(Topic ID: 286030)

Ghostbusters Premium: "9 Node Not Found" was the beginning...

By DruTheFu

3 years ago


Topic Heartbeat

Topic Stats

  • 50 posts
  • 11 Pinsiders participating
  • Latest reply 3 years ago by DruTheFu
  • Topic is favorited by 7 Pinsiders

You

Linked Games

Topic Gallery

View topic image gallery

5552C947-4D58-4763-BE11-420C3AF40E5A (resized).jpeg
SpikeNode9_Dx (resized).jpg
75A458F0-A8C9-4859-B046-B6ADA53067BF (resized).jpeg

You're currently viewing posts by Pinsider druthefu.
Click here to go back to viewing the entire thread.

#1 3 years ago

I am the original owner of my Ghostbusters Premium purchased in October 2016.  I turned the game on last night, and the game would not complete the boot process, displaying "Locating Node Boards, 9 Node Not Found".  I restarted the machine three times, receiving the same message.

After searching Pinside for similar reports, I proceeded to unplug and reseat all Ethernet cable connections in the backbox, in the cabinet, and under the playfield.  The same message displayed when attempting to boot when the machine was turned back on.  This time, after displaying that node 9 is not found for about 20 seconds, the game completed its boot process, but the majority of the lights in the back section of the playfield were dark, and I noticed smoke emitting from under the playfield accompanied by a burning smell.  Looking under the playfield, I saw the smoke coming from the left magna-sling magnet.  The metal bracket holding the magnet in place was extremely hot to the touch, and the black tape-like covering surrounding the magnet looked like it had melted some.  I disconnected the molex connecter of the magnet assembly to remove its power source to prevent any further damage since it appears that the magnet may be locked on.

I have also removed the Node 9 board (520-7017-72A) to check for any damage, and nothing looks burnt/blown and I did not see any solder issues. All transistor legs look to be intact with the board.

I have installed the recommended Service Bulletin kits years ago as they were released (#189 "Core Node Stabilization Kit 502-7061-00", #190 "Capacitor Service Kit 502-7060-00").

Is there anything else I should check? Is my only hope to replace node board 9 and the magna-sling magnet assembly?

#2 3 years ago

I was able to remove the left manga-sling assembly once it cooled off. The good news is that the core, bracket, and core screw all seem unaffected by the heat from the coil locking. The bad news is that the magnet coil will need to be replaced, and even worse, it’s a part that looks to be 3x the price of other magnet coils!

75A458F0-A8C9-4859-B046-B6ADA53067BF (resized).jpeg75A458F0-A8C9-4859-B046-B6ADA53067BF (resized).jpeg
#4 3 years ago

Bump for any other ideas or confirmations of my suspicions...

#7 3 years ago

So with my original Node 9 board error message I received, and the the Left Sling Magnet burning up, it was pretty easy to confirm in the user manual that this coil was attached to Node Board 9.

I've removed Node 9 board and pulled out my Fluke multimeter. Not sure if I performed the readings correctly, but I placed the meter into Diode mode, with the black lead touching the top of the transistor, and the red lead on each of the legs. From reading on this site and some outside searching, readings should be in the .4 - .6 range. Looking at my readings, not sure if I'm doing this correctly.

Can any of you more knowledgeable board/transistor testers teach me to fish?

SpikeNode9_Dx (resized).jpgSpikeNode9_Dx (resized).jpg

#8 3 years ago
Quoted from Lermods:

It’s a spike game so tough to fix/diagnose. A coil or magnet locked on would typically be a transistor issue caused by a short. If you look in the manual, see if you can identify where the transistor is that protects/drives that magnet. Then get out your meter and start testing the transistor values against the other nearby ones.

Would this short be on the board itself, or somewhere along the wiring runs that connect to this board, specifically the Orange/Blue or Gray/Brown wire runs that connect to the left magna-sling?

#10 3 years ago

Making more sense to me after staring at the different pages in the GB user manual for a while.

