Mostly a bump for you
But, any chance you have LED bulbs in there?

Looks like an LED ghosting problem since at some point in the video all lights turn on and off. All those lamps are in the same column along with the Right Road sign on the playfield.

If the Right Road sign on the playfield is working correctly it could be a problem with the connector on the light board with cold solder joints. Beyond that, run lamp matrix test and test each lamp. If they come on as they should, it's probably an LED issue. Sys11 doesn't do well with standard LEDs.

viperrwk

Maybe pop some incandescents in there before testing. Can't LEDs also ghost in test mode?

That's a good idea to confirm the issue. Having said that since all of these lamps are in the same column, the ghosting wouldn't appear on the display panel, it might appear on any LED lamps installed in the playfield in the next column. For example, if the jackpot light 1 is on, the captive ball 2 light might ghost if its an LED. It might not because Jackpot 1 is at the end of the row and in the last column of the matrix - depends upon the LEDs used.

Bottom line though is you're right - if the OP puts incandescents in the panel and the problem goes away he'll know for sure. I know that the Jackpot lights on my Cyclone are as spastic as what the OP posted with LEDs. Problem solved with LED OCD.

viperrwk

Here is the section from Clay's guide regarding two lamps on instead of one. This is what is happening, right?

Two Lamps On Instead of One.
If a lamp diode is shorted on (or installed incorrectly), this can cause two lamps to act as one. This can be seen in the "Single Lamp Test". Each individual lamp in the lamp matrix (as displayed on the screen) should flash. The "start game" button will move the test from one lamp to another. If TWO lamps flash in this test instead of just one, suspect this lamp has a bad or mis-installed lamp diode.

Testing a Lamp Diode.
In order to test a lamp diode, use your DMM set to diode test. Turn the game off and put the black test lead on the banded side of the diode. You should get a reading of .4 to .6 volts. Reverse the leads and put the red lead on the banded side of the diode. You should get a null reading. Any other reading and this lamp's 1N4004 diode should be replaced. You don't have to remove the light bulb or desolder the diode to perform this test. You can also test the diode soldered to the circuit boards that hold the 555 lamps in the same manner.

PM me if you need the entire document, it goes into detail on how to test each lamp at the board level to try and isolate the issue to the board or the play field.

Also from Clay's guide, general common issues with System 11 lamps:

Most Common Problems with Lamps.

Bad bulb. Any light bulb can burn out. Often you can visually see the bulb is burnt, but sometimes you can't. Test the bulb with your DMM, set to continuity. Put your test leads on the bulb. No continuity, and the bulb is bad.

Wire broken away from the socket. This happens quite often and requires re-soldering the wire back to the socket lug.

Diode broken away from the socket. If the lamp diode becomes disconnected from its socket, the lamp will not light.

Corroded or Bad Socket. Games imported back into the US from other countries exhibit this problem more often. Re-seating the bulb in its socket often fixes this problem. On 555 plug-in sockets, bend the contact tabs slightly for better contact.

Blown Fuse. If several GI lights don't work, check the fuse associated with them. If all the CPU controlled lamps don't work, check the fuse by the BIG 30,000 mfd capacitor and bridge bolted to the back of the backbox.

Burned Connector on the power supply or interconnect board. This happens most often with GI (general illumination) lamps. See Burnt GI Connectors for more info.

Lamp matrix power resistors. On the CPU board, lower right corner, there are eight .4 ohm, 3 watt wire wound sand resistors (R113, R116, R119, R122, R125, R128, R131, R134). Sometimes these resistors get hot and de-solder themselves from the CPU board and fall off (or the solder joints need re-soldered).

Bad Column Transistor. The TIP42 transistors that control the lamp matrix columns often fail. If this is the case, all the lamps in a particular column will be brightly locked on.

Two Lamps act as One. If a lamp diode has a shorted on, this can cause two different lamps to act as one.

All good points and given the video and explanation, if it's not a connector issue it is probably one or more of the diodes on the 7D lamp board on the speaker panel. Perhaps one was damaged when removing/replacing the panel. You can get replacements at Rat Shack. Won't know for sure until we know what's happening with the Right Road sign.

viperrwk

Just so I understand...

Have you tried changing the lamp on the Right Road? It never lights up no matter what?
Do the jackpot lights keep lighting for the entire single lamp test as you describe ie Jackpot 1 is on with Captive Ball 2, Bonus 3x, Building 8, etc, Jackpot 2 is on with Captive Ball 3 Bonus 4x, Building 9, etc etc?
Are any other single lamps on even dim other than the tested lamp and the jackpot lamp?
Do you have a multimeter with diode function? Have you tested the diodes on the display panel as jalpert has described above?

viperrwk

Its #58 thru #64 on your lamp matrix.

Earthshaker lamp matrix.jpg

Check those diodes. I think you can test them in circuit, correct? Very easy to do and quick.

Pinwiki has a good troubleshooting section -

http://www.pinwiki.com/wiki/index.php?title=Williams_System_9_-_11#Lamp_problems

I would check transistor Q52 then the matrix column and rows.

I think the flow is:

U52 / #7408 (to) Q51 / 2N6427 (to) Q52 / Tip 42

Start downstream at Q52 and work your way up.

Hello!

I purchased this pin from Kurisu (it was nice to meet you, BTW) and thought I would follow up for anyone else who might be having a similar problem.

It was pretty clear to me that it had something to do with column 8, as all 8 of those bulbs were malfunctioning, mostly staying on much brighter than normal but not always and not consistent. There was a lot of hacked board work including a large green 100 ohm Clarostat VK100N resistor, which is about a half foot long and was attached using a couple screws above the main board with two wires snaked down and soldered where R82 goes. R82 is supposed to be a 27ohm white cement resistor. I found the traces around R82 and Q51 were basically gone. Q51 was shorted as well. I put the correct resistor in and replaced Q51, but it didn't seem to help. I replaced the TIP42 at Q52 and that didn't help. I then replaced the 7408 at U20 and still no good.

At this point I was pretty well stuck. I considered replacing the 6821 PIA at U54 but I didn't have a replacement and I wasn't convinced it was bad. I still thought there was a problem with the hacked up traces around R82 and Q51, although everything toned out OK. Finally, I found Q51 had shorted again! ARGH. Probably the 7408 was good after all.

I didn't have another 2N6427 so I had to use a 2N3904 at Q51. Not ideal, but it worked and suddenly the column started working correctly! Finally. I'll have to order some 2N6427's and replace it with the correct part eventually.

Kevin