Sega G-loc deluxe motor power supply

[perjmolsen]

Hi

If someone needs the eprom code for the eprom that is located on the motor driver board.
Then I have made a dump of my eprom

https://www.dropbox.com/s/87oitu23utdpemj/Sega-gloc-motor-eprom-tms27c256.zip?dl=0

It need to be burned on a 27C256 eprom.

Kind regards
Per

[perjmolsen]

Hi

Just a small update to the repair log, there is a typo R24 is a 2.2ohm NOT a 220 ohm

https://www.dropbox.com/scl/fi/jdwr58zhg2tofkmensh2q/Nomenclature.ods?dl=0&rlkey=tm49vx0ax3kz10vesr5eg5ngx

[ArcadeMachinist]

Hello everyone,
I'm sorry for writing in English, but my French is 0.

I'm trying to repair G-Loc motorboard, and little bit stuck, becuase my understanding of analog electronics is not so good.

I have 2 board, one of them now work fine, anotherone has problems with delivering voltage to Q1 gate.
On working board there is +3v after R21b, on the bad one - +0v
Something is sinking +3v to Ground, so Q1 does not open.

Also, on my board R23 and C15 are in series, not parallel as posted here before. D6 is reversed.
My board had Q1 dead (blown), I have replaced it, also replaced L2 transformer with a new one, thanks to Per.

I have changed C15, R23, ZD2, R25, PC1, D4, R22, C16, D6, R28, R27, ZD1, D5 and Q1 with new parts.
Did not help. Something still sinks 3v after R21B to GND.

If I (for testing) remove L2 transformer and R22 resistor - then Q1 gate gets exactly 3 volts, as it should.

What can be wrong?

All circuits after the L2 are tested and work fine.
As a bonus info PC910 optocouplers can be replaced 1-to-1 with Everlight 6N137, they are the exactly same and tested to work fine.

[gc339]

A l'époque où j'ai relevé le schéma du convertisseur BT, j'aurai dû accorder plus d'attention au schéma transmis par pn_jeux, il m'avait semblé à première vue trop dissemblable. J'aurai dû me rendre compte que la circuiterie autour du primaire du transformateur était quasiment la même ce qui m'aurait évité les erreurs qui ont été trouvées.

J'ai donc redessiné ce schéma en corrigeant ces erreurs pour qu'il soit plus propre et plus compréhensible.

@ArcadeMachinist

Are you sure the new components are all OK?
Have you checked them before soldering? Especially if they were bought from Chinese sellers on eBay or AliExpress.

Are you also sure that one of the 15 volt power supplies is not shorted? Have you tried to unsolder diodes D3, D9..D11 one by one to check it.

[f4brice]

J'ai donc redessiné ce schéma en corrigeant ces erreurs pour qu'il soit plus propre et plus compréhensible.
[...]

Je trouve ce schéma fait main magnifiquement superbe ! 

[ArcadeMachinist]

@ArcadeMachinist
Are you sure the new components are all OK?
Have you checked them before soldering? Especially if they were bought from Chinese sellers on eBay or AliExpress.

They are all new, from Mouser, not from China.
I was testing them one-by-one with that transistor/diode/capacitor/resistors tester device.

Are you also sure that one of the 15 volt power supplies is not shorted? Have you tried to unsolder diodes D3, D9..D11 one by one to check it.

I have tested all secondary circuits, by removing L2, applying 18v from lab power supply to L2 outputs 11,10 and 14,13 -  checking that voltage regulators (IC5, IC6) correcty drive down this 18v to 15v.

12v in case of "fan" output, all of them work.
15v in case of 8,7 outputs.

No shorts. L2 is new from Per.

[ArcadeMachinist]

In case anyone want's to sell me a spare board, or repair mine for a good price - let me know please! 

[ArcadeMachinist]

It also looks like not only the enable pin 6 is PWM controlled, but direction pin alternates polarity all the time.

Idle - up/down alternates every 32μs. 32μs up, 32μs down. 64μs full cycle.

When the cab goes up on startup, I'm getting 37.5μs "up", 26.5μs "down", so around 58.5% up.
When it slowly returns back down, it is 30.2μs "up", 33.8μs "down".
There is PWM too, but it is actually insignificent, almost always 100% duty.

So, I guess, the speed is controlled by how much % of the time it spends in up or down direction.





Or I might just be measuring something wrong

[ArcadeMachinist]

I have reproduced the PCB, as a physical 1-to-1 copy.
First used front traces layer from this forum, but it has multiple mistakes, which are now fixed.

Low voltage part works as intended, verified it does trigger MOSFETs in correct order.


High voltage part - with L2 removed - it gives correct 163V** DC voltage at L2 terminals 3 and 4 (main coil).
Once I connect L2 - the board blows 6A fuse, I have set before it (so 8A triac survives each time).



I guess L2 is a part of a "switching power supply" circuit, where "feedback" coil is supposed to help switch Q1 (via Q2) on/off constantly, limiting the input voltage to L2, to the level where it produces desired output.

With something being bad in the "feedback", constant 163VDC is applied to L2, which shortcircuits the full thing.
I have replaced EVERY part in the feedback circuit, with testing new part before soldering.
All the traces and connection verified to match original PCB 1-by-1.
I must be missing something and would appreciate ideas how to test it further.

I have uploaded my KiCAD design here:
https://github.com/ArcadeMachinist/GLoc-Motor-PCB/tree/main/kicad-reproduction



Mistakes in top layer, which was published previously in this forum:
L2-1 trace incorrectly goes to via, while it shoud go to ZD2
VR2-3 trace incorrectly goes to C24, while it should go to C23
PC7-2 missing trace to R59

All fixed in the github project.

** See below

[ArcadeMachinist]

Turned out I was getting 163V, becuase TR1 never latched and R1/R2 are removed.
In my good board R1/R2 were dead, I have decided to put new ones - and they are burned immediatly. wtf.
However good board works without R1/R2.

I was able to get 2.7V on Q1 gate and Q1 worked, launching L2 and I was getting 7.1V between pins 5 and 3 or L2 coil. Same as in working board.
But at the time of this experiment I had TR1 removed and after some seconds my Q1 started smoking.
The tester shows it is fine, but looks like it is actually not. Waiting for new Q1 to arrive.

[dio©]

hello, with a friend we are fixing a pcb that moves one of the two motors. can anyone confirm that these in the red circle are 18V zener or simple diodes?
thanks

EDIT: we found the solution here, in the last post
https://www.gamoover.net/Forums/index.php?topic=28406.96

[dio©]

hello again, can anyone tell me how much this variable resistor should be set to and what it is used for in this pcb?
thanks