Node 9 board, Transistor Q7 is associated with the left magna-sling.

The GI lights that I instantly noticed out on the playfield are using Q15-17.

The readings in my pic look to correlate to what I'm seeing, and with Q8's readings, it looks like I caught the right magna-sling locking on in time before it burned up that coil too.

Turning the board over, locating the labeled solders and following the traces, and inspecting the board even closer, I see that there looks to be some irregularity to the areas of the above-mentioned transistor solder connections.

It looks like I've located the blown transistors, and it seems like they were blowing in some kind of a lined order.

Now it looks like I need to trace some wires and look for any exposed copper, any bare wires crossing/touching, pinches.... Would a bad bulb be a culprit?

What else should I look for that would lead to transistors blowing on a board? I'd hate to install new parts only for them to suffer the same fate.

Is there a way f guessing/knowing wish transistor was the first to blow and concentrate on those wires?

Thanks for all the tips and help!

#12 3 years ago
Quoted from northvibe:

Was the “tsb” done for the board where the Ethernet looking cable was pulled too tight?

I don't think I've seen that service bulletin. Is there a link to this, or something to verify with my machine?

I've installed
#189 "Core Node Stabilization Kit 502-7061-00" (that classy looking zip-tie in my picture)
#190 "Capacitor Service Kit 502-7060-00"
... (and #187 Ecto Goggles Shield)

#15 3 years ago

I thought the same, and with the play field up and peaking through the captive ball area, i verified the Mylar fix is installed.

#16 3 years ago
Quoted from Lermods:

If your gi lights are out, there have been a few threads recently where the socket was causing a short so I’d start there.
I’m not sure what could have blown the magnet, maybe it just failed internally and blew the transistor. Maybe use a meter and compare readings with the other magnet/wiring to the magnet for clues.
Spike is a tough system to troubleshoot.
Have you done any work on the game prior to this happening?

I’ll poke around the opposing magnet coil wiring, see if any clues.

This has been my first real “gotcha” moment with my GB after 4 years NIB. I’ve never added any mods, so no alligator clips or out of factory soldering jobs.

Just wish I could chalk it up to the board living a good life and move on, waiting for replacement order to install. But the “why” in me wants to keep searching for the cause so I don’t have to have another $300+ Pinball Life order on my hands.

I hate to say that working on my former TZ seemed a little easier to pinpoint than this Spike on-board scheme.

#19 3 years ago

Thanks for your input. It's helped get me closer.

I wish it were as easy as going light by light down the wire strand checking diodes. With this Spike, it looks all wires are directly connected to each socket without a protective diode. Each light socket is screwed into the playfield underneath, so unscrewing all of them lights per strand to remove the bulb to troubleshoot is going to be a process requiring some time and effort. Going one by one per socket, per board connection, until I get to the short.

Time to start my search on troubleshooting Spike.

I've visually inspected all the wire strands connected to the affected 5 blown transistors, all look fine. The soldiers from each wire to socket looks clean. I didn't see and exposed wires touching along the search.

Super exclusive ad from the Pinside Marketplace!
#21 3 years ago

Following up from a previous comment of mine regarding the underside of the Node 9 board...

5552C947-4D58-4763-BE11-420C3AF40E5A (resized).jpeg5552C947-4D58-4763-BE11-420C3AF40E5A (resized).jpeg

From reading the "thoughts on Spike" thread, one would think that the board has the logic to shut down the affected source of the issue/short. (I think I even read that you said similar). It doesn't look like that occurred, and it looks like the board started getting burnt up where I tested failed transistors.

#25 3 years ago
Quoted from Schwaggs:

The GI driver transistors look good to me from your readings. However of you want to order some just in case they are here:
https://www.mouser.com/ProductDetail/ON-Semiconductor-Fairchild/FQP20N06L/?qs=%2Fha2pyFadujhaj0xBReI9xUAnlzFOBeKo69xT23Kvbk%3D

Thank you for the part number. I don't have much experience with board repair, and my soldering skills may do more damage to this board than fixing it.

