My Nishijin just stopped working

[atlanticom]

Not to hijack the thread, but I have a quick question on Obake since there are several owners commenting. I purchased one last year and discovered it does not pay out any balls during game play. Am I correct in my assumption that this model SHOULD be paying balls when a ball passes through the start gate? (I've determined that balls are not stuck.)

[atlanticom]

As a follow up, I noticed the youtube video of Obake Land gameplay referenced in the beginning of this thread had the player continually adding balls by hand, with no balls paying out even when balls entered the start gate! Hmmm... maybe I'm trying to fix a problem that doesn't exist? Would love to get a definitive answer on this from other Obake Land owners. thanks!

[Moparformances]

As a follow up, I noticed the youtube video of Obake Land gameplay referenced in the beginning of this thread had the player continually adding balls by hand, with no balls paying out even when balls entered the start gate! Hmmm... maybe I'm trying to fix a problem that doesn't exist? Would love to get a definitive answer on this from other Obake Land owners. thanks!

had an o-bake a few years back.. as i recall it should play and pay as any normal pachinko.. no special rules or lack or pay..

just to be sure the back top hopper is full and the balls are making there way to the payout motors.. is it a cr model?? if so do you have a daverob dongle you can test the ball loan system??

[atlanticom]

Thank you for the quick reply. Yes, the top hopper is full and balls are at the payout motors (solenoids in this case.) It is a CR model, but I do not have a daverob dongle. Would love to try that though. Looks like dongles are unobtainable at the moment, but will look into it when Pinball Dave's store is taking orders again.

[MrGneiss]

Can you post pics of the back of your machine?? Someone might be able to spot something..

[atlanticom]

Can you post pics of the back of your machine?? Someone might be able to spot something..

I have it in pieces at the moment since I've been troubleshooting to the circuit-level. However, I will gladly send pics when I get it back together. Happy Monday everyone!

[Drunkenclam]

Clam's basic test procedure.
Ball in start pocket do the reels spin.
No. Duff pocket sensors
Yes. Sensor good.

Does it payout
No.

Is the tray full.

Yes.
Is the tray/ball feed sensor ok (check both tracks as most machines have two) easiest way is to empty the tray and see if you get an error light or indicator.

Then check the bottom tray full sensor
This is harder as it varys per frame. Might have to artificially pus a lever, break a light beam. Or something to simulate this.

Is the payout motor seated correctly. Is it jammed. Do the balls make.it all the way to the motor. Etc.

[atlanticom]

Just a quick update.

To be thorough I progressed through Clam's basic test procedure:
Ball in start pocket starts reels spinning? - yes - success.
Does it pay out? - no - failure.
Is the tray full? - yes - check.
Is the tray/ball feed sensor ok - yes (they are optical, work in tandem, and are working properly. Tested electrically and functionally with error light condition on empty of either track.)
Check bottom tray full sensor. It is a microswitch that failed open (which is the condition for 'not' full). So switch is currently faulty, but in a good way (I can simulate tray full if I short the pins.)
Is the payout motor seated correctly/jammed/ball make it all the way, etc. - yes. It's actually two solenoids with ball sensors. I really suspected these were part of the problem, but while mapping out all the wiring I discovered the ball release switch (I was wondering were that was It is an optical switch that when triggered informs the payout circuit board to release balls until empty. Worked perfectly! (It even proved the third solenoid (to redirect balls away from payout channel) worked.

One other item I found was the microswitch on the front glass door (to sense if it is open or not) was faulty and stuck in the condition of 'door open.' Unfortunately forcing the circuit to 'door closed' condition did not resolve the payout problem. Does anyone know what typical normal behavior is if the front glass door is 'open' during play?

So.... I know the payout board CAN control the payout solenoids when it wants to, I'm now out to discover why it doesn't want to! Still tinkering at it as I have the time.

Have a great day!

[Peteybob]

I'm a bit late to this conversation but have a couple thoughts as I have this frame with another cell.

Did you have a number or letter in the rear lower right display? "0" seems to indicate normal operation. Have you tried the reset button to the left of the same display? It is tiny and on the side of the colored box. Hold it in for five to ten seconds while powering up.

20210127_200339.jpg20210127_200733.jpg

As a quick check my payouts still work with the glass door open and dropping balls in various pockets. I don't know if my door sensor is intact or what just yet. Just a quick check hoping it would have failed.

One more thought for what it's worth on the door. On a much newer frame I disconnected the door alarm which in turn disabled the payout system.

[atlanticom]

Thanks for insights. My display does indeed show 0, so it thinks it is 'normal.' I've pressed the reset button before, but never tried holding it in for 5-10 seconds on power up. I'll reassemble components and try that tomorrow.

Regarding disconnecting the door alarm... I've not experienced any 'alarm condition' while I've had the machine door(s) 'open.' Can you suggest a procedure to induce such an alarm on this frame? I would love to test that hypothesis.

Thank again!

[Peteybob]

Well the door alarm issue was with another and much newer frame. I did mess around with this door switch and had no changes when disconnected or not. It might only be functional in a parlor setting.

I did happen to break a wire which caused the balls to stop paying out though. In the picture below look at the blue connector with two wires. This is attached to the payout motor cartridge. Perhaps you can check either side of this for a break causing a similar issue of balls not paying out. With this break the light in the front upper left indicating low balls in the hopper is flashing. A steady "on" light would be low balls or a faulty micro switch or connection.

