Sea story with no fish

[Drunkenclam]

I turned on my Super Sea Story in the Mediterranean at the weekend and everything fired up but the screen it's completely blank. I can't see any sign of a back light, or if I shine a bright light at it. I can't see any graphics in the darkness.

Done the basic things of removing and reseating ALL the connectors all the way from the screen back to the PSU. I had to get past the security tabs to do it. That board contains all the connectors to the cel lights and motor, and the frame lights and speakers. Also the CPU and fan. Everything works but the screen. I did the trick with the Star Wars and left it on for 2 hours, then did a reset. But no effect. The screen seems to have 4 sets of connectors. One ribbon cable, Two connectors for the back lights, One connector that seems to connect to strips on the back of the screen.

Any thoughts

[heima]

Is the sound there as if the video display was working?
Checking it with a flashlight is a brilliant idea, but nevertheless something is telling me backlight inverter.
I would not think that if the inverter failed, that the driver for the LCD would as well. (Hence the flashlight)
But sometimes the flashlight trick doesn't work. (you just can't get it at the right angle.)
I am thinking, if the fluorescent bulbs burned out, but the inverter still worked, would it still make the high pitched whine? Maybe not.
Lets say that both the inverter is not powering on, and the drivers are not powering the LCD. In this case, the problem is upstream where the two share something. Like a power source, or a signal from the CPU. That is, if the CPU can power off the LCD. So I would start with the power source.
So, I guess the challenge in that aspect is to find the fuse (it might be an SMT fuse) or the circuit that supplies power to both the LCD and the inverter.
A DMM and an oscilloscope are your friends.
Good Luck, you'll get it nailed pretty quick.

[Drunkenclam]

Yep, sound works great as well as the animatronics. Everything works great but no video. No humming from the back. I'll take some pictures later at. Of course with all the multi plugs it's hard to have it set up on a test bench. Now if only I had an oscilloscope and knew what a DMM was.

[daverob]

My first step at tackling a fault like this would be with the backlight. Shining a light at the screen can show you if the display is there and the fault is definitely the backlight, but doesn't rule out a backlight fault even if you can't see any trace of the display. It's also worth looking around the edge of the display (from the front and the back) with it powered up in a dark room to see if there is any trace of light from the backlight.

I assume you don't have any other machines using the same display panel for a quick swap out of the panel and the backlight inverter.

If you still suspect the backlight is not working, then look to see if there is a separate inverter, and measure the voltage going to the low voltage side of the inverter. DO NOT attempt to measure anything on the high voltage side (the side with the thicker wires leading to the back of the LCD panel) unless you really know what you are doing, you usually need several hundred volts to light a fluorescent tube and often the striking voltage is over 1000v. If there's nothing on the low voltage side, then look for a wiring problem or fuses in the power supply.

If the low voltage is there, then you'll either need to figure out a safe way of testing the inverter (possibly on a known good LCD panel of similar dimensions?) or replace it.

If the backlight is working, then you are going to need to check that power is getting to the LCD panel, and the graphics CPU board. Data sheets for the panel should reveal which pins are the power connections, but you are unlikely to find anything for the CPU board, so will have to identify the power pins of the components you can find data for (Display RAM/ROM chips or any generic logic present on the board).

Post some good pics of the various parts, and we'll see if we can figure out what to set the meter to and where to point the probes.

[heima]

DMM = Digital Multi Meter, its a combination voltage, ohms, current, whatever meter. If you lived in the states, I would tell you to go buy one at Harbor Freight Tools for $5. Are they called VOMs in Brittain?
7 Function Multimeter

[Drunkenclam]

DMM = Digital Multi Meter, its a combination voltage, ohms, current, whatever meter. If you lived in the states, I would tell you to go buy one at Harbor Freight Tools for $5. Are they called VOMs in Brittain?
7 Function Multimeter

Oh a multimeter. Yes I have a couple of them

Picture Time. if you want higher res, Go to my photobucket page.
Pachinko Photos by quigmaster | Photobucket


DSC02873_zpsf6395b31.jpg DSC02875_zpse535ed4b.jpg

The main board in the middle with the fan on it, has all the connections for the cel and frame lighting and sound.
Power is fed from the main PSU via the cable on the bottom left of the board. The large white multi plug above the fan feeds the cpu and screen.
Removing the board gets you to the second layer with the fan and main cpu. All fixed to the back of the screen.

DSC02881_zps91a145a8.jpg DSC028832_zpsdaaea659.jpg

Screen and inverter for the back lights. The main board connector. I checked both fuses and have continuity through both.
I disconnected the lead from the invertor and positined it outside the casing, then refitted the board and back into the machine. then turned it on. I was getting 12+ on pin1 with pins 2-4 being ground/low/negative depending how you read it.

DSC02884_zps7904f276.jpg DSC028802_zpsbe8a29cd.jpg

With the screen and board back out. I thought I better check the continuity between the pins on the sockets out to the tubes. Or what I gather would be the transformer to up the voltage to drive the lighing. Assuming it runs on the principle of copper wrapped round an iron core. Would I not get continuity between pins 1 and 2 on each socket? Well the answer is. I don't.

[daverob]

With the screen and board back out. I thought I better check the continuity between the pins on the sockets out to the tubes. Or what I gather would be the transformer to up the voltage to drive the lighing. Assuming it runs on the principle of copper wrapped round an iron core. Would I not get continuity between pins 1 and 2 on each socket? Well the answer is. I don't.

You won't. There will be something in the ground side of the circuit to measure the current being drawn. so the only way to measure across the windings of the transformer is to make the measurement on the pins of the transformer, and not from the output connector.

I can't find any data on that particular inverter so can't tell if the input voltages are correct or not.

