(Topic ID: 302620)

AS-2518-35 LED locked on

By JethroP

2 years ago


Topic Heartbeat

Topic Stats

  • 34 posts
  • 6 Pinsiders participating
  • Latest reply 2 years ago by JethroP
  • Topic is favorited by 3 Pinsiders

You

Linked Games

No games have been linked to this topic.

    Topic Gallery

    View topic image gallery

    680x_Reset_Voltage.png
    A65DBCD4-7BCC-4D3B-9C74-DCD0871F6A8F.gif
    6CEC804F-FF25-4BBD-B9DA-5A421A8E24F2 (resized).jpeg
    MPU_Resistance_Readings_in_ValidPowerCircuit.jpg
    IMG_1395 (resized).JPG
    IMG_1363 (resized).JPG
    IMG_1362 (resized).JPG
    #1 2 years ago

    Power the board up and the LED comes on immediately. Have followed the wiki troubleshooting procedures but still unresolved. Pulled all chips except U6, U9, and U11. When the LED didn't turn off I replaced all three of these chips with what I believe to be good chips. The reset appears to be working. Short pins 39-40 on U9 the LED stays on. Q1, Q2 and Q5 test good. Have checked continuity on traces from sockets, especially in the old corroded area and all check good. I have pulsing on U9 pins 3, 36 and 37. What would be the next troubleshooting step?

    IMG_1362 (resized).JPGIMG_1362 (resized).JPGIMG_1363 (resized).JPGIMG_1363 (resized).JPG
    #2 2 years ago

    have you got the correct jumpers for the ROMs you are using?

    also see acid leakage still on the board, this will travel, needs to be completely gone.

    the solder on the rear of board for Q5 looks close to neighbouring traces, i'd be cleaning that up.
    some of the pins on the 6800 U9 look to be quite tarnished/oxidised, i'd be cleaning them up.

    you may have a short/eaten away trace under the U8 socket, personally i'd be replacing both the U7 & U8 sockets with socket strips.

    #3 2 years ago
    Quoted from Rikoshay:

    have you got the correct jumpers for the ROMs you are using?

    Yes

    Quoted from Rikoshay:

    also see acid leakage still on the board, this will travel, needs to be completely gone.

    I had abated the acid previously (looks worse than is it), but washed again with stronger acid this morning. Looks a little cleaner.

    Quoted from Rikoshay:

    the solder on the rear of board for Q5 looks close to neighbouring traces, i'd be cleaning that up.

    Even though Q1, Q2, and Q5 tested good, I removed them, cleaned up the board, and replaced with new transistors. Checked connections for shorts. No problems.

    Quoted from Rikoshay:

    some of the pins on the 6800 U9 look to be quite tarnished/oxidised, i'd be cleaning them up.

    I removed all the chips and cleaned the legs.

    Quoted from Rikoshay:

    you may have a short/eaten away trace under the U8 socket, personally i'd be replacing both the U7 & U8 sockets with socket strips.

    I had pried off the socket plastics and looked at the pins on U8 and U11. They looked good and there were no traces gone or shorted. I went ahead and replaced the socket at U8. All the continuity checks at U8 are correct.

    I also replaced U6, U9 and U11 from another working board. No change. LED is locked on and the reset appears to be working. CR5 tests good. I measure 7.82V across VR1.

    #4 2 years ago

    VR1 is an 8.2V zener.......i'd be replacing that, honestly with all those components in the reset section 40 odd years old, i replace them all and raise R11 the 82ohm up about 20mm.

    have you seen with a logic probe or scope pin 40 go from low to high on the 6800?

    shorting pins 40 to 39 should turn off the LED, if it doesn't the problem looks to be the reset circuit.

    i may have missed something, but how do you know the reset is working?

    #5 2 years ago

    I’d put all the ROM chips in their proper sockets. If still locked then try pulling the PIA and install the other one. I’d also try LEON’s test ROM and dig out the logic probe.

    #6 2 years ago
    Quoted from Rikoshay:

    VR1 is an 8.2V zener.......i'd be replacing that,

    not questioning your advice, I appreciate it, but I'm trying to learn here...I measured 7.8 volts on the zener. Wouldn't that be considered good?

    Quoted from Rikoshay:

    have you seen with a logic probe or scope pin 40 go from low to high on the 6800?

    yes, with a logic probe and DVM

    Quoted from Rikoshay:

    shorting pins 40 to 39 should turn off the LED, if it doesn't the problem looks to be the reset circuit.

