Restore of a nice Donkey Kong - COMPLETE

[Maximus]

Looks like no high score saves or D2k for me, Z80 isn't socketed

There's more than one way to skin a cat

So you're saying I should write down my high scores on my cat then

[CoryBee]

Looks like no high score saves or D2k for me, Z80 isn't socketed

There's more than one way to skin a cat

So you're saying I should write down my high scores on my cat then

You have to knock it out first, trust me you don't want to engrave a conscious cat.


If you need to replace the chip you could do a myriad of things to get it off (hehe) if that is what you want.

Drag solder, heat gun, 

[yotsuya]

He wants you to use the cat skin as a replacement for those socketed chip, duh!

[Maximus]

Did more testing on the dead PSU yesterday. Using the troubleshooting flow chart I now know that TR1 and TR3 are good which lead me on to test D1, D7 and D2. I pulled up one leg on D2 and D7 to test them and they were working so I resoldered them. D1 showed working without having to pull it from the board.

This leaves me with only one last thing to test from the troubleshooting sheet, a Silicon Controlled Resister (or Thyrister) at SCR1 which, ironically is physically right next to the blowing R1 resister. The component is an NEC 5P2M. I had a hell of a time trying to work out which if the legs are the cathode, anode and gate as finding a datasheet with that information seems almost impossible, also I am struggling finding a replacement for it, closest I can find is a 5p4m on eBay.

I found a pinout for other NEC thyristers (5p4m) which indicate that with the component facing you the pins from left to right are...

1- Cathode
2 - Anode
3 - Gate

http://pdf.dzsc.com/5P4/5P4M.pdf

Assuming this is the same for the 5p2m I get the following results when I have my multimeter set to ohms

Red lead on Cathode > Black lead on Anode = infinite resistance
Read lead on Cathode > Black lead still on Anode and bridged to Gate = 222ohm
Red lead on Cathode > Black Lead on Gate = 222ohm
Red lead on Anode > Black lead on Gate = infinite

Black lead on Cathode > Red Lead on Gate = 222ohm
Black lead on Anode > Red Lead on Gate = Inifinite

Any insight?

[g0pkh]

Hi all

This may seem strange to you all that a complete newbie (who has only ever built a mame cab) posts offering help here.
But I have been following this thread for a while now with great interest.

I trained in electronics many years ago, and spent a long time gaining a lot of experience repairing switch mode PSU's  just like this.

I would like to fill in a few blanks which might help explain a little more what is going on here.

Reading from the schematic posted earlier on this thread, and seeing the photos. It is clear that this PSU is built over 2 circuit boards
but the schematic doesn't show this (no interconnections shown, is this the difference between the A and B units ?)

I will add the following comments but understand that I can only go from the schematic shown of course.

1) the resistor which is burning up appears to me to be a startup surge limiter. It is effectively shorted out by the thyristor mentioned
(SCR1) within a few cycles of the PSU starting up. The supply for the thyristor is provided via D1, R3 and C8 from a winding on the transformer T1.

2) therefore if there is a problem (short) on the main PSU board in that the supply wont start then the SCR will not turn on and the resistor will pass all the wasted current and burn up. In fact if there was a full short on the main PSU board (T1 or T3 short collector to emitter)  then I would expect that resistor to die violently in a shower of sparks, it didn't appear to do so.

3) I think your next course of action would be to test again (to be totally sure) T1, T3, the diodes  D2, D7 and D1. You should really do this out of circuit to completely verify that there are not shorts on these devices (use the diode test range of your meter).

Should there be no shorts on these components then we should take look at the secondary outputs, and associated rectifiers

D4, D5 and D9, note that D5 and D9 have two diodes in the package. The common cathode pin would normally be the centre pin.

As a note, I am sure you know but the Japanese miss 2 characters from transistor numbers usually.

The devices T1 and T3 would be 2SC2525 and 2SC2555 respectively.  2SC, 2SD  being NPN types and 2SA PNP.

I appologise for jumping in to this thread like this as a complete newbie. But (call me weird and maybe even sadistic if you like) I find SMPSU's fascinating.