I have a new board and magnet coil on the way from Pinball Life (yay $315 ). I'm hanging on to this current board and will attempt to repair it at a later time.

My focus now is to understand diagnosing the cause, and finding any short that would have caused this, if this is indeed the cause. The Spike diagnosis that appeared on the screen was very uninformative, reading that it couldn't find Node 9. I wish it would have pinpointed better like "transistor X detected...", or point me in a direction to look.

#26 3 years ago
Quoted from Schwaggs:

Can you CAREFULLY look at D9 (the one with the zip tie through it at the top of your picture) and see if either lead is broken off? These would vibrate until one of the legs would crack, causing an overvoltage to the 6v section (this includes GI).
If that is broken, you should get a replacement for it as well:
https://www.mouser.com/ProductDetail/Vishay-General-Semiconductor/MBR760-E3-45/?qs=%2Fha2pyFaduiIANWfweiZVnqRONx8UmAa7nSQK1vptCZzj%252B8lEkARlQ%3D%3D

The D9 diode that has the zip-tie service bulletin "fix" looks like it is still solid, with each of the two legs soldiered into their respective connection points.

#28 3 years ago
Quoted from Schwaggs:

Look at the leads not at the board but as they enter the black body of the Diode. Push on it side to side to see if one of the legs is broken off.

Gotcha. Thanks!

The leads connecting to the black body of the diode appear intact, and has no wiggle or movement to it while gently trying to do so.

#36 3 years ago
Quoted from dsmoke1986:

I had the same thing happen on my Stranger Things Pro, it was the shaker motor that fried the board.
Do you have a shaker in this game?

I DO have a shaker motor on this game, and it is connected to the Node 8 board. Would that affect Node 9?

#38 3 years ago

Reporting back with some more findings.

Using my multimeter, I did some continuity testing with all of the GI (CN15) connected to this Node 9 board, as well as the drivers connected on CN7 (blown transistors Q7 and Q8). My testing sourced from the molex connector that attaches directly to the board, and the destination was each individual end-point connection. I did this for both the source power (+) wires and the return (-) wires.

CN15
- GI 118 (red/yellow - ) (red/black + ) GOOD
- GI 119 (blue/yellow - ) (blue/black + ) GOOD
- GI 120 (green/yellow - ) (green/black + ) GOOD

CN7 (blown transistors Q7 Left Magna Sling, Q8 Right Magna Sling)
- Left Magna Sling (orange/blue - ) (gray/brown + ) GOOD
- Right Magna Sling (orange/violet - ) (gray/brown + ) GOOD
- Left Pop Bumper (orange/black - ) (gray/brown + ) GOOD

I didn't seem to find any loss in continuity on any of the above (beeeeeep, .3 or .4), so there doesn't look to be any shorts in the wiring. Please correct me if I'm wrong.

Out of curiosity, I tested each of the magna sling magnet coils. The left magna sling that burnt up read 1. The right magna sling that looks to be ok read 3.5. I received those same exact numbers on 5 seperate tests. Left magnet coil is dead to me HAHA.

I mounted the Node 9 board back to the underside of the playfield, plugging in ONLY the 2 Ethernet cables (CN4, CN5) for Spike BUS communication, and CN6 for the main power to the board. Powering the game ON, I still received the message "9 NODE NOT FOUND". I went into the Node Test Menu, and there were no errors reported except for Node 9 ("9 NOT RESPONDING"). I attempted the Node BUS Test to no avail on Node 9. I performed the Node BUS Test on all other node boards with no errors reported, and communication present to all.

Not sure what else I can do at this point until I get the new board delivered and installed. This board isn't being recognized even without any load connected.