    I am confused then by the instructions in the pinwiki. It says if you short 39 and 40 and the LED goes out, then concentrate your efforts on repairing the reset section. In my case, the LED does not go out, so I assume that means my reset is working properly.

    #7 2 years ago

    The "E-720-20" ROM in U6 pictured above will lock the LED on and never flicker with those jumper settings - that ROM is from a -17 MPU board which has different addressing.
    Without changing jumpers you need a 9316 PROM marked with either 720-30 or 720-35 at U6 to get the flicker.

    #8 2 years ago
    Quoted from Quench:

    Without changing jumpers you need a 9316 PROM marked with either 720-30 or 720-35 at U6 to get the flicker.

    Thank you for that information. I installed two different 9316 PROM's marked 720-30-U6 and got the same results. No flicker. Locked on.

    Attaching a pic of the board with the socket pins exposed. I don't see anything unusual. No solder whiskers, open traces, etc.

    IMG_1395 (resized).JPGIMG_1395 (resized).JPG
    #9 2 years ago

    Referring to the above postings, would my reset section appear to be working or not? Should I replace VR1 or not? Anything I can probe or measure to determine the specific problem or defect?

    #10 2 years ago
    Quoted from JethroP:

    Referring to the above postings, would my reset section appear to be working or not?

    If you're getting 5 volts at pin 40 of the CPU then I'd say the reset is probably ok.
    I normally replace VR1 when it reads 7.8V or lower but it's only because boards won't consistently boot properly on my bench power supply otherwise.

    The trace running between pins 1 and 2 of U11 has a screw driver nick above it. Is it cut through?
    U11 pin 40 and pin 29 receptors have brown spots, is that just a reflection?

    I don't see anything obvious, you'll have to check all the trace via holes in the corroded area. And make sure you have proper address and data signal continuity from the CPU to U6. Put your logic probe on the VMA signal at pin 5 of the CPU and see if there's activity after powerup indicting the CPU is trying to do something.
    Infact, plug in the CPU only - no ROMs, RAMs or PIAs and make sure every address line is pulsing at each of the sockets.

    #11 2 years ago

    I went ahead and replaced VR1. Now reads 8.3V across it.

    Quoted from Quench:

    The trace running between pins 1 and 2 of U11 has a screw driver nick above it. Is it cut through?

    No

    Quoted from Quench:

    U11 pin 40 and pin 29 receptors have brown spots, is that just a reflection?

    Not a reflection. Those are brown spots just at the tip, not on the receptor wipe.

    Quoted from Quench:

    Put your logic probe on the VMA signal at pin 5 of the CPU and see if there's activity after powerup indicting the CPU is trying to do something.

    There is no activity.

    Quoted from Quench:

    plug in the CPU only - no ROMs, RAMs or PIAs and make sure every address line is pulsing at each of the sockets.

    There is no activity on the address lines at the CPU.

    #12 2 years ago
    Quoted from JethroP:

    There is no activity.

    Quoted from JethroP:

    There is no activity on the address lines at the CPU.

    Is the CPU stuck in reset? i.e. what's the reset line voltage at pin 40 on the CPU?

    #13 2 years ago
    Quoted from Quench:

    what's the reset line voltage at pin 40 on the CPU?

    3.9V

    #14 2 years ago

    Time to get busy:

    Voltages in blue.
    Resistance readings are in red and black.
    If I remember right, red resistance readings are with black meter lead on ground and the red meter lead at the point of interest. And black resistance readings are with the meter leads swapped around.

    MPU_Resistance_Readings_in_ValidPowerCircuit.jpgMPU_Resistance_Readings_in_ValidPowerCircuit.jpg

    #15 2 years ago

    Here ya go…

    6CEC804F-FF25-4BBD-B9DA-5A421A8E24F2 (resized).jpeg6CEC804F-FF25-4BBD-B9DA-5A421A8E24F2 (resized).jpeg
    #16 2 years ago

    At the junction of Q1, R138 and R140 you have resistance readings of "inf/95k" meanwhile on the other side of R138 you have "inf/3M". These don't add up since R138 is a 47k ohm resistor. Can you recheck the resistances at these two points?

    Also what type of transistors have you got at Q1 and Q5

    What voltage do you measure at the top right leg (that's next to the big R11 resistor) of Q5?

    #17 2 years ago

    Rechecked. At the junction of Q1, R138 and R140 I read inf./74K. On the other side of R138 I read 7.5M/3M (and the longer I hold the probes for these readings they slowly raise). R138 does measure 47K.