I've got some old Panasonic Toughbook laptops that have 12.1" 800x600 screens, and I think the part numbers are similar. Will see if I can find one of them and check the data sheets to see if they'll be any use.

[Drunkenclam]

Well as another test. I plugged the backlight/screen in to the invertor on one of my other machines and the backlight did illuminate. So I guess there is something in the invertor or further down the supply line. It wasn't safe to connect the screen to one machine and the back light to another. But I wonder if its worth getting hold of one of these to test the screen back in the machine Cold Cathode Invertor Even if is only for a short period of a few seconds, it will prove if the graphics are good.

I'm also watching this auction on YJA Super Sea Story as its a stripped out cel with the screen and psu, plus some ball sensors incase I want to build a ball counter in the future

[heima]

It wasn't safe to connect the screen to one machine and the back light to another.

I am confused. Why not? If the only the fluorescent tubes were powered by one machine (an isolated circuit), and everything else by another, what is the problem?

[Drunkenclam]

The machines are screwed to 2 different cabinets on different sides of the room and there is no way I'm going to cut and shut the supply leads on a perfectly good Sankyo to test my Sanyo now if I had the right plug and socket to make an extension. That's perfectly sound. After all, as Daverob said, these inverters can put out 1000v and with the screen out and wiring, the high voltage terminals are exposed. So best to play things safe

[heima]

Its a fair cop, but I am crazy enough to run some zip cord across the room. Only need to power on one fluorescent tube anyway.

[Chumsize]

found this... Sharp LQ121S1LG55 12.1" inch LCD Screen Display Panel | eBay

some others on there too...

[Drunkenclam]

Well I refitted the screen with the aftermarket inverter and the screen fails to illuminate anything, its just a pale glow from the back lights so much for the simplest option.

[heima]

Sorry to hear that DC. But at least we now know that it is an upstream problem. That is, that something that feeds both the inverter and the display is at fault.

So, how is the power feeding into the board common to both the inverter and the display?

[Drunkenclam]

All the power links and fuses check out of from the main PSU to the board on the back. The only thing I can see wrong is there is a board led out near the 20 pin ribbon connectorand screen driver. The board tracks are impossible to trace as this seems to have 3 layers on it and all the board connections seem to disappear into the middle layer

So there could be a fault with the main board that everything plugs into, or the board underneath that holds the cooling can and the large processor beneath it.

[heima]

It seems so odd that it would "just stop working". I mean, something could have happened when you last powered it off, or when you powered it on, but with all of the protective circuits, its hard to believe that a transient voltage spike could have fried the pachinko.

One of Murphy's laws states that the most difficult part to check or replace will be the one that has faulted. I guess it is time to check there. But before you do, you did check ALL fuses and connectors, not just the most logical ones, right?

[heima]

A crazy idea hit me. This is very unlikely, but still possible. To make an extremely long story short, is there any conductivity on the surface of the circuit boards? That is, what is the resistance between two insulated areas, 1 cm apart. You will need to check in many areas.

I will explain.

A number of years ago, the electronics manufacturing industry was making a shift from lead-based solder and tree-sap based fluxes to more environmentally friendly substances. Water soluble fluxes were developed. Unlike traditional fluxes, these water-soluble fluxes were conductive, and needed to be washed off. Sometimes, the flux would not get completely washed off, but testing showed no faults with the boards. The water-soluble fluxes were also hydroscopic, in that they absorbed water from the air. After being put into service for several years, the flux residue on the boards had absorbed enough water that the residue became conductive. The residue could not carry a lot of current, but just enough to affect logic circuits.

If you ever had a Mac that powered off for no reason, and then would not power on unless you held the button for a while. That over time, you could hold the button and the Mac would continuously power cycle, this was the cause. The ADB bus was going kattywompus. Swapping keyboards and power supplies indicated the problem was with the motherboard. The solution was to pull the motherboard and wash it with hot distilled water. Did Apple follow this corrective measure? Of course not. They replaced the motherboard, that in a couple of years, or less, exhibited the same problem, because it had the flux residue as well.

Take a guess how I know all of this.

[Drunkenclam]

Well yes. I did unplug and reseat ALL the connectors. From where the 24VAC goes in (1x 15A fuses there) Then both ends of the lead from the power switch to the main psu, then all the plugs from the psu to the payout board, cr board and the cel board. Then all the plugs from the cel board that connect it to the frame and cel lighting and speakers. even all the connections from the main board to the motor aniamtronic board. Also broke open the security seals to access the screen and board behind it, then removed and reseated all the plugs there.

Apart from the 2 glass fuses (the 15A one and the 1A one for the dongle) all other fuses are soldered to the main board and all the ones I can find ( there may be some hidden) I have done a continuity check on. and all test ok. There is no sign of any staining on either side of any board only some dust around the CPU where the cooling fan is. The boards are really clean

Out of interest. I did have a fault with the dongle on this one and have returned it to Daverob, But I was able to play it fine for 2 weeks after the dongle was replaced.

[Drunkenclam]

Changed the cel, and machine all working again, I'll keep the old cel for any useful spare, Or unless I can find a stripped out cel in Japan (Just the boards) I'll rebuild it

See here for the goodies box

[Drunkenclam]

Package from Japan today containing a screen and circuit board. Brought for 2000 yen.
Put the original cel into a frame I bought off YJA for 1 yen

IMG_2620.jpg

Did a straight swap, but didn't fix the problem, although the back lights came on and an LED did light up on the board on the back of the screen, But no sound or animatronics.
Swapped the clip on board back to the original, Now had sound and animatronics, Back light and LED on but no picture.

IMG_2618.jpg IMG_2619.jpg

Opened the case and swapped the screen only back for my original one. Put back togther and with a memory clear it all lit up like a charm

IMG_2621.jpg

Yah Fish