Uh oh...he's done it again...geez...

This is the place for any magazine-related discussions that don't fit in any of the column discussion boards below.
Enzo
Posts: 276
Joined: Tue Sep 14, 2004 1:01 am
Location: Lansing, Michigan, USA
Contact:

Re: Uh oh...he's done it again...geez...

Post by Enzo » Thu Oct 07, 2004 5:58 pm

When things get real hot, it is because way too much current is flowing. Generally this is because either a failed part or an unwanted connection is shprting across the power source.<p>Look at your circuit. AC comes on the board, goes through a diode becoming DC, then on to a filter cap for smoothing and eventually to the circuit it powers. If any point in this path is shorted to ground, excess current will flow.<p>In the Pac Man circuit, there are two DC supplies, each having a pair of diodes for full wave rectification. There are a couple other diodes associated with power supply. One is a overvoltage protection zener on the regulator, and another is with the drive transistor for the coin counter.<p>But looking at my simple description, If the AC end of the diode was shorted to ground, the AC would be shorted and its fuse would blow. But the diode would not get hot, because the current that would have come through it was shorted to ground before it could go through the diode. Your diode is hot, so the excess current is flowing through it.<p>That means the short would be on the other side of the diode. If the cathode end of the diode - the DC side - is shorted to ground, then the flow of current would be through the diode on its way to ground. The diode would then be the short circuit across the AC source. <p>How can the diode cathode be grounded? Your solder work or something related might be grounding it. Or one of the filter caps may be shorted, either as a failed part, or one wired backwards. Or something out in the game could be shorted - something that uses the DC supply in question.<p>So what I suggested was to measure the resistance to ground across the main filter caps or from the stripe end of that diode to ground. If the resistance is very low, that means something is shorting the whole deal to ground. If not, there is something else wrong.<p>D8 is one of a pair of main recitifiers for the 7.5V supply that is ultimately regulated down to +5VDC for the logic. It and D7 rectify the 7VAC which goes right to the two large caps. Right next to those two diodes is a third diode - a 1N4004. Is that one shorted?<p>Hold on, look on the coin door for the coin counter. Disconnect the wires from it and see if it makes a difference. Or are you running the board out of the game? A shorted coin counter can load down the 7.5V line.<p>Here is my thinking. Since the AC comes in through both D7 and D8 and they are wired together at the cathodes, anything affecting one should affect the other unless the diode itself is bad. If only one gets hot, could it be the AC is not getting to the cold one due to a burnt trace or open fuse? The +7.5VDC supply that this diode serves only goes a couple places, so there are a limited number of points for trouble. The two caps - negative ends towards the outside edge of the board. The regulator xstr. The 1N4004. And offboard for use by the coin counter via edge connector pins 5 and E.<p>SO measure to see if the cathode and of the diode is shorted to ground. This would be the same as the mail filters being shorted. So you could take the reading at either of those points. I would disconnect the wire from pin 5 of the edgeconn. (Is it a blue wire with a red trace line?) The transistor on the heatsink is likely OK, if it shorts, it probably won't melt diodes anyway. If that supply is shorted, we have to find out where from the limited possibilities.<p>And for that matter, since we were getting hot and all, pull the edgeconn off the board - and disconnect any extra wires you might have added for the AC - and use the diode test function of your meter and test every diode you can see on there. I only see eight of them of several sorts on the whole board.<p>Now as my poor brain conrinues to plod along, it occurs to me we have been concentrating on backwards capps, but we did not mention checking the diode polarity. Are you sure all your diodes are in there the right way? The three diodes in a row next to the heat sink all have their anode ends closest to the edgeconn. All facing the same way.<p>If the 4 ohm, 10 watt resistor is getting real hot - and it gets hot normally - then the load is on the logic side and can be hard to find. But in that case, the diodes ought not to desolder themselves. A dead short to ground on the logic side of the big resistor would result in only a bit less than 2 amps, which wouldn't melt the 3 amp diodes.