I can take you further into fault diagnosis if you wish, but I will leave it there for now.

Kind regards

Pete

[yotsuya]

Kudos to you, Pete!

[CoryBee]

@Pete

 

[Maximus]

Hey Pete,

Thanks so much for jumping in, please please continue as I really do need someone of your skill level to be able to break down these steps into dummy level concepts and explanations so I can learn from the ground up. At this point I don't care about how much time I have to take or money I have to spend on this supply as I have a replacement coming for the cab itself. This is an opportunity for me to learn a new set of basic skills, and there's nothing better to motivate learning than a problem that needs solving.

Yep I worked out the proper descriptions for TR1 and TR3 and bought NTE equivalents at Frys to replace them, so that's why feel confident that they are now good (unless I killed these new ones too).

In case this gives any insight, the board blew when I already had the plugs for CPU and Audio plugged in. The PSU was powered on and I foolishly plugged in the Video plug (10 pin I believe) that's when the short came about - not sure if that helps narrow things.

[Maximus]

Also small update, ThisOldGame are shipping me replacement side vinyls this week, they agreed the vinyls are usually cut to 72" in length so they didn't know what happened with my pieces, they are taking care of it and I appreciate that for sure.

[g0pkh]

Hi all

Thanks for the kind welcome here.

I will try to be as much help as I can then. Before I continue though. I will say that I have no experience in the game cabs themselves. As far as I am concerned a cab just contains sub assembly's that connect together.

Psu, logic board, some wiring. And a colour monitor. In the event of any problem. It I simply a matter of breaking the unit down into individual units.

You mentioned that you had NTE equivalents. I must had missed the part that mentioned the original transistors were faulty. Can you tell me what the ohms readings were between the c,b & e

We're both TR1 and TR3 faulty?

What equivalents did you fit? And are they still ok? We need to be certain that the package types and lead outs are the same. It would be helpful to see pictures of the originals and the replacements.

If anyone has the schematic of the A unit psu showing the interconnections between the two boards, that would be very helpful.

Pete

[Maximus]

Here you go...

c2535 equiv fitted... NTE379
http://pdf1.alldatasheet.com/datasheet-pdf/view/10331/NTE/NTE379.html

c2555 equiv fitted... NTE2308
http://www.digchip.com/datasheets/parts/datasheet/333/NTE2308-pdf.php

[Maximus]

Not sure if there is any help on this layout...



When I was plugging the video board socket into the PSU I think the two +5V pins hit first, maybe before the ground pins which is what start the fry up. I've yet to find any schematic that shows the interconnects between the two PSU boards

[Maximus]

Found a thread on KLOV with very similar issue...

http://forums.arcade-museum.com/showthread.php?t=193370

No resolution posted, but this trouble shooting info was offered...

Ok so I am at the library and looked at supply schematics.
Check

br1, c6, c7, scr1, r2, r3, d1, tr1.

visually check c5, c8.

If nothing there is shorted to grnd then we will have to look after the xformer

d4, d5, d6, ic3, tr2

[g0pkh]

Well the equivalents certainly look compatible at first glance.

Did you make sure you fitted the insulation washers,or pads if there were any? And if so be sure that the collector ( middle lead) is insulated from the metal heatsink.

Do these devices still read ok? Are you completely familiar now with how to test a transistor using a meter with a diode test facility. I assume your has this. ?

If I am going too fast. Do stop me.

The originals were both faulty, yes?  I find that a little strange if they were. Since the unit actually comprises of two almost separate supplies.

Thanks for the diagram, that is a little useful since it shows the entire game.
What is really needed is the PP7-A power unit diagram. That is what contains the two boards we re concerned with

I assume the B unit contains the entire psu on one board rather than two. Can anyone confirm this for me.

Pete

[Maximus]

Hi Pete, no the 7B and 7A are virtually identical bar a few small tweaks as I understand it.

TR3 was conducting in all directions, all legs, that was how I came to the conclusion it was fried. I must admit I just replaced TR1 as I was in the store so I just bought that and threw it in too, bad science I know but I'm so lost in this I was grasping at anything.

