(Topic ID: 79650)

TECH—TZ Only Resets When Cold (Doesn’t Reset When Warm Despite 5V Only At 4.8V)


By NM

5 years ago



Topic Stats

  • 79 posts
  • 20 Pinsiders participating
  • Topic is favorited by 9 Pinsiders

You

Linked Games

Topic Gallery

There have been 10 images uploaded to this topic. (View topic image gallery).

photo-979.JPG
DSC01548b.jpg
DSC01439b.jpg
DSC01530b.jpg
photo-944.JPG
DSC01516b.jpg
LM323KHackSolderSide.jpg
DSC01517b.jpg
DSC01511b.jpg
anigif_enhanced-buzz-24877-1368536475-9-973.gif

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

-1
#1 5 years ago

(Initial Post Below Now Updated Reflecting All The Tests Performed So Far/Latest Conditions

My TZ resets the instant both flippers are pressed simultaneously (even at 0 points on the DMD with Ball 1 in the shooter lane)…this happened EVERY time without fail--but only when the game is cold (IE--10 out of 10 times.)

But after the game warms up, it doesn't reset despite the 5V rail being at only 4.80V (I thought these games most certainly reset if under the 4.92 V)?

Anyway, I started methodically trouble shooting the problem by going through all the usual steps:

1. Line voltage: 119.9 volts; with the furnace on: only drops to 118.7 V.
2. 5V TP2 on PDB: shows a pathetic 4.80 Volts.
3. 5V on the CPU on EPROM pin 32: no change, 4.80 Volts.
4. Reseated J101 on the PDB and used a pusher: same voltage on pin 32: 4.80 V.
5. Reseated J114 on the PDB and used a pusher: roughly same voltage on pin 32: 4.79 V.
6. Reseated J210 on the CPU and used a pusher: roughly same voltage on pin 32: 4.77 V.
7. Z connector was previously eliminated.
8. 18-Volt TP8 was between varying between 15.75 and 18.5V.
9. Decided to reseat all 5 connections coming out of the transformer, same voltage afterwards: 4.80 V.
10. All PDB ground screws are tight.
11. Earlier thought I might have a failing DMD or Controller Board, swapped in 2 known good parts; still resets when cold.
12. Thermistor: jumpered around it and game still resets when cold.
13. All other connectors in the power box are secure; in-line fuse holder looks fine.
14. All 8 flipper diodes present; no cold solder joints.
15. DMM reading on BR2: reading on positive leg was 9.86V.
16. C5: reads 0.286 with DMM on AC Auto.

Thanks in advance.

-1
#4 5 years ago
Quoted from markmon:

Also, could be smal meter variance. And could be the machine doesn't reset once warmed up.

UPDATE:

Wow, what a 180-degree turn:

Markmon, your instinct was 100% right--I just tried the game after it sat all night and it's back to resetting every time both flips are pressed simultaneously (with Ball 1 in the shooter lane and 0 points on the DMD)!

I'll continue going through the guide later where I left off to pinpoint the cause, but is there any typical culprit in a case like this (when a game only resets when cold & not after it's warmed up)?

Thanks.

-1
#8 5 years ago
Quoted from burningman:

Maybe the Thermister?

Good suggestion re the cold vs. warm angle but I just ruled out the Thermister by jumpering around it (game still reset with a jumper while cold). Thanks.

And thanks for the suggestion Mantra of where your issue was, I'll check it out.

I actually have both PinWiki & Clay's guides and will get back on working through them later.

-1
#17 5 years ago

UPDATE:

I think I found the problem: the positive leg of BR2 is reading only 9.86V...this is suppose to be at 12V isn't it?

BTW, there seems to be conflicting info between the two guides on BR2 (unless my eyes are failing me):

Clay's Guide says (in the section entitled: "Testing a Bridge WPC-S and prior, Under Minor Load, In the Game"):

"Any less than 12 volts, and the bridge (or the connection to the bridge) is bad."

But PinWiki says in Section 4.8.10 (an in-game, power-on test also):

"The meter should read about 9V. This will be true whether C4/C5 are working or not. If the bridge rectifier is bad, it will read about 7V".

So which is right?

If 12V is correct, would cracked solder joints here possibly explain why the game doesn't reset when warm?

I guess it's time to now pull the Power Driver Board & examine it for cold solder joints, cracked header pins, etc.--unless there's something I missed that I should try first? (Initial post has now been updated showing all tests done to the game so far/latest conditions.)

Thanks.

-1
#20 5 years ago
Quoted from terryb:

Sorry to say this, but I think the pinwiki info is correct--9 volts would be about right for a supply voltage that is being regulated down to 5 volts.