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Fri Oct 08, 2004 8:45 am

Oh boy! Enzo...you've really given me a bunch there to work with...Thanks!! My thick skull is starting to get a "little" of it...haha.
I can't wait to try more testing now.<p>In this process, I did not touch any diodes--all I did was replace all the caps. I only resoldered the diode that had come loose. (I've not seen any info anywhere on the size of these two didodes D8 and D7...you mentioned 3amp...is that what they are?)<p>I'll check the 1N4004...and I'll disconnect the coin counter..<p>Since you're so good at explaining this, can you give me a version on why the monitor is not displaying anything? I got no sounds of anything. Just complete deadness, except for a wee bit of speaker hum... I mean, since all I did was replace the caps on this main board (and oh yea, two small caps on the auxilliary board which mates to the logic board by way of a ribbon cable--I've also tested my work with the logic board in the game and powered, with and without this aux. board coupled to the logic board...)<p>You see, my way of thinking was this...the manual told me of some minor problems with video being dried or leaky caps and such, I decided to just replace them. That's also my way of thinking on my jukebox to undertake one day--replacing the caps in its amp....I may be off-base here in my limited thinking.

Enzo
Posts: 276
Joined: Tue Sep 14, 2004 1:01 am
Location: Lansing, Michigan, USA
Contact:

Re: Uh oh...he's done it again...geez...

Post by Enzo » Fri Oct 08, 2004 9:41 pm

That circuit board is really just a computer dedicated to producing the Pac Man game. The 7.5VDC supply is the source ultimately for the +5VDC that the computer runs on. No 7.5 - no 5 - no computer - no picture.<p>Turn off the computer and leave the monitor going on your personal computer and it is the same thing - blank screen.<p>When we encounter a "dead" game we always do one test. We coin the game and hit start. If we see no video but still hear the game sounds, then we know the computer is working and the power supply must be OK. We would then check the monitor. If there are no game sounds, or no clicks on the coin counter, then we suspect there is no powr to the board.<p>We also turn up the briteness on the monitor to look for the background level or "raster." If the heater is glowing in the neck of the picture tube, we know that most of the monitor is working.<p>Now that you mention it, I am not sure if they are 3A diodes. There are 5A fuses in the AC, whjich would be a lot for 1 amp parts. SO 3 sounds about right to me.<p>If the game was not melting this diode before and now it is after your recap, it would be a VERY large coincidence if it was not due to something you did.<p>You said you resoldered a diode that unsolderd itself. If it did that, I would not expect it to still be very good. Did you check the diodes as I described?

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Sat Oct 09, 2004 10:12 am

Enzo...I appreciate your way of explaining...makes a lot of sense...<p>I have been so busy at work (MENTAL exhaustion and lots of sleep to follow) that I've not had the time to start my "new" testing procedures...
Hope to get to it either late tonight or tomorrow...
Results to come...<p>Thanks again.

toejam
Posts: 253
Joined: Wed Dec 05, 2001 1:01 am
Location: n.c.
Contact:

Re: Uh oh...he's done it again...geez...

Post by toejam » Sun Oct 10, 2004 4:13 am

If I' getting your information correctally, yoy are takind the leads from tht vom and potting them on those of the capacitors and either getting a slowly decreasing resistance reading.Thas shows the wiring to the caps is not shoeted.You now should measure the voltage across the large caps and see if it is what is in the schematic. Than go down the line and do the same.