I made sure both transistors were insulated from the heat sink with the rubber cover that goes over them all.

I've tried testing the transistors with my multimeter and think I have done it right, but to be honest I really don't know for certain again due to lack of experience.

As for a layout showing the interconnects between the two internal PSU boards I think we are SOL

[g0pkh]

Your last post refers almost exactly what I said in my first.
When I talked about the secondary side.

Secondary side. Is all parts after the transformers. In particular the circuit which is isolated from the mains supply.

Pete

[g0pkh]

Ok

So if TR3 showed continuity between all three connections then that would cause a burn up of the resistor.

You shouldn't,t have replaced TR1 at that point without testing. As you could have introduced another fault.
But you understand that now 

Have you tried the PSU since. If not. Please don't yet. We should do a few further tests first.
I could use some nice close ups of the circuit board containing the transistors. We also need to confirm that none of the diodes are faulty.

Ok about the A and B units then.
The difficulty we have is determining exactly what one you have.

[Maximus]

I know for sure it's a 7B, due to the location of some of the caps, (and it says 7B on the outside  ).

Yes I tried it after installing the two new transistors and poof went R1 again. I think I still have TR1 in teh trash can next to my desk so I can fish it out maybe later this evening and try and retest it.

[g0pkh]

Ok. If R1 failed again. You must recheck those transistors again.

[Maximus]

Ok. If R1 failed again. You must recheck those transistors again.

Good times 

[g0pkh]

Also D2 D3 D7 D8 and D9

D9 will look similar to TR3 and be mounted on the heatsink under the rubber strip.

Don't at this stage remove the diodes just make sure the board is unplugged and completely removed.

Come back to me with meter readings in both directions on all above parts. .

We can then remove any if the readings look wrong to confirm further.

You should use the diode test range on your meter. And wait for readings to settle. Measuring in circuit will charge capacitors. So just leave meter probes on for awhile.

Pete

[g0pkh]

Testing D9 The dual diode in circuit will show a short across the two outer connections. This Is
due to the transformer secondary winding.

Just test each outer connection to the centre pin, both ways

[g0pkh]

Sorry.

You will need to remove D9 to test. There is a 100ohm load resistor which will scupper your test.

That lot should keep you busy for a while 

[Maximus]

Ok will do, thanks for all the info, I'll report back with results if I get to it this evening. Also d2 and d7 wouldn't test in the board last time I had to lift a leg on them - lol

[Maximus]

Going to pick up one of these for $19 to keep in my toolkit, seems like it will prove very useful as I move forward with this area of the hobby...

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=400431622482&ssPageName=ADME:L:OC:US:3160

[Maximus]

Assuming pins left to right with transistor front facing are 1,2 and 3

Old removed C2555
1 neg 2 pos - 0
1 neg 3 pos - 0
1 pos 2 neg - 0.531v
1 pos 3 neg - 0.531v

Old removed C2535
Nothing in any combination on any pin

[Maximus]

C2535 in circuit, new component

1 neg 2 pos 0.001v
1 neg 3 pos 0.001v
2 neg 3 pos 0v
2 neg 1 pos 0.001v
3 neg 1 pos 0.001v
3 neg 2 pos 0v

C5555 in circuit new component

1 neg 2 pos 0.005v
1 neg 3 pos 0.005v
2 neg 3 pos 0v
2 neg 1 pos 0.005v
3 neg 1 pos 0.005v
3 neg 2 pos 0v

[Maximus]

D2 good 0.443v
D3 good 0.443v
D7 good 0.443v
D8 good 0.452v

Had to lift these out of circuit to test as in circuit they gave bouncing readings in both directions

[Maximus]

D9 out of circuit

1 neg 2 pos 0v
1 neg 3 pos 0v
2 neg 3 pos 0v
2 neg 1 pos 0.135v
3 neg 1 pos 0.135v
3 neg 2 pos 0v

[Maximus]

Still can't work out how to test scr1 properly

[welash]

you could try one of the two methods on this page:

http://www.circuitstoday.com/how-to-test-an-scr

The multimeter only method seems like it might take a little coordination but doesn't require you to build a test circuit.