Quoted from Borygard:

PinWiki is correct.

Thanks for the clarification.

Any ideas on what I should try next (given my updated work-through list in my initial post--other than the cold spray idea on the suspect areas on the PDB or CPU)?

Or is it now time to pull the boards--and look for cold solder joints/cracked header pins, etc.?

Is there something I missed--before I pull the boards individually/or sub-in known good boards?

Thank you.

-1
#23 5 years ago
Quoted from ChrisHibler:

Take a look at both sides of the J101 connector.
If either side is tarnished, replace them.
Thanks for working thru the PinWiki steps!

Thank you for PinWiki!

Will do on J101 (but FWIW, observed J101 has already had the IDC replaced with Trifurcons/new housing).

-1
#26 5 years ago

Will do.

FWIW guys, the work on J101 was done by a PO/unknown how long ago.

Thanks.

-1
#31 5 years ago

UPDATE: decided to pull the PDB and swap in a known good one.

Recall the previous reading on TP2 on the suspect board was 4.8V--and with the known-good board in the game that same reading has now gone up to 4.97V...and the game hasn't reset when cold as of this writing.

Visual inspection of the board revealed no cold solder joints.

BR1 and BR2 have previously been replaced--so has J120.

Remember: BR2 tested fine in the game, and the solder side looks fine.

J120, J121, and J115 appear to have been reflowed.

There are no cracked power or ground header pins or cracked solder at J101, J102, or J103.

The LM323K regulator has no fractured solder joints.

Fuse 113 also has no fractured solder joints.

However, what I did find is that on multiple connectors the plating has worn off right where the female connector mates with it--found this condition on J101, J102, J103, J114 and on many more .156 pins...could that alone account for that much of a difference between the two boards (of 4.8 to 4.97V)?

If not, it might just be time to send in this tired old board in for professional refurb.

Thanks.

-1
#33 5 years ago
Quoted from ChrisHibler:

Yes.
When the connectors have reached this stage, I replace them. If it was a coil drive connector, no big deal. But the system requires a nice clean ~5VDC, and this can impact it. What you found also means that it is likely that the female side of that connector needs to be replaced too. Similar damage occurs on both sides of the connection.
This problem is exactly what SteveP3 was aiming at, above.

Very good.

What are the critical ones if I wanted to take a minimalist approach for now: J101, J102, J103, and J114?

Yes, I wanted to try SteveP3's excellent suggestion but just decided to proceed directly to pulling/swapping PDBs.

Thanks again.

-1
#38 5 years ago
Quoted from ChrisHibler:

J101 and J114 are the most important.

OK; I'll make this repair in a week or so and report back (had to order a new adjustable-temperature soldering station).

Many thanks for all the help Chris (and again for PinWiki...it's a FANTASTIC resource!)...and thanks to all responders in the thread!

NM

2 weeks later
-1
#39 5 years ago

UPDATE--replaced the header on J101 and that wasn't the culprit.

Recap:

This game's TP2 was at 4.82V on its PDB--and the game had no drop from the PDB to the CPU at Pin 32.

Swapped in a known-good PDB and TP2 soared to 4.97V (with no more resets of course).

BR2 tested good and it's evident BR2 & BR1 have been replaced by the PO (but looks like C5 & C4 are original).

Female connector at J101 also replaced (by the PO) besides the J101 header I just replaced on the PDB.

Did all the other stuff on the check-list from A to Z (details listed in my initial post).

So what's next: go back and reflow/replace the solder at BR2, C5 & C4 before replacing parts?

Should I add BR2 jumpers to C5 while I'm there?

Anything else?

Thanks.

-1
#43 5 years ago
Quoted from rkahr:

NM,
PS seems to break the link (or maybe operator error on this side) - search for item number 281270779341.
Rob

Thanks Rob, very interesting--here's a working link:

ebay.com link

Noted from the YouTube it's plug n play on the CPU--guessing the design accounts for if the problem originates from the PDB?

-1
#45 5 years ago

PS--remember this game resets only when cold.

Did some RGP research and found that cold resets:

"Could indicate that the filter caps C5 on the Power Driver PCB ... could be dryed out. Dry caps don't filter well until they warm-up."

And I THINK my C5 is original, but not sure (sloppy work on BR2 & BR1...C5 appears untouched but who knows).

BTW, my Thermistor tests fine.

So plan is now: reflow/replace solder on BR2, BR1, C5, C4, and add jumpers from BR2 to C5.

And if that craps out, replace C5.

Sound good/anything else?