20210128_171838.jpg

[atlanticom]

Thanks Peteybob. I really appreciate your time and ideas. That set of wires are good (and connect to the solenoid that redirects balls from payout to the rear hopper - as when the ball release is activated.) I can also confirm that (except for the broken wire) my ball path looks identical to yours.

[grinwing]

Using google translate:

0 = Normal F
8 = Stop
A = Accessory board not connected
F = Lower plate jam
H = No tank ball (hopper empty?)

[Peteybob]

Can you post pics of the back of your machine?? Someone might be able to spot something..

I have it in pieces at the moment since I've been troubleshooting to the circuit-level. However, I will gladly send pics when I get it back together. Happy Monday everyone!

Do post some pics when it is back together. (2mb size limit per picture)

[atlanticom]

I'm still tinkering on my Obake Land payout problem as time permits. I've ruled out bad connection, sensor, solenoid, or switch issues at this point. I'm now trying to figure out how the main game cpu board communicates a payout to the CR board. The connection between the two points highlighted with yellow arrows seems to be the communication interface between the two boards. Ignoring my random makeshift breakout wire colors in the picture, the normal wiring harness has five wires. Yellow (pin1) and four black wires. Tracing the circuits on both boards (and exploring with a meter and scope) I suspect the following pinout:

Pin Purpose
1 Ground (yellow wire)
2 125 Hz Clock signal (opto-isolated)
3 Game CPU to CR CPU communication signal (opto-isolated)
4 CR CPU to Game CPU communication signal (opto-isolated)
5 Power

Throughout my diagnostic testing, I've never seen any communication on pins 3 & 4 (my assumed transmit and receive between the two boards.) Testing the data bus on the Game CPU, I also discovered the CPU it is NOT trying to communicate with the CR board when a ball traverses the start hole sensor. I've tried 'plunking' some random signals on pins 3 & 4, but with no result.

Does anyone in the community have a similar enough machine and o-scope to test if & what signals pass on pins 3 & 4 when a payout happens? Maybe @Daverob has some insight since it's CR related?

Also Peteybob, while I don't have a picture of the assembled machine, I do have a pic of my kitty Pricilla, she appears to be a relative of your avatar pic

Happy Monday Everyone.

CRConnectionWAnnotation.jpgIMG_0026.JPG

[atlanticom]

Obake Land low-level payout tinkering continues...

Yesterday I placed my o-scope on pins 3 & 4 (referenced above) and set it to capture on pin 4 falling. Doing so provided a new result!

Using the 125 Hz clock signal as a reference, I was able to see the following pattern when a ball physically dropped through a normal gate and then travelled to the bottom of the frame
Pin4(CR) sent 1,3 then Pin3(CPU) responded 3,1,1,3,1,1

When I dropped a ball through the fever gate
Pin4(CR) sent 1,3 then Pin3(CPU) responded 4,4

I then confirmed that this communication was initiated when a ball passed through the bottom frame ball sensor (pic attached for reference, but rotated 90 degrees by forum) which is directly wired to the CR board. So I interpret this to be an anti-cheat mechanism to stop someone from triggering a gate without releasing the ball.

I also monitored the communication channel on system power on and CR board reset (using the button on its upper left side.) When I reset or power cycle,
Pin4(CR) sent 1,1,1 then Pin3(CPU) responded 1,1,1

If I lift Pin4 (so that it can't send any signal), when I power on the machine, the CR board displays error A (expected complaint.) If I then reconnect Pin4, I again immediately see the 111->111 communication above and the CR board displays 0 (normal.)

So... all this leads me to believe the communication channel between the two components is functional (since messages passed both ways.) However, I've never seen the CPU initiate communication to the CR board - only respond to it.

My initial expectation was that the CPU would tell the CR board to pay out balls, but I'm not seeing that right now. If anyone has a similar style Y&O frame, and the test equipment to sample communication from their working machine, I would LOVE to get some insight into the normal communication dialog between these two boards.

Wishing everyone a happy day!
IMG_0316.JPG

[Drunkenclam]

Hmmm. if you had a daevrob dongle. You could try to see if that makes a payout, as I don't think that needs communication. with the cel game board. Hmmm. someone might thave to try that out first.

[atlanticom]

I just ordered one from daverob

Hmmm. if you had a daevrob dongle. You could try to see if that makes a payout, as I don't think that needs communication. with the cel game board. Hmmm. someone might thave to try that out first.

[atlanticom]

OBAKE LAND PAYOUT PROBLEM FOUND!!!!

Everything on my Nishijin Y&O Obake Land (that mattered) was electrically working as it should. The problem was a mechanical failure. After acquiring a similar machine to use as a reference, I was able to determine the assembly that senses a ball has travelled through a paying gate had a part with an edge worn down/broken. This missing edge would physically stop the winning ball within the proximity sensor until the payout process was complete. Then a solenoid would retract it and allow the winning ball to pass.

In my case, the winning ball would fall immediately through the proximity sensor and trigger normal communications between the CR & CPU boards, but the CR board would not payout because the ball did not remain present during the payout cycle. Based on the number of youtube videos I've seen of this model pachinko that did not appear to payout - I bet this is a very common failure for these machines. I'm going to try drawing a replacement part in freeCAD and have it 3d printed. If I'm successful, I'll post a link to it on thingiverse for anyone else with this problem.

PachinkoProblem.jpg

BTW: I received my rev.3 daverob dongle this past week - love it! Well done daverob.

[atlanticom]

I was successfully able to recreate the broken part in FreeCAD. If anyone needs it, you can find it here https://www.thingiverse.com/thing:4929657

NishijinPayoutPart.JPG