One of the things you said above confuses me.  Did you replace the transistors, and plug the power supply in, and everything was ok until you connected the video board, are are you saying that originally the power supply was ok untill you connected the video board.

After you replaced the transistors, were you able to plug in the power supply and measure the outputs, and were they ok?

One thing that g0pkh mentioned, are both transistors mounted to the same heat sink?  If so, is there any insulator between the transistor and the heat sink?  If not, it is likely that the 2 transistors collectors are shorted to each other, which wouldn't be good (at least if I read the schematic correctly. You could check if this is happening by checking continuity between the collectors of the two transistors.  If when you removed the transistors there was something between the transistors and the heat sink, make sure to put it back.

[g0pkh]

Ok thanks for the readings.

I am a little confused by your representation of the readings however

When you tested the 2SC2555 you got these readings

Old removed C2555
1 neg 2 pos - 0
1 neg 3 pos - 0
1 pos 2 neg - 0.531v
1 pos 3 neg - 0.531v

What does 0 mean ? Open or short ?

What does your meter display when
1) leads shorted together = short circuit
2) leads open = open circuit

The readings of 0.531 actually look good (it is close to the junction voltage of a diode.
I can see that 1 is therefore the BASE and clearly the junctions look good.

You must always test between COLLECTOR and EMITTER as well, both ways round between 2 and 3

This transistor at first sight looks good.

NEXT.
This test

Old removed C2535
Nothing in any combination on any pin

when you say nothing on any pin do you mean a short circuit or an open circuit ?
Remember meter leads joined together = short circuit and meter leads apart = open circuit.
I suspect you mean short circuit.



The test here

C2535 in circuit, new component

1 neg 2 pos 0.001v
1 neg 3 pos 0.001v
2 neg 3 pos 0v
2 neg 1 pos 0.001v
3 neg 1 pos 0.001v
3 neg 2 pos 0v

confuses me a bit what does 0V mean compared to 0.001V referring to what I asked about meter readings with leads shorted or open.


Does 0V means meter leads shorted, I suspect it does ?

Since you have the board out can you remove this transistor and retest please, post the readings here.

I would expect to similar readings to the old 2555 which you removed.
especially the .531v or similar between base collector and base emitter.

For guidance on testing transistors you may find this page helpful.

http://www.wikihow.com/Test-a-Transistor

I would suggest you study and practice it using the old 2555 which you removed (looks good) and the 2535 youve just removed (looks suspect).


Let me know your test results.

Its looking like the fault is on the 5V part of this supply now.

Pete

[g0pkh]

Something looks strange about D9

While I see two junctions (0.135V)
they appear not to be on the correct pins which you indicated earlier

Can you recheck this for me ?

I would expect to see the .135 reading between 1 and 2 and 3 and 2 in one direction only.
With nothing between 1 and 3 in either direction.


Pete

[g0pkh]

Also

while TR3 is out can you check R19 and R20

according to the diagram

R19 = 51 ohms
and
R20 = 12 ohms.

You will have to lift one end of each to check them because of transformer T4

[g0pkh]

Looking at the circuit diagram, it should be possible to power up the supply with TR3 removed. You are not ready to try this yet however, so please only read and try to understand.


I feel at this stage that I must highlight some very important safety information. There are dangerous voltages within this unit.

I would like to express that the information I give is for information only, it is entirely your responsibility should you wish to act on it. It must be understood that I will take absolutely no reponsibility for any damage, injury or even death!

LETHAL VOLTAGES ARE PRESENT.
PLEASE BE VERY CAREFUL. IF YOU ARE IN ANY DOUBT DON'T DO IT...

 
Ok. ON with the show ...

I will give you a run down of how it works.
I appologise if I loose you through this blurb but I will try to keep the description as easy as possible. You will I am sure need to read this two or three times, maybe even more.
If you can understand this eventually then it will stand you in good stead for future projects too.