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Sun Oct 10, 2004 6:58 pm

guys: I've done some more "testing"...I've checked all the diodes I can find with my DMM on "diode" setting...all but two read as you would say would be normal. But two are different...if you have access to that schematic from the site that Edd mentioned, the diodes are:
D1, which is a 1N914B, and D9, which is a 1N4735.
The D9 reads high readings on both sides. The other diodes read as you have suggested "about .5 volts or so when measured one way; OL when read with the meter leads switched. Now the D9 reads very low on BOTH sides. And just for the heck of it, I did a cont. test, and it beeped with both leads on; the others did not. Does this tell you anything?<p>With the board out on my bench, can I just use some little wallwart power supply of 7.5 volts, and attach it to the edgecard leads for the 7.5 volts, and try and pick up readings from that point on on various points on the board? This is probably silly, but again, I'm trying to teach this old brain some "new" things...<p>thanks again...

Enzo
Posts: 276
Joined: Tue Sep 14, 2004 1:01 am
Location: Lansing, Michigan, USA
Contact:

Re: Uh oh...he's done it again...geez...

Post by Enzo » Sun Oct 10, 2004 7:15 pm

Well now there you found something. The little 1N914 is not blowing your board, it is not in a place to affect the power supply.<p>But the zener sure is, it is dead smack across the +5VDC rail of the board. Unsolder one end or it and test it that way - so it is no longer in the circuit. No need to remove it, just undo one end and pull it up. In the trade we say "lift" the part. Now if it still measure low both ways it is shorted and loading down the supply. If it measures OK, then make a resistance reading across it - well, across where it solders to anyway. You are then measuring resistance across the 5V buss. I don't have a board handy, they are all out in the warehouse, but all those chips add up to a fairly low resistance, so you may simply be running into that in which case we keep looking.<p>Are you sure the two 5A fuses down in the bottom are still OK? I worry that only one diode is melting. They should either both melt or neither. If one diode is melting and the other remains unpowered, then only the one melts. On the outboard ends of the two main diodes do you measure 10VAC on each with respect to ground. If one has no 10VAC, then it is unpowered. On the inboard ends, you should have about +7.7VDC.<p>Is that big 10 watt resistor getting hot when the diode does?<p>If you want to try to power the board, you could hook 7-8V to it. I would go to the two main filters - easy spot to clip to. But that won't work. The regulator has its own supply. That is the GUIC part draw below the main diodes. The other pair of diodes power it.<p>The way to power the board for real is to apply +5VDC (regulated) to the 5V buss directly. The easiest spot is the end of the 10 watt resistor - the end not connected to the diodes. The end that IS connected to the zener. If you do, then watch close because the bus may be shorted.

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Mon Oct 11, 2004 12:06 am

Ok...I lifted one leg of the suspect Zener...but on diode setting of DMM, I get .7 one way, and OL the other. So I take it that it's not the problem. But with one leg still lifted out, I set the DMM to ohms and measured the across the two pads the zener sets into, and got a reading of 113 one way, 127 the other...<p>I checked all the fuses at the bottom of the cabinet the other day, and all checked ok cont. with meter. The last time I hooked up the board and powered up, I noticed that the two bead diodes did not get AS hot, but still hot, and the area just above it just under the connector was not as hot. But remember I replaced the tab with a wire straight from the harness soldered directly onto the pad on the anode side of the bead diode. I did notice a small amount of heat coming from that 4ohm, 10watt resistor. The large heat sink did not seem hot anywhere on it, nor the top of the transistor that I could sense.<p>Let's see if I understand this ...powered up, I should take my meter and test the points of the cathode end of one bead diode (and the other meter lead touched to a ground point) to hopefully read 7volts DC? And then test the anode side (one lead on ground contact and other on the anode side of the diode) to find 10 volts AC? And "GUIC"..is that the big heat sink assembly with the little transistor in it?<p>Again, I appreciate your efforts in this...I'm like a dog that's biting...I can't let it go...haha.

User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Re: Uh oh...he's done it again...geez...