#48 5 years ago
Quoted from Borygard:

Unfortunately this solution is a band-aid to the actual problem. I suspect that because it's a very quick and easy band-aid that some will latch onto it and consider it the greatest thing ever. I also suspect it will severely tax the 12V regulated power and cause issues there.
One thing that really bothers me though, is the sales pitch blames the CPU's watchdog. The watchdog is actually very rarely the problem with WPC reset issues. It's very easy to disable the watchdog, do that and watch the resets continue.
--
Rob Anthony
Pinball Classics
http://LockWhenLit.com
Quality Board Work - In Home Service
borygard at gmail dot com

Yes, if I was looking for a quick fix I think I would have gone directly to the "Last Resort" of boosting the 5V to 5.15V with a 22 ohm 1/2 watt resistor & cutting the traces (still might have to LOL). Ridiculously cheap too: only about $1 in parts.

But I really want to find the cause of this now.

It's gotten personal...man vs. machine.

Thanks for the assistance.

-1
#51 5 years ago
Quoted from rkahr:

Don't cut the traces - use vinyl screws/nuts to isolate the heat sink from ground so the resistor hack can be un-done more easily. I get the man/machine logic...I'll be around if that calculus changes.

Great idea; thanks.

1 week later
-1
#52 5 years ago

OK, after further testing/re-flowing/replacing solder, my conclusion is there's now no resort on this but the Last Resort.

Question: on the pics below of the Last Resort mod done to the same board/different version (A-12697-1 vs. my TZ's A-12697-3), what's the green jumper for? (PinWiki doesn't show this jumper as being necessary in 4.8.17.)

Thanks.

DSC01511b.jpg

DSC01517b.jpg

DSC01516b.jpg

-1
#54 5 years ago
Quoted from donjagra:

Jumpers fix bad connections. Any damaged area could require a jumper. Pin wiki just shows the common ones.

Well yes, I realize that but I just thought it was a possible related variant to the Last Resort mod--because the left side of the jumper runs to the negative leg of the capacitor at C9.

Also: on the right side of the jumper, there's no related component--it's just spliced into the trace ahead of the 22-Ohm 1/2 Watt Resistor.

Is this perhaps just an alternative to cutting the two traces on the back as shown in 4.8.17 below?

LM323KHackSolderSide.jpg

#56 5 years ago
Quoted from ChrisHibler:

The hack in the second picture is one that I did.

Curious what's your recommended method of cutting the traces?

-1
#60 5 years ago
Quoted from RobKnapp:

OMG my ST:TNG is now rebooting when flippers are presses. And in attract mode. This is not going to be easy. My Technical skill is lousy .

You can do it--just methodically follow the steps of the guide.

Thanks to Chris and all contributors for PinWiki!

-1
#62 5 years ago
Quoted from rkahr:

Don't cut the traces - if you are going to do the hack do it in a way that has a hope of being undone. Replace heat sink screws with vinyl screws/nuts from home depot/lowes. If I remember correctly I put a vinyl nut between the board and the heat sink AND on the rear of the board. This isolates the regulator from the copper traces. Tie resistor between ground test point on the top side of the board and the casing and to the heat sink. . Incidentally I did this and my PZ still reset... which drove me to design the circuit board I mentioned earlier in the thread.

Sounds great in theory--let's see if I've recapped these steps correctly:

Component side: vinyl screws through the heat sink, then vinyl washers against the solder mask under the heat sink.

Solder side: threaded screws poke through the PCB, then simply tighten with vinyl nuts.

(In other words, just basically just float the heat sink off the solder mask with vinyl washers--so no metal is touching solder--and secure the vinyl screw threads with vinyl nuts.)

And in so doing, this precludes having to cut the two traces on the solder side and the single trace on the component side leading to C9.

Then, just solder a ring terminal end to one end of the 22 Ohm 1/2 Watt resistor and place it under the lower vinyl screw--and lastly, solder the other end of the resistor to the ground test point (TP5) directly below TP7.

-1
#65 5 years ago
Quoted from rkahr:

NM, you got it - particularly like your improved wording of "float the heat sink" as it is more descriptive. Vinyl screws come in limited lengths so plan on cutting down the screw so your board fits back in the back box.
Rob.

Very good Rob.

Looking forward to rounding-up the vinyl parts at Lowes and giving your very clean, Non-Hack Version of the Last Resort Mod a whirl!

Quoted from joe2012:

hi nm,keep us posted if you do rkahr's fix. thanks joe

You got it!

#66 5 years ago

EPILOGUE :

Happy to report this game is now fixed via the Last Resort Mod (and with no hacks thanks to Rkahr: what a fantastic idea)!

The voltage on my PDB's TP2 has now soared from the initial pathetic reading of 4.8 V to now a stout 5.17 V!