Print off the schematic diagram first and have that in front of you.
 
You should be able to see that this PSU actually contains 3 separate power circuits.

1)A switching power supply to provide -5, +12 and +24
The transistor for this is TR1 a 2SC2555.

2) Another switching supply to provide the +5V for the game board, this is quite a high current supply hence it is separate.    Switched of course by TR3 the 2SC2535.
This supply also has    overvoltage protection built in to protect the logic (more of that later if you are interested).

3) A low voltage (linear) supply comprising mains transformer T5 and diode bridge BR2. This provides around 12V to power the control IC (IC1) which provides the switching signal for TR1 (via transformer T3)

The supply for IC2 (switching signal for the 5V logic PSU TR3)is actually provided from the +12V supply of the other PSU via a 7812 voltage regulator (IC4) from the +24V rail. Hence if the 24V rail is not there then neither will be the +12V or +5V .

The two main PSU's are fed from about 180V DC which is present across the large caps C6, C7. This input voltage is provided from the rectifier BR1 through our old friend the surge limiter R1.

SCR and R1

   As soon as the 24V supply is present then an AC voltage    will be present across the transformer winding T1 and    rectifed by D1. It is this voltage which fires the SCR    effectively shorting out R1 until power is completely    removed and next restored.

Perhaps at this stage maybe I should stop for a breather. If you are finding the info useful then I can go deeper if you wish.


Pete

[Maximus]

Ok this is a lot to go through, thanks for taking all this time to lay it all out. First I'll start by trying to answer the questions in the order you asked them. Now I understand what OL means on my meter lol...

c2555 - old one
1 neg 2 pos - OL
1 neg 3 pos - OL
1 pos 2 neg - 0.531v
1 pos 3 neg - 0.531v
2 neg 3 pos = OL
2 pos 3 neg = OL

c2535 - old one
0.000v on every leg combination so I assume that means they are all shorted, otherwise it would say OL

c2535 - new - test out of circuit
1 neg 2 pos 0.002v (SHORT?)
1 neg 3 pos 0.002v (SHORT?)
2 neg 3 pos SHORT
2 neg 1 pos 0.002v (SHORT?)
3 neg 1 pos 0.002v (SHORT?)
3 neg 2 pos SHORT

looks dead again I guess

D9 retest out of circuit
1 neg 2 pos OL
1 neg 3 pos OL
2 neg 3 pos SHORT
2 neg 1 pos 0.140v
3 neg 1 pos 0.140v
3 neg 2 pos SHORT

R20 out of circuit - 12 ohms
R19 out of circuit - 51 ohms

SO looks like TR3 is also getting killedi n the 5v circuit. This would also coincide with the circumstances witnessed where it first blew when I said that I think the +5v pins hit before anything else.

[g0pkh]

OL yes that's better LOL

D9 looks bad too. There is one diode which is shorted in it

The one between pin 2 and 3

because of the short you are reading the good one twice.

We need to try to source a replacement.

That would be what is killing the transistor too, I expect.

Will see what I can find out about it.  It is a dualt diode with the common cathode connection at the centre. Will be a schottly type too.

Pete

[Maximus]

Ok D9 is marked as C10P048 I did a quick search on that but having trouble finding info on it.

[g0pkh]

CTB33S is a dual 12A  35V schotty barrier diode, with common cathodes to the centre pin looks like a TO220 package but cant be sure.

An equivalent should be easily obtainable.
Since I am in the UK I would likely find a different equivalent.

You need another transistor too. Along with R1 again.

Don't rush to power it up afterwards you should make cold checks first, it's all too easy to blow the whole lot up again.

Pete


EDIT

Hmm I would guess a 10A 48V device. Different to that shown on the diagram then.

[g0pkh]

Something like this from Digikey (I know you guys use them across the pond)

http://www.digikey.com/product-detail/en/VS-25CTQ045PBF/25CTQ045PBF-ND/812210


or this

http://www.digikey.com/product-detail/en/VS-MBR3045WTPBF/MBR3045WTPBF-ND/811514

assuming of course that the package matches