Results 1 to 10 of 10

Thread: Neo Geo AES csync low? (RESOLVED/FIXED)

  1. #1
    Great Puma (Level 12) Bratwurst's Avatar
    Join Date
    Jan 2003
    Posts
    4,445
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default Neo Geo AES csync low? (RESOLVED/FIXED)

    I have one of the early iterations of the Neo Geo AES that on bootup only shows a black screen and was previously mute.

    Bit of a long shot but does anyone around here know if pin 126 of the LSPC-A0 chip should be showing 5 ohms of resistance to ground? On replacing the CXA1145 video encoder chip, I can now get sound but still no video, and I measured csync input at 0.4 volts which seems kind of low for what pin 10 of the CXA1145 is expecting.
    Last edited by Bratwurst; 05-17-2019 at 10:11 PM.

  2. #2
    ServBot (Level 11) Niku-Sama's Avatar
    Join Date
    Aug 2005
    Location
    Deadford, OR
    Posts
    3,656
    Thanks Thanks Given 
    2
    Thanks Thanks Received 
    21
    Thanked in
    19 Posts

    Default

    I dont have any experience with a AES but in general it sounds like you have a short some where...
    I get the feeling you already know that though.

  3. #3
    Great Puma (Level 12) Bratwurst's Avatar
    Join Date
    Jan 2003
    Posts
    4,445
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default

    It looks like the LSPC-A0 is faulty, but I have no other system on hand to reference. Worst case scenario is rob the part from an early MVS unit.

    I'd also like to know if the RGB channels before they go into the CXA1145 (pins 2, 3 and 4) are supposed to be 0.7vpp or if they should be stronger. Curiously this replacement chip isn't outputting the RGB on the other side (just like the original, which was confirmed faulty) and the specsheet insinuates it's just a passthrough, so I have more CXA chips on the way.

  4. #4
    Great Puma (Level 12) Bratwurst's Avatar
    Join Date
    Jan 2003
    Posts
    4,445
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default

    I hope this information may be helpful to others in the future which is why I'm posting a followup. I was able to source a battery corroded MV1 board and determined that csync is normally 5 M ohms to ground pin on the LSPC-A0, and should NOT be 5 ohms as it was on the faulty AES. So I performed a transplant:





    The Neo Geo AES now works as it should. So my suspicion is someone plugged or unplugged a video cable into the system while it or the display was on, or there was a power surge- bridging the audio output and csync input of the CXA encoder. Which bled into the LSCP-A0 chip's csync output pin and eventually shorted it to ground.

    Video of the now fixed system in action:

    https://www.youtube.com/watch?v=leMB9AX9v-Q&

  5. #5
    Cherry (Level 1) -^CrožBow^-'s Avatar
    Join Date
    May 2003
    Location
    Ivory Tower, Fantasia
    Posts
    294
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Nice work!

    You got any ideas where I need to start with this one I've got on my bench of ToDo stuff? It is an older one that used the +5 3amp PSU with it. Client didn't think they one they had was working and tried to power it up using an SMS power supply...so... it had a full +9 - +13+ unregulated voltage applied to it near as I can tell. Nothing seems burnt out visually on the board itself, but the original PSU does smell cooked and isn't providing any power so that is at least confirmed to be dead. I was able to put a +5 1amp adapter to it and now all I get with it is the following in the brief video:

    https://drive.google.com/open?id=1YD...MLql9OoO2yPl0g

  6. #6
    Great Puma (Level 12) Bratwurst's Avatar
    Join Date
    Jan 2003
    Posts
    4,445
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default

    Check to see what's getting excessively hot while it's powered and running, fingertip touch to the tops of chips if you don't have an IR thermometer, be careful not to short anything.

    Also check to see if everything is getting five volts that should be getting it, the bios chip pin 1, the DIP package 68000 CPU pins 14 & 49, etc.

    Somes guesses are: a circuit trace has burned open from overvoltage, one of the processors (68000) could be damaged without tripping the watchdog reset/click of death, or the power rail has something wrong with it, check Q1 transistor, Q2, Q3, Q4, etc.

  7. #7
    Cherry (Level 1) -^CrožBow^-'s Avatar
    Join Date
    May 2003
    Location
    Ivory Tower, Fantasia
    Posts
    294
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    Now that other work requests has lightened up a bit, I can take another look at this soon. I don't recall any chips getting hot to the touch at all when I had it powered up before. I can certainly check for voltage across the ICs. (Just have to look up their pinouts to know which pin or pins on them contain Vcc). The original PSU wall wart smells cooked but I couldn't see anything obviously wrong with it visually. But you plug it in and you get nothing on the output from it. Fuse inside was still good though...

    I really didn't want to take this AES on, but the client insisted I take a look a look at it. So yeah I'm green around the AES and Neo hardware in general.

  8. #8
    Great Puma (Level 12) Bratwurst's Avatar
    Join Date
    Jan 2003
    Posts
    4,445
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default

    I would consider the original power supply a write off and disregard it for the moment. Unless the transformer has developed a fault it's probably the power filtering capacitor that's given up the ghost in there, which could be an easy and cheap fix attempt to try later. Use another known and good 5v supply in the meantime on the system for troubleshooting.

    My guess that the overvolt has taken out something in the system's power rail is optimistic, maybe the rest of the system's getting only 3 or 4 volts for example.

    Worst case scenario is that multiple ICs were taken out by feeding it more than 5 volts, the early AES had next to no protection against that kind of thing. I still have that MV1 board which could be a potential candidate for chip transplants so long as the AES is using the LSPC-A0/PRO-B0/PRO-C0 etc. chipsets.

    The fact that you're seeing anything is indicative that the video out and sprite generation probably means those are working ok.

  9. #9
    Cherry (Level 1) -^CrožBow^-'s Avatar
    Join Date
    May 2003
    Location
    Ivory Tower, Fantasia
    Posts
    294
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    0
    Thanked in
    0 Posts

    Default

    I was advised by an arcade board repair guy that repairs a LOT of Neo-Geo arcade boards that the results I'm getting most likely indicate the vrams are trashed for starters followed by the CPU since it would seem the bios isn't even being accessed yet. The client is hoping I can get this working and then they are are wanting me to put a Uni-bios into it somewhere down the line. I appreciate the info and assistance with this Bratwurst!

  10. #10
    Great Puma (Level 12) Bratwurst's Avatar
    Join Date
    Jan 2003
    Posts
    4,445
    Thanks Thanks Given 
    0
    Thanks Thanks Received 
    10
    Thanked in
    8 Posts

    Default

    Please feel free to post any progress you make on the AES in this thread because I'm always keen to learn about the symptoms to associate with the illness.

    I would look hard at that CPU (the 68000) before focusing on the VRAM. You can have bad VRAM and still hear the game playing, background music, etc.
    Last edited by Bratwurst; 05-29-2019 at 02:36 PM.

Posting Permissions

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