    Quoted from Quench:

    Also what type of transistors have you got at Q1 and Q5

    Q1 2N3904
    Q5 2N4403

    Quoted from Quench:

    What voltage do you measure at the top right leg (that's next to the big R11 resistor) of Q5?

    Measure 5.3V

    #18 2 years ago
    Quoted from JethroP:

    Rechecked. At the junction of Q1, R138 and R140 I read inf./74K. On the other side of R138 I read 7.5M/3M (and the longer I hold the probes for these readings they slowly raise). R138 does measure 47K.

    Looks like you have an open circuit there somewhere. If you read "inf/74K" at the Q1, R138 and R140 junction and "7.5M/3M" on the other side of R138, the right side figures don't add up.
    Since R138 is 47k ohms, /74K measured on one side plus or minus 47k should give you 27k or 121k respectively on the other side of the R138.

    Open circuit again. The top right leg of Q5 is the reset line, yet you're measuring 3.9V on that signal at pin 40 of the CPU.

    #19 2 years ago
    Quoted from Quench:

    If you read "inf/74K" at the Q1, R138 and R140 junction and "7.5M/3M" on the other side of R138, the right side figures don't add up.

    Every time I measure those points the numbers are different. And the longer I hold my probes on those points, the readings continue to increase. Is that indicative of something?

    I have been doing continuity checks and haven't found an open circuit yet. More checking planned for today.

    Also, does the resistance reading at the cathode of CR7 indicate a problem?

    #20 2 years ago
    Quoted from JethroP:

    Also, does the resistance reading at the cathode of CR7 indicate a problem?

    I have seen some boards give that type of reading so I think you can ignore that one. You can probably ignore that other changing resistance too since you're getting the right voltage at the top right leg of Q5.

    Concentrate for now on why you have 5.3V on the top right leg of Q5 but only 3.9V at pin 40 of the CPU. Do you measure zero ohms continuity between the two points? From your pictures, on the back of the board from the top right leg of Q5 the trace goes across to U8 pin 19 then across to a through hole via above the battery area (suspect that spot because of corrosion) then on the top side of the board the trace goes up to U11 pin 34, then U10 pin 34 then U9 pin 40 and onto J5.

    #21 2 years ago

    This is me watching this thread….

    A65DBCD4-7BCC-4D3B-9C74-DCD0871F6A8F.gifA65DBCD4-7BCC-4D3B-9C74-DCD0871F6A8F.gif

    Can’t wait to see the fix guys!

    #22 2 years ago
    Quoted from Quench:

    Do you measure zero ohms continuity between the two points?

    Measure .4 ohms from the top right leg of Q5 to U9 pin 40. Not zero but that’s lower than my working board.

    #23 2 years ago
    Quoted from JethroP:

    Measure .4 ohms from the top right leg of Q5 to U9 pin 40. Not zero but that’s lower than my working board.

    Does that now mean you're getting 5.3V on the reset line at pin 40 of the CPU?

    #24 2 years ago

    Yep. 5.4v now. (!$&$()

    #25 2 years ago

    MPU LED still stuck on? Any activity showing on the CPU now?

    #26 2 years ago

    No activity. I measured 5.4v without u9. Installed U9 and measured 3.8v. I replaced u9 with another and get same results: about 3.8v.

    Now here is what I’ve learned: only U9 is installed. When I power on the board I get 5.4v at U9 pin 40, then immediately it starts to decay over about a minute and eventually stops around 3.2v

    #27 2 years ago
    Quoted from JethroP:

    No activity. I measured 5.4v without u9. Installed U9 and measured 3.8v. I replaced u9 with another and get same results: about 3.8v.

    Does that mean the top right leg of Q5 now measures 3.8V too?
    Hook up a jumper wire to test point TP5 on the MPU board which is the 5V supply. Put the logic probe onto pin 5 of the U9 CPU. Carefully hold the other end of the jumper on pin 40 of the U9 CPU. Be careful not to slip against pin 39 which is ground otherwise you'll short the power-supply.
    By doing this you are forcing the board out of reset.

    Do you see any activity on pin 5 of the CPU? Are the address lines now doing anything?

    If yes, install U11 and a good ROM at U6. Does the MPU LED now switch off when you jumper pin 40 of the CPU to 5V?
    Try this process a few times in case the LED doesn't respond the first time.

    #28 2 years ago
    Quoted from Quench:

    Does that mean the top right leg of Q5 now measures 3.8V too?

    Yes. Top right leg of Q5 mirrors U9 pin 40.

    Quoted from Quench:

    Do you see any activity on pin 5 of the CPU? Are the address lines now doing anything?