Being that the GI wires test good for shorts, and the associated transistors do not seem to be blown, is it safe to believe that the GI is not the cause of the Node 9 board failure? If that holds true, it would save me a lot of time unscrewing each and every GI light socket from under the playfield for visual inspection for issues that other Spike users have recently reported. And it would save me even more time not removing each bulb for additional testing.

Is there a chance that the left magna sling lived a good life, and out of nowhere decided to end it all and lock on, resulting in its transistor pulling the cord, and the board making the ultimate sacrifiice in the end?

I'm thinking that once I receive my new board, that I'll reinstall, starting with the smaller loads, one at a time, building up to adding more connectors one at a time. Thoughts?

Any other ideas for testing and diagnosing?

#39 3 years ago
Quoted from PinMonk:

Very unlikely. Your melted magnet is pretty much the clue. It probably failed and took out the transistor.

This is what I'm thinking too.

#42 3 years ago
Quoted from PinMonk:

What do the LED status lights on the Node 9 line up with on the LED status table here:
https://sternpinball.com/wp-content/uploads/2019/02/520-7017-72D-Core-Node-Hall-3amp.pdf

The red LED is solid on (48V present).
The green LED is solid on (6V present)
The amber LED is off/dark.

The LED Status Table description: "48V present, 6V present, processor not running: check 5V and 3.3V supplies."

Looking in the GB user manual and the Node 9 pinout on page 25, I see the 48V input power on CN6, but I do not see any pinout for 5V/3.3V inputs. I assumed the board received it's power all from the 48V input and performed the electrical step downs within the board architecture.

Does this indicate that the component(s) on the board that perform this function have failed and unable to do so? AKA my board is shot.

And I appreciate everyone's input and ideas. Thanks

#43 3 years ago
Quoted from Schwaggs:

I'd expect a coil that looks as burned as that one to be shorted internally. You do not know if this occurred first and blew the node board transistor or if the node transistor went first, overheated the coil which resulted in the coil shorting.

This is the chicken or the egg question I want to know! Was it the coil that went first or the transistor? Knowing would hopefully help me figure out if I have a bigger issue at hand or it was a simple and inevitable part failure.

1 week later
#46 3 years ago

I wanted to jump back in to add an update and some closure to my issue.

I received a new Spike Node board and magnet coil from my Pinball Life order. I started by installing the new board, only connecting the two ethernet cables for the Spike bus communication (CN4, CN5) and the main 48V power supply (CN6). The game booted without the "9 NODE NOT FOUND" error message, and all of the board's LED status lights were now functioning as anticipated. I then connected all of the CN molex connections with the exception of the CN7 connection powering the magna sling coils. Upon booting up, there was still no report of any errors. I then connected the magna sling coils, and proceeded to connect CN7 to the new board. I powered the machine back up, and still no errors present!

I played a couple of games to test it out, and everything looks to be in good working order now.

Thank you to all that chimed in with ideas and tips to my issue. It's great to have a supportive community when it comes to actual tech troubleshooting and fixing of issues without any of the "noise" that some threads spiral into. I appreciate everyone's help!

2 weeks later
#50 3 years ago
Quoted from Pinhead1982:

Late to this thread.. but was it possible that the lead switch in the slings got stuck closed by the rubber, causing the magnet to be ON and burning up?

I checked the switch and sling rubber, and there didn't appear to be an issue.

Since replacing the magnet coil and Node 9 board, all seems to be good and functioning as expected.

Promoted items from the Pinside Marketplace
$ 16.00
Boards
Lermods
Boards
$ 28.00
Playfield - Other
Pin Monk
Other
From: $ 14.95
Playfield - Toys/Add-ons
ULEKstore
Toys/Add-ons
$ 299.95
Lighting - Led
Pin Stadium Pinball Mods
Led
$ 38.00
Various Novelties
Pinball Photos LLC
Various novelties

You're currently viewing posts by Pinsider druthefu.
Click here to go back to viewing the entire thread.

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/ghostbusters-premium-9-node-not-found-was-the-beginning?tu=druthefu 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.