Rather than cutting/hacking the three traces, I used Rkahr's excellent suggestion of floating the heat sink off the board with nylon hardware to eliminate the ground--and then installing the 22 Ohm 1/2 watt resistor on the component side of the board (with one end of the resistor connected to a ring terminal under the bottom nylon screw securing the LM323K, and the other end of the resistor soldered to the ground at TP5).

Got the nylon parts at Lowes and the inventory is simple: two 8-32 X 1/2 nylon machine screws (Lowes item #138981) , two #8 nylon flat washers (use the thin ones, Lowes item # 139061), and two 8-32 nylon hex nuts (Lowes item #138980).

Reason to use the thin nylon flat washers is if you're reusing your LM323K like I was (the legs going through the through-holes were already cut of course and were almost too short...made it work with the thin flat washers mentioned above).

Anyway, many thanks to all contributors and a big thank you to both Chris Hibler for his help here (and for PinWiki!)--and many thanks to Rkahr for his fantastic hack-free version of the Last Resort Mod shown directly below!

DSC01530b.jpg

TP2 "Before":

DSC01439b.jpg

TP2 "After" With The Last Resort Mod Installed (Rkah's Hackless Version shown!):

DSC01548b.jpg

-1
#70 5 years ago
Quoted from joe2012:

thanks nm.if I do it I'm going to use heat shrink on both leads and hot glue that sucker in place. thanks again joe

That's a good idea Joe--I thought of that too after looking at the pics: the ground lead is dangerously close to that 20V TP @ TP7!

Although the solder connection is firm, might as well do it while you're there--as the investment is just a penny or two (LOL)!

Thanks again to all responders!

-1
#72 5 years ago
Quoted from rkahr:

Clean work NM. Remember though, a meter can hide what an O-scope can show. This blurry picture is about 2 years old - it shows what a brief collapse in the 5 volt line looks like on a PDB - yes, with the resistor jumper in there to push up line voltage. It's tough to see for a lot of reasons (I took it with my old blackberry, my 30 year-old yard sale-purchased scope (that I revived from the dead) doesn't have screen captures, resets were pretty random so I kinda had to get lucky to catch it), but the steep drop and rapid recovery are clear enough. And it can be enough to take 5.1+ volts down to a reset. Even with the hack you may want to consider using my new daughterboard to move the MPU off the 5 volt line. It is certainly as well thought out as my "Float the heat sink" update to the resistor hack.

Neat shot of the scope dip.

Indeed: it will take many games on the clock yet before an All-Clear can be issued.

If I run into further problems, I'd be happy to purchase and try one of your daughterboards--as you've clearly demonstrated a successful ability to think outside the box/present new solutions to old problems.

Many thanks again Rob!

-1
#75 5 years ago
Quoted from kbliznick:

Gah, I wish people would stop calling the added resistors to the LM323 circuit a hack or a last resort

I think the Last Resort nickname is here for good: Clay's Guide first called this mod the "Last Resort" many years ago--and when he removed his guides, PinWiki referred to it by the same name.

Quoted from kbliznick:

The cuts to the traces on the board can be simplified down to the single cut on the right side as pictured here
http://img-f.pinside.com/201403/1504456/200755.jpg
Instead of making the cut on the left side in the picture and on the front side of the board simply cut off the small cap at C9 just between the regulator and C5 on the front of the board. This cap is no longer in circuit when the other 2 cuts are made so it's simpler and cleaner to just remove the cap.

If I'm hearing you right you're saying only one cut only on the solder side (the one in the right circle).

But your last sentence above refers to two cuts...guessing by second cut you mean cutting the ground leg on C9?

So in summary: your method involves only cutting the single trace, and then removing C9?

Quoted from kbliznick:

Also to the OP please make sure that you soldered the resistor to the ground test point in your picture. If that is just wound around it can come loose and if it does come loose you will potentially be in a lot of trouble.

Thanks, yes mine is wound and soldered to TP5.

#78 5 years ago
Quoted from ChrisHibler:

I do like the idea of lifting C9 instead of cutting. Although, repairing a cut trace is trivial.

Chris: what do you mean by "lifting C9 instead of cutting"?

(Perhaps through breaking the ground of the heat sink via Rkahr's nylon hardware idea?)

Thanks!

Promoted items from the Pinside Marketplace
$ 20.99
Lighting - Interactive
Lee's Parts

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

Hey there! Got a moment?

Great to see you're enjoying Pinside! Did you know Pinside is able to run thanks to donations from our visitors? Please donate to Pinside, support the site and get anext to your username to show for it! Donate to Pinside