    Forcing CPU out of reset (5V jumper to pin 40) does not turn off the LED. Probed all pins on U9 with 5V jumper at U9 and only activity I get is on pin 3, 36, and 37.

    #29 2 years ago

    Almost every board that I have run into a locked on LED with known good CPU, PIA and ROMS and rom jumpers (and good ground traces to the PIA's) has been a bad chip at U15 which was a 3459 chip. This chip will test as a 7400 if you pull it and put it in a TTL tester, but you can't use a 7400 as a replacement. Suitable replacement is a 7437.

    I did run into a board today on the job that has a locked on LED but booted fine in the game. replaced the LED and Q2 and now it shows the boot flashes.

    As far as the reset section goes, if the only thing that is wrong with the game is the reset section then jumpering pins 39 to 40 will allow the board to boot. If jumpering those together does NOT allow the game to boot then there are other problems to fix first. So many people get hung up on this on these Bally boards and the Game Plan boards. The reset bypass allows you to immediately determine whether the REST OF the board is OK. When the game doesn't boot then fix WHATEVER ELSE is broken, then worry about the reset section later.

    #30 2 years ago

    the 6800's you are using in U9, can you confirm they good by them working in another mpu?

    it may be nothing, but in your first post, on the rear of the board where pin 40 and 39 of U9 the 6800 join to the top header J5, it looks like a possible short?

    #31 2 years ago
    Quoted from Quench:

    Do you see any activity on pin 5 of the CPU? Are the address lines now doing anything?

    I replaced U15. I now have activity on all the address lines. There is no activity on pin 5. I installed U6 and U11. Shorting 39-40 does not turn off the LED.

    #32 2 years ago

    Other common issues I run into are bad traces running from U11 PIA to the U8 5101 RAM back up to the eproms. The amount of corrosion on that board almost guarantees that the 5101 socket should be replaced with machine pins so that you can solder the traces on the front and back of the board.

    One other super common issue (which appears OK here) is the ground trace running to pin 1 of the two PIA from the bottom ground plain. I found a spot on the back of the board where you can connect the eprom ground lines to the PIA ground lines from (looking at the front of the board) the left side of C11/C81 to the top of C8/Pin 12 of U3. This reinforces/replaces the ground that gets lost from corrosion. Lots of boards booting again just from adding this.

    #33 2 years ago
    Quoted from kbliznick:

    As far as the reset section goes, if the only thing that is wrong with the game is the reset section then jumpering pins 39 to 40 will allow the board to boot. If jumpering those together does NOT allow the game to boot then there are other problems to fix first.

    I'd take the other approach with this board.
    The reset line spec for the 6802/6808 CPU says it needs to be above 4.0V. The 6800 datasheet doesn't specify it but you can be sure it's the same.

    OP is measuring between 3.1 - 3.9V on the reset line. Shorting pin 39 to pin 40 isn't going to reliably bring the board out of reset - I would resolve it first so I know it's always ready to properly kick start.

    680x_Reset_Voltage.png680x_Reset_Voltage.png

    Quoted from kbliznick:

    One other super common issue (which appears OK here) is the ground trace running to pin 1 of the two PIA from the bottom ground plain.

    Certainly worth checking as it is a common problem.

    Quoted from kbliznick:

    Other common issues I run into are bad traces running from U11 PIA to the U8 5101 RAM back up to the eproms.

    The U8 socket has been replaced so there's the possibility a via through hole at the socket's been taken out. Replacing the new socket again might cause more damage. Diagnose the fault first.

    Quoted from JethroP:

    There is no activity on pin 5.

    Monitor pin 5 the moment after you reset, sometimes activity can stall after a second or three.

    #34 2 years ago

    Board is fixed and working. Went back to check the address lines and found a short between A5 and A6 under U8 socket (my earlier repair). Guess I should have found that a couple days ago

    Thank you all for your help!

    Reply

    Wanna join the discussion? Please sign in to reply to this topic.

    Hey there! Welcome to Pinside!

    Donate to Pinside

    Great to see you're enjoying Pinside! Did you know Pinside is able to run without any 3rd-party banners or ads, thanks to the support from our visitors? Please consider a donation to Pinside and get anext to your username to show for it! Or better yet, subscribe to Pinside+!


    This page was printed from https://pinside.com/pinball/forum/topic/as-2518-35-led-locked-on and we tried optimising it for printing. Some page elements may have been deliberately hidden.

    Scan the QR code on the left to jump to the URL this document was printed from.