HIGH and the output will go LOW. When HSYNC goes low, the XOR gate will see a LOW on pin 1 and and a HIGH on pin 2 ( because the capacitor is already charged) and ouput a HIGH. Once the capacitor discharges, there will be LOW on both inputs and output will go LOW again.

        Thus, assuming I understand it correctly, the output of the first set of XOR gates is such that they generate a HIGH to LOW pulse for each change of the input (HSYNC or VSYNC) signal.

        HSYNC => HHHHLLLLLHHHHLLLL => XOR => HLLLHLLLHLLLHLLL

        The second level XOR gate combines the signals and outputs a HIGH only when they are not equal. The third level XOR (with one input tied to ground) just passes the input and thus acts as a buffer.

        I assume the transistor network just adjusts the output voltage back down to 1 Vpp and provides the appropriate output current levels. If you're getting clean logic levels out of the last XOR gate, then I'd guess that your problem resides in the transistor resistor net. Double check your connections and/or try replacing one or both transistors. It's possible they are defective.
        2004 4runner

        Comment


        • #5
          Rando,

          I was thinking the same thing about the transistors... if I get some time tonight, I am going to work on it. Luckily I purchased some extra transistors

          Can anyone confirm that the transistor network is in fact supposed to bring the output from 5V down to 1V?

          Comment


          • #6
            I e-mailed Tomi Engdahl asking him for some guidance. I figure - hey, he created the circuit, so he's the best person to ask

            Comment


            • #7
              Originally posted by Mr Castulucci
              Well, I'm not sure yet. I think I can do it
              So what is your idea then? I am pretty anxious to hear it. I was thinking about the possible T-flipflop-alike workaround, but I also don't think it will make it work.

              Please, if you have an idea, let me know it, maybe I can have a go on it to test it. (I have a Basic Stamp laying around here that I can use to program some of these shift/deviding functionality)

              Also, I didn't even check the voltage that has to be input to the OEM screen in my Galaxy. I just 'went ahead' and took the chance. My OEM screen still works
              Needstorage
              --------------------------

              Who the hell is general failure?!?
              And what the f@#k is he doing reading my drive C:\?

              Comment


              • #8
                http://nakoruru.rudora.net/carcomp/serio_lcd.htm

                Check that out. I'm going to use a similar procedure with the program PowerStrip.

                Basically trial & error - it should work eventually

                Comment


                • #9
                  Originally posted by Mr Castulucci
                  [url]with the program PowerStrip.
                  This is the one I tried too... but I didn't manage to do it.
                  Needstorage
                  --------------------------

                  Who the hell is general failure?!?
                  And what the f@#k is he doing reading my drive C:\?

                  Comment


                  • #10
                    Well, I've decided to put the project on hold for now.

                    I spent pretty much the whole day working on it, and I haven't really made any real progress.

                    Firstly, the RGB color signals are not in the correct format. I don't have access to an oscilloscope, so I can't see what the difference between the PC and the OEM Navi ECU's RGB is. I hooked the PC RGB singals to the OEM screen and the screen remained black. The screen is only black when it has no RGB color signals connected, which means that the RGB singals probably aren't adequate. I don't know what to do with them.

                    As far as the CSYNC goes, I decided to give it a whirl. While the CSYNC coming out of my converter does affect what is displayed on the screen, it doesn't seem to affect it in a good way. Let me explain: I leave the OEM Navigation ECU's RGB connections connected and I connect the CSYNC line to the sync converter (which is connected to the PC's VGA). The distorted Navigation Image gets even more distorted. Furthermore, as I adjust the sync rates on the PC, the image on the screen does not change. This tells me that there is something wrong with the sync conversion circuit. The OEM equipment is obviously nonstandard. I don't really know enough about complicated circuits to continue work at this point.

                    I was thinking that it would be nice if I had a video card that had a CSYNC output, as that may solve a lot of my problems. Then again, if the OEM equipment is nonstandard (as it seems to be), then it might not make any difference at all.

                    If anyone has any ideas on this project I'd love to hear them! Thanks for all your help guys!

                    Oh, and I'm not putting the CarPC on hold in general - just the intergration with the OEM equipment. As an alternative, I am considering buying a Xenarc or Lilliput and putting them in the Center Console Storage Cubby (replacing the cubby with the screen). I think that would be a nice way to do it.

                    Comment


                    • #11
                      The center cubby is fairly low and will give you a less than desirable vertical viewing angle to the screen. Also, you'll lose access to the LCD when you use the cup holders. Finally (and I know you've already seen what's behind there), recall that there isn't much space behind that cubby.

                      Maybe there's another approach you can take? What features do you want from your carpc? I assume navigation isn't one of them since you'll be keeping your OEM system. If you're just going to use the PC as an audio/video juke box, perhaps you can get away with a composite video connection to your oem screen? If you skin it right and pick a big enough font, this might be acceptable? Have you looked into tapping the oem touch screen interface?
                      2004 4runner

                      Comment


                      • #12
                        Hey Rando,

                        Sorry it's been so long since I've been on here. I went on vacation and now I am back at school taking summer classes.

                        You're right about the center storage cubby. I guess that wouldn't be the best route to go. I haven't completely given up on this yet though. What I am going to do is find the professor down here (U of IL in the top 3 schools in the USA for computer engineering) that teaches ECE 418 - Image & Video Processing (a class I intend to take, but will not take until spring 2005 or so). Hopefully he/she will be able to help me out or at least point me in the right direction. I know integration with the OEM equipment is possible - I just don't have the right equipment or knowhow to proceed on my own.

                        I'll keep you guys posted. I'll write that prof an e-mail in the next couple days.

                        Comment


                        • #13
                          By the way, I heard back from Tomi and below is what he said regarding the transistor network:

                          >> Is the function of the transistors to just step the voltage down from 5V to 1V?

                          The idea is to step down the voltage and give more current driving capability. Driving 1V to 75 ohm load would take more than 10 mA of current output capacity, something that the logic IC can't reliably provide. Transistors have no problem in giving that..

                          Comment


                          • #14
                            Originally posted by Mr Castulucci
                            By the way, I heard back from Tomi and below is what he said regarding the transistor network:

                            >> Is the function of the transistors to just step the voltage down from 5V to 1V?

                            The idea is to step down the voltage and give more current driving capability. Driving 1V to 75 ohm load would take more than 10 mA of current output capacity, something that the logic IC can't reliably provide. Transistors have no problem in giving that..
                            Makes sense ... but we pretty much knew that part already (almost). More importantly, did you find out what was wrong with your implementation of the circuit (i.e. why wasn't it working)?
                            2004 4runner

                            Comment


                            • #15
                              No, unfortunately... that's what I'm hoping the prof can help me with

                              Oh, by the way - I got my Opus in the mail Rando, I'm planning on making a PC case like yours. I'll have to read up on what you did when I get a chance.

                              Comment

                              Working...
                              X