Post by Edd » Mon Oct 11, 2004 1:23 am

Well I see that there has been quite a bit of activity on this unit. In the interim a deadline has kept me as busy as a one pawed cat with the diarrhea, however, I have a free second now. At the time of my reply I also didn’t have reference access to the schematics, just my poked up notes.
Now considering that you didn’t find your D9 Zener to be shorted after lifting it out of circuit.(BTW..I lift a lead from a board by taking the short length of a very small Allen wrench under the lead of the component and then heat the solder joint and then the Allen’s long end is levered to permit a very controllable lifting up of the lead. Typically, when using a pair of needle nose, oftimes results will be a wild component fracturing jerk, if a sticking/stubborn lead releases suddenly.)
If you find D9 dead shorted, expect the Red G.E. D44VM4/(TIP31) series power pass xstr on the larger heat sink to be shorted C to-E.
Aside…Your errant reading experienced on the small sig ’914 is probably due to a shunting resistance tied into its circuitry, not the diode proper.
I think the most useful analysis information could come by your taking some voltage readings. In which case you would only be powering up the unit for just adequate duration to accomplish a reading and then logging it for reference.
Initially place the meter in AC mode and run a test clip from meter negative to the 10,000u filters negative ground lead for all readings.
1…Get the AC voltages on the anodes (*22 pin connector side *) of 1N4004’s diodes D3&4.
2…Move to the anodes (*)of GenInst A15’s/(1N5404) diodes D7&D8.Log those AC voltages.
3…Are those voltages up in the proper ranges or are they seriously loaded down and reduced ?. If towards their norm, then:
4... Switch to DC voltage mode and see if common cathode conjoined D3 & D4 are in the 16VDC range …or what is it?
5...See if common cathode conjoined D7 & D8 are in the 7.5 VDC range …or what is it.
6…Get the DC voltages on both sides of Pass transistors ceramic block 4 ohm 10W current shunt/sharing resistor R53.
That should be adequate info before investigating switchable regulator
78GU1C, as it doesn’t turn on unless its turn on voltage threshold has been met by its passage thru R53.<p>Testee…testee..time with F/B.<p>Schematic referencing is made to.…page…22…of:<p>
http://www.arcadedocs.com/vidmanuals/P/ ... -Part2.pdf <p>73's de Edd
[email protected] ........(Interstellar~~~~Warp~~~Speed)
[email protected]........(Firewalled-Spam*Cookies*Crumbs)
;) ;)<p>[ October 11, 2004: Message edited by: Edd Whatley ]</p>

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Mon Oct 11, 2004 8:57 am

Oh this is getting good...I'm getting goosebumps...haha...<p>Thanks again Edd...here are some measurements:<p>(AC voltages)
anodes of diodes D3&4 1N4004.....12.21<p>anodes of diodes A15's 1N5404 D7&8....4.73 on D8
low or mv on D7<p>I went a little further and checked DC on:<p>cathodes of D3&4 are both at 14v<p>cathodes of D7&8...1v<p>both side of ceramic block 4ohm 10watt....1v.<p>Hmmmm...I feel like we're getting somewhere now.<p>Thanks for another great step-by-step instruction.

User avatar
Edd
Posts: 885
Joined: Sun Jan 27, 2002 1:01 am
Location: Dallas Tx
Contact:

Re: Uh oh...he's done it again...geez...

Post by Edd » Mon Oct 11, 2004 12:01 pm

[[[Oh this is getting good...I'm getting goosebumps...haha...]]]….hold on REAL tight …this is going to be a wild ride!<p>Well looking at the info on the D3-D4 circuitry, we can leave it alone, as the D7-8 circuitry is wherein we need to concentrate. First pull the 22pin connector from the board and then check for static AC voltage derivation on that connector using shared pins A&1&B&2 on that connector for your AC metering common lead test point. Check for the AC voltage present on C & 3 connectors…log it…then the AC voltage on the questionable connectors….D & 4….??? if not the same as the adjunct windings voltage, track all the way back to the T-formers leads thru series fusing and wiring to find why not.
At a juncture now, of not knowing your findings, but playing ahead….the next procedure would be to have a test on D7/8 with the 22 pin connector still left unconnected, so that no low shunt transformer winding resistance could confuse your interpretation of readings. Enact a test of your full wave rect diode pair D7 and D8 using your meters diode test function mode. Now do we have about the same measured junction voltage on both diodes or possibly a shorted or verrry raaarely…open junction…. within D7.
Standing by, to see if we have to go on to the load/component shedding/isolation route in accordance with your F/B data.<p>73's de Edd
[email protected] ........(Interstellar~~~~Warp~~~Speed)
[email protected]........(Firewalled-Spam*Cookies*Crumbs)
;) ;)<p>[ October 11, 2004: Message edited by: Edd Whatley ]</p>

Enzo
Posts: 276
Joined: Tue Sep 14, 2004 1:01 am
Location: Lansing, Michigan, USA
Contact:

Re: Uh oh...he's done it again...geez...

Post by Enzo » Mon Oct 11, 2004 4:39 pm

Yep, you have verified that one of the diodes is unpowered - the zero VAC reading at the anode. If the 5A fuses in the bottom are OK, make sure the holders are working. One end of a fuse clip gets spread and it no longer makes contact for example. Like the man says, inspect the edgeconn where the traces for the diodes contact.<p>Even with only one of the rectifiers powered, the _+7.5 should not load down to 1V. The fact that the 10 watt resistor has the same voltage at each end tells me the load is before it, unless the pass transistor is shorted. That leaves us with the main filters, solder bridge to ground, and external loads that might be on pins E and 5 of the edgeconn, right above D8 on the drawing.<p>Lift the main filters from circuit one at a time to see if either is loading us down. You can remove them both for test. The game won't work that way, but you can see if the 7.5V rail comes back up. With no filters it would only measure 4-5V by the way.

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Mon Oct 11, 2004 5:48 pm

Oh boy...I'm getting giddy...haha<p>Here are my latest findings...
After pulling the connector and powering up; with gnd. lead stuck into the A1 &/or B2 slot, and the pos. lead touched to c3 and d4, I am now getting 7.5v ac at the ends of the wires. I had found that one of the 7v fuses had gone (ironically one I had replaced already--I may have shorted it when trying to test around that area)
Then I replaced the connector back onto the board, powered up, and started reading again. Here are the results: (gnd. probe on gnd. end of 10,000 cap)<p>still getting 12. VAC at anodes of diodes D3 and D4...<p>Now getting about 7. VAC at anodes of the A15 diodes D7 & D8...<p>In DC mode...<p>cathodes of D3 & D4 I'm getting 15 volts...(again to make sure, that's with the gnd. probe attached to the ground side of one of the big 10,000 caps, and the pos. lead touching just one of the diodes)...<p>and 7.4 volts on each common cathode of D7 & D8...<p>Now this one I'm not so sure...readings taken at each end of the ceramic 4ohm 10watt resistor are 5.volts nearest the edge of the PCB, and 7.4 towards the middle of the board...<p>More thanks...

User avatar
ModRob
Posts: 195
Joined: Sat Jan 11, 2003 1:01 am
Location: Chattanooga, TN
Contact:

Re: Uh oh...he's done it again...geez...

Post by ModRob » Mon Oct 11, 2004 5:54 pm

OOPS....I forgot one item...I just realized I had taken those previous readings and not put the leg of the zener diode back in the circuit...
I'm heating it now..and will test again to see if any difference...

Ed B.
Posts: 45
Joined: Wed Sep 24, 2003 1:01 am
Contact:

Re: Uh oh...he's done it again...geez...

Post by Ed B. » Mon Oct 11, 2004 7:34 pm

I'm still willing to bet that one or ALL of his caps are connected in reverse. How did you locate the negative lead of the caps ? Was that negative lead connected to the ground side of the power supply ? Please double , triple check.
Ed B.

Post Reply

Who is online

Users browsing this forum: CeaSaR